ICINCO 2021 Abstracts

Area 1 - Industrial Informatics
Full Papers Short Papers
Area 2 - Intelligent Control Systems and Optimization
Full Papers Short Papers
Area 3 - Robotics and Automation
Full Papers Short Papers
Area 4 - Signal Processing, Sensors, Systems Modelling and Control
Full Papers Short Papers

Area 1 - Industrial Informatics

Full Papers
Paper Nr: 12
Title:

Energy Consumption Modeling for Specific Washing Programs of Horizontal Washing Machine using System Identification
Authors:

Yongki Yoon and Sibel Malkoş
Abstract: This paper presents the application of an energy consumption modeling technique using a system identification method regarding the washing program settings for a horizontal washing machine. The observer/Kalman filter identification/eigensystem realization algorithm (OKID/ERA) method is employed to identify the linear discrete state-space model by choosing the system order computed by the significant singular values. The identified model is used as an estimator to figure out the energy consumption level for washing programs with the full loading condition, and results show the feasibility of the method in energy consumption modeling.
Download
Short Papers
Paper Nr: 33
Title:

Glue Level Estimation through Automatic Visual Inspection in PCB Manufacturing
Authors:

Bruno P. Iglesias, Mario Otani and Felipe G. Oliveira
Abstract: Nowadays, the increasing use of automatic visual inspection approaches in the manufacturing process is remarkable. The automation of production lines implies profitability and product quality. Moreover, optimized human resources result in process optimization and production increase. This work addresses the problem of optimizing the glue tube replacement in Printed Circuit Boards (PCB) manufacturing, warning a human operator only just in time to replace the glue tube. We propose an approach to estimate the glue level, in the glue injection process, during PCB manufacturing. The proposed methodology is composed of three main steps: i) Pre-Processing; ii) Feature extraction; and iii) Glue level estimation through machine learning. The proposed predictive model learns the relation between visual features and the glue level in the tube. Real and simulated experiments were carried out to validate the proposed approach. Results show the obtained Root Mean Square Error (RMSE) measure of 0.88, using Random Forest regression model. Furthermore, the proposed approach presents high accuracy even regarding noisy images, resulting in RMSE measures of 3.64 and 4.15 for a Salt and Pepper and Gaussian noise, respectively. Experiments demonstrate reliability and robustness, optimizing the manufacturing.
Download
Paper Nr: 124
Title:

Profit Maximized Network Optimization at SAP System: A Real-life Implementation in Cement Industry
Authors:

Eren Esgin, Volkan Ozay and Gorkem Ozkan
Abstract: What if we told you that “you already have 27% of net profit trapped in your misleading business”? As common de facto state in production planning, subjective human judgments play a significant role on demand point:plant assignments at product replenishment and this is mostly driven by myopic transportation minimization paradigm, disregarding production and profitability determinants. In this paper, we propose an integer programming characterized Network Optimization solution to find global optimal assignments that maximize the profitability in terms of contribution margin or net profit by taking sales, transportation and production planning perspectives into account and concerning potential capacity constraints. According to the experimental results obtained at a real-life implementation in cement industry, Network Optimization solution increases contribution margin by an average value of 6.33% and net profit by 26.3%. Moreover, proposed solution architecture promises a seamless network optimization experience over a large canvas that wholistically integrates SAP system, optimization logic and Microsoft Power BI tiers. As a result, our clients can concentrate on more value adding operations such as variance analysis and what-if scenario evaluation rather than manual, time consuming and error-prone data preparation.
Download
Paper Nr: 38
Title:

Estimation of the Features Influence on Cluster Partition
Authors:

Daria Kolesnikova, Yuri Andreev and Radda Iureva
Abstract: The use of machine learning and clustering tools for production management, operational and strategic planning is an urgent task. Industrial automation and Industry 4.0 in general stimulate the use of new technologies. So, for the analytics of many business processes and tasks, it is possible to use clustering. This paper evaluates the clustering performance for supplier evaluation considering the influence of preference features. Clustering is mostly unsupervised procedure, and most clustering algorithm depend on some certain assumptions. Subgroups present in the dataset are formed on the base of these assumptions. Consequently, in most cases, the resulting cluster groups require validation and reliability assessment.
Download
Paper Nr: 85
Title:

A BCI Driving System to Understand Brain Signals Related to Steering
Authors:

Enrico Zero, Simone Graffione, Chiara Bersani and Roberto Sacile
Abstract: In the last years, the manufactured vehicles were designed to focus on prevention of some risky situations caused by a human driver. The aim of this paper is to illustrate the design and implementation of a BCI system which can detect the arm movements by the EEG signal during a simulated driving session. The proposed approach to realize a classifier able to recognize the arm movement by EEG feature analysis is based on the consecutive application of a Time Delay Neural Network (TDNN) and a Pattern Recognition Neural Network (PRNN). Preliminary tests are shown on three different participants between 24 and 45 years old.
Download

Area 2 - Intelligent Control Systems and Optimization

Full Papers
Paper Nr: 16
Title:

An Evolutionary Calibration Approach for Touch Interface Filter Chains
Authors:

Lukas Rosenbauer, Johannes Maier, Daniel Gerber, Anthony Stein and Jörg Hähner
Abstract: Touch interfaces are human machine interface (HMI) that can be found in a wide range of products ranging from mobile phones over cars to home appliances. Many of these HMIs measure digital signals which are used to detect touch events. These signals are processed using filters in order to decide whether there is a touch event or not. The filterchain must be functional even if the signal contains heavy noise. Thus a precise calibration of the individual filters is necessary. We employ a genetic algorithm (GA) to choose the filter parameters automatically. We evaluate our approach in a series of experiments which includes simulated as well as real data. We additionally compare our GA with manually calibrated parameters and thereby show the superiority of our method in terms of the accuracy of the calibration provided. A cost-intensive manual calibration can thus be avoided.
Download
Paper Nr: 29
Title:

Multi-target Optimal Control Problems for a Tentacle-like Soft Manipulator
Authors:

Simone Cacace, Anna C. Lai and Paola Loreti
Abstract: We investigate the optimality of the configurations of a tentacle-like soft manipulator ensuring the contact with a target object, while avoiding an obstacle. The main novelty consists in treating the contact sub-region of the manipulator as an unknown of the problem and, at the same time, in allowing the target to be disconnected. We set the optimization problem in full generality, then we focus on the case of a multi-target problem, in which the goal is to simultaneously reach a finite set of points. Numerical simulations complete the paper.
Download
Paper Nr: 36
Title:

The Furtherance of Autonomous Engineering via Reinforcement Learning
Authors:

Doris Antensteiner, Vincent Dietrich and Michael Fiegert
Abstract: Engineering efforts are one of the major cost factors in today’s industrial automation systems. We present a configuration system, which grants a reduced obligation of engineering effort. Through self-learning the configuration system can adapt to various tasks by actively learning about its environment. We validate our configuration system using a robotic perception system, specifically a picking application. Perception systems for robotic applications become increasingly essential in industrial environments. Today, such systems often require tedious configuration and design from a well trained technician. These processes have to be carried out for each application and each change in the environment. Our robotic perception system is evaluated on the BOP benchmark and consists of two elements. First, we design building blocks, which are algorithms and datasets available for our configuration algorithm. Second, we implement agents (configuration algorithms) which are designed to intelligently interact with our building blocks. On an examplary industrial robotic picking problem we show, that our autonomous engineering system can reduce engineering efforts.
Download
Paper Nr: 44
Title:

PrendoSim: Proxy-Hand-Based Robot Grasp Generator
Authors:

Diar Abdlkarim, Valerio Ortenzi, Tommaso Pardi, Maija Filipovica, Alan M. Wing, Katherine J. Kuchenbecker and Massimiliano Di Luca
Abstract: The synthesis of realistic robot grasps in a simulated environment is pivotal in generating datasets that support sim-to-real transfer learning. In a step toward achieving this goal, we propose PrendoSim, an open-source grasp generator based on a proxy-hand simulation that employs NVIDIA’s physics engine (PhysX) and the recently released articulated-body objects developed by Unity (https://prendosim.github.io). We present the implementation details, the method used to generate grasps, the approach to operationally evaluate stability of the generated grasps, and examples of grasps obtained with two different grippers (a parallel jaw gripper and a three-finger hand) grasping three objects selected from the YCB dataset (a pair of scissors, a hammer, and a screwdriver). Compared to simulators proposed in the literature, PrendoSim balances grasp realism and ease of use, displaying an intuitive interface and enabling the user to produce a large and varied dataset of stable grasps.
Download
Paper Nr: 46
Title:

Empirical Evaluation of a Novel Lane Marking Type for Camera and LiDAR Lane Detection
Authors:

Sven Eckelmann, Toralf Trautmann, Xinyu Zhang and Oliver Michler
Abstract: Highly automated driving requires a zero-error interpretation of the current vehicle environment utilizing state of the art environmental perception based on camera and Light Detection And Ranging (LiDAR) sensors. An essential element of this perception is the detection of lane markings, e.g. for lane departure warnings. In this work, we empirically evaluate a novel kind of lane marking, which enhances the contrast (artificial light-dark boundary) for cameras and 3D retro reflective elements guarantee a better reflection for light beams from a LiDAR. Thus intensity of point data from LiDAR is regarded directly as a feature for lane segmentation. In addition, the 3D lane information from a 2D camera is estimated using the intrinsic and extrinsic camera parameters and the lane width. In the frame of this paper, we present the comparison between the detection based on camera and LiDAR as well as the comparison between conventional and the new lane marking in order to improve the reliability of lane detection for different sensors. As a result, we are able to demonstrate that the track can be detected safely with the LiDAR and the new lane marking.
Download
Paper Nr: 88
Title:

A Flexible Structured Solver for Continuous-time Algebraic Riccati Equations
Authors:

Vasile Sima
Abstract: The solution of algebraic Riccati equations (AREs) is a fundamental computation in optimal control and other domains. The available solvers lack the flexibility in choosing a solution technique, or specifying options and parameters. The quality of a computed solution depends not only on the problem conditioning, but also on the various decisions made by the solver designer. This paper proposes a flexible solver for continuous-time AREs that allows the user to choose among several structured solution approaches, orthogonalization methods, and balancing options and parameters. No selection ensures the best results for all problems. Therefore, it is sometimes useful to try alternative pathways and find the best solution. The new solver has been used to solve the examples from the SLICOT CAREX benchmark collection. The numerical results in extensive tests illustrate the good performance of the proposed flexible solver.
Download
Paper Nr: 96
Title:

Learning-based Optimal Control of Constrained Switched Linear Systems using Neural Networks
Authors:

Lukas Markolf and Olaf Stursberg
Abstract: This work considers (deep) artificial feed-forward neural networks as parametric approximators in optimal control of discrete-time switched linear systems with controlled switching. The proposed approach is based on approximate dynamic programming and allows the fast computation of (sub-)optimal discrete and continuous control inputs, either by approximating the optimal cost-to-go functions or by approximating the optimal discrete and continuous input policies. An important property of the approach is the satisfaction of polytopic state and input constraints, which is crucial for ensuring safety, as required in many control applications. A numeric example is provided for illustration and evaluation of the approaches.
Download
Short Papers
Paper Nr: 3
Title:

On the Design and Fabrication of a Voice-controlled Mobile Robot Platform
Authors:

Saleh Ahmad, Mohammed K. Alhammadi, Abdulla A. Alamoodi, Ahmed N. Alnuaimi, Saif A. Alawadhi and Abdulla A. Alsumaiti
Abstract: The ‘voice’ is the most popular form of communication for human beings is. In the field of service robots, the application of speech recognition is a natural choice. Service robots minimize the physical work that people render with their day-to-day tasks. The development of a mobile robot platform that can be controlled using speech commands is presented in this paper. The human voice commands are recognized by the AI JARVIS voice recognition software and translated into motion commands sent to the mobile robot platform using RF communication. Not only does the mobile robot recognize the voice commands and execute them, but it also provides acknowledgment by voice output. The designed mobile robot can carry out various movements, turns, and shifting objects from one location to another. The voice commands are processed using the modified AI JARVIS voice recognition software. Using an RF module's wireless end-point networking, voice signal commands are directly transmitted to the Robot. The mobile robot is developed on a Microcontroller based platform. Performance evaluation showed promising results from the initial experiments. Possible improvements in the future deployment of such a robot in households, hospitals, and other sectors are also explored.
Download
Paper Nr: 11
Title:

Optimization-based or AI Task Planning for Scenarios with Cooperating Mobile Manipulators?
Authors:

Stefan-Octavian Bezrucav, Yifan Liu and Burkhard Corves
Abstract: Task planning has become one of the most important components of the control system for teams of cooperating robotic systems in complex scenarios. It plays such a critical role, as it must determine, order and assign a high variety of different tasks to the involved actors, such that at the end, the goals are reached while some metric, as the total execution time, is minimized. In this paper the analysis of three task planning approaches, mixed-integer linear programming (MILP), constraint programming (CP) and automated temporal planning (TP), with respect to three criteria is targeted. These criteria are the CPU time for plan generation, the plan makespan and the flexibility of the modelling approach. For the analysis, an intricate scenario in a kitchen environment with a team of multiple mobile manipulators is developed. The models and results are then compared to derive the advantages and disadvantages of each strategy.
Download
Paper Nr: 34
Title:

Task-motion Planning via Tree-based Q-learning Approach for Robotic Object Displacement in Cluttered Spaces
Authors:

Giacomo Golluccio, Daniele Di Vito, Alessandro Marino, Alessandro Bria and Gianluca Antonelli
Abstract: In this paper, a Reinforcement Learning approach to the problem of grasping a target object from clutter by a robotic arm is addressed. A layered architecture is devised to the scope. The bottom layer is in charge of planning robot motion in order to relocate objects while taking into account robot constraints, whereas the top layer takes decision about which obstacles to relocate. In order to generate an optimal sequence of obstacles according to some metrics, a tree is dynamically built where nodes represent sequences of relocated objects and edge weights are updated according to a Q-learning-inspired algorithm. Four different exploration strategies of the solution tree are considered, ranging from a random strategy to a ε-Greedy learning-based exploration. The four strategies are compared based on some predefined metrics and in scenarios with different complexity. The learning-based approaches are able to provide optimal relocation sequences despite the high dimensional search space, with the ε-Greedy strategy showing better performance, especially in complex scenarios.
Download
Paper Nr: 37
Title:

Genetic Optimization of Excitation Signals for Nonlinear Dynamic System Identification
Authors:

Volker Smits and Oliver Nelles
Abstract: Two new methods for optimization of passive step-based excitation signals for system identification of nonlinear dynamic processes via a genetic algorithm are introduced - an optimized Amplitude Pseudo Random Binary Signal (APRBSOpt) and a Genetic Optimized Time Amplitude Signal (GOATS). The investigated optimization objectives are the evenly excitation of all frequencies and the uniform data distribution of the space spanned by the system’s input and output. The results show that the GOATS optimized according to the uniform data distribution outperform the state-of-the-art excitation signals standard ARPBS (APRBSStd), Optimized Nonlinear Input Signal (OMNIPUS), Chirp and Multi-Sine in the achieved model quality on three artificially created Single-Input Single-Output (SISO) nonlinear dynamic processes. However, the APRBSOpt only exceeds the Chirp, Multi-Sine and APRBSStd in the achievable model quality. Additionally, the GOATS can be used for stiff systems, supplementing existing data and easy incorporation of constraints.
Download
Paper Nr: 42
Title:

Stability Analysis for State Feedback Control Systems Established as Neural Networks with Input Constraints
Authors:

Lukas Markolf and Olaf Stursberg
Abstract: Considerable progress in deep learning has also lead to an increasing interest in using deep neural networks (DNN) for state feedback in closed-loop control systems. In contrast to other purposes of DNN, it is insufficient to consider them only as black box models in control, in particular, when used for safety-critical applications. This paper provides an approach allowing to use the well-established indirect method of Lyapunov for time-invariant continuous time nonlinear systems with neural networks as state feedback controllers in the loop. A key element hereto is the derivation of a closed-form expression for the partial derivative of the neural network controller with respect to its input. By using activation functions of the type of sigmoid functions in the output layer, the consideration of box-constrained inputs is further ensured. The proposed approach does not only allow to verify the asymptotic stability, but also to find Lyapunov functions which can be used to search for positively invariant sets and estimates for the region of attraction.
Download
Paper Nr: 49
Title:

The ALNS Metaheuristic for the Maintenance Scheduling Problem
Authors:

David Woller and Miroslav Kulich
Abstract: Transmission maintenance scheduling (TMS) is an important optimization problem in the electricity distribution industry, with numerous variants studied and methods proposed over the last three decades. The ROADEF challenge 2020 addresses a novel version of the TMS problem, which stands out by having multiple time-dependent properties, constraints, and a risk-based aggregate objective function. Therefore, the problem is more complex than the previous formulations, and the existing methods are not directly applicable. This paper presents a method based on the Adaptive Large Neighborhood Search metaheuristic. The method is compared with the best-known solutions from the challenge qualification phase, in which more than 70 teams participated. The result shows that the method yields consistent performance over the whole dataset, as the method finds the best-known solutions for half of the dataset and finds solutions consistently within 5‰ gap.
Download
Paper Nr: 52
Title:

Towards Real Time Bottleneck Detection using Miniterms
Authors:

J. Llopis, A. Lacasa, E. Garcia and N. Montés
Abstract: The Sub-Bottleneck concept is introduced in this article for the first time. The literature defines the bottleneck concept through the cycle time in which, as a general rule, the slowest machine with longer cycle time, is classified as a bottleneck. Depending on the cycle time, the machine, the production line, the plant taken into account, etc., the literature has defined the concept of bottleneck in plant, bottleneck in production line, bottleneck in machine, etc. This article presents the Sub-Bottleneck concept for the first time. This concept uses the mini-term, a cycle time of each component that makes up a machine to determine which is the slowest and focus on future improvements that will optimize the efficiency of the production line. In order to validate this proposal, the mini-terms have been implemented in a production line at the Ford factory in Almussafes (Valencia, Spain), made up of 4 welding robots. The tests show the variable nature of the components and that the typical bottleneck studied in the literature does not have to coincide with the Sub-Bottleneck concept.
Download
Paper Nr: 54
Title:

Identification of Gait Phases with Neural Networks for Smooth Transparent Control of a Lower Limb Exoskeleton
Authors:

Vittorio Lippi, Cristian Camardella, Alessandro Filippeschi and Francesco Porcini
Abstract: Lower limbs exoskeletons provide assistance during standing, squatting, and walking. Gait dynamics, in particular, implies a change in the configuration of the device in terms of contact points, actuation, and system dynamics in general. In order to provide a comfortable experience and maximize performance, the exoskeleton should be controlled smoothly and in a transparent way, which means respectively, minimizing the interaction forces with the user and jerky behavior due to transitions between different configurations. A previous study showed that a smooth control of the exoskeleton can be achieved using a gait phase segmentation based on joint kinematics. Such a segmentation system can be implemented as linear regression and should be personalized for the user after a calibration procedure. In this work, a nonlinear segmentation function based on neural networks is implemented and compared with linear regression. An on-line implementation is then proposed and tested with a subject.
Download
Paper Nr: 58
Title:

UAV Inspection of Large Components: Indoor Navigation Relative to Structures
Authors:

Martin Schörner, Michelle Bettendorf, Constantin Wanninger, Alwin Hoffmann and Wolfgang Reif
Abstract: The inspection of large structures is increasingly carried out with the help of Unmanned Aerial Vehicles (UAVs). When navigating relative to the structure, multiple data sources can be used to determine the position of the UAV. Examples include track data from an installed camera and sensor data from the orientation sensors of the UAV. This paper deals with the fusion of this data and its use for navigation alongside the structure. For the sensor fusion, a concept is developed using a Kalman filter and evaluated simulatively in a prototype. The calculated position data are also fed into a vector flight control system, which dynamically calculates and flies a trajectory along the component using the potential field method. This is done taking into account obstacles detected by the onboard sensors of the UAV. The established concept is then implemented with the Robot Operating System (ROS) and evaluated simulatively.
Download
Paper Nr: 64
Title:

Plan Recovery Process in Multi-agent Dynamic Environments
Authors:

Leonardo H. Moreira and Célia G. Ralha
Abstract: Planning is the process that focuses on the choice and organization of actions through their expected effects. Plans can be affected by unexpected, uncontrolled, non-deterministic events leading to failures. Such challenging problem boosted works focusing agent distribution, communication mechanisms, privacy, among other issues. Nevertheless, the plan recovery process does not have a defined standard solution. Thus, in this work, we present a three-phase plan recovery process to provide resilience to agent plans by supporting a staggered solution. Whenever an action execution fails, agents try to solve individually through their own capabilities. But when not possible, agents start an interaction protocol to ask for help. Finally, when previous two phases were unsuccessful, a centralized planning process is trigged. Regardless the phase in which the solution is found, agents’ plans are coordinated to guarantee cooperation maintaining information privacy. An empirical analysis applying metrics such as planning time, final plan length and message exchange was conducted. Results give statistical significant evidence that agents’ autonomy is better explored in agents’ loosely coupled environments. The contributions of this work include: a three-phase plan recovery process, a simulation tool for benchmarks, and a statistical robust evaluation method to multi-agent planning.
Download
Paper Nr: 79
Title:

Online Facility Service Leasing Inspired by the COVID-19 Pandemic
Authors:

Christine Markarian and Peter Khallouf
Abstract: In response to resource shortages caused by the COVID-19 pandemic, many communities have been leasing health facilities such as hospitals, clinics, and other centers in order to meet the needs of their patients. The goals have been two-folded: leasing costs had to be optimized and patients had to be served as soon as possible. Decisions as to when to lease which services at which facility locations shaped the success of these goals. At the heart of these decisions lies a complex optimization problem, which we call the Online Non-metric Facility Service Leasing problem (non-metric OFSL), a generalization of the well-known Online Non-metric Facility Leasing problem (non-metric OFL) in which facility locations are leased for different facility-time durations. In non-metric OFSL, each facility location may provide a number of services leased for different service-time durations. Additionally, each service is associated with a dormant fee that needs to be paid for each day at which the service is not leased. The optimization goal is to minimize the total leasing costs, dormant fees, and the distances between patients and the facilities they are connected to. We develop the first online algorithm for non-metric OFSL, evaluated using the notion of competitive analysis. The latter is a worst-case analysis used to measure the quality of online algorithms, in which the online algorithm’s output is compared to the optimal offline solution for all instances of the problem.
Download
Paper Nr: 107
Title:

System Proposal for Integrating Quality Control Data of Components of the Brazilian Oil and Gas Industry
Authors:

Mario Ricardo Nascimento Marques Junior, Eder Mateus Nunes Gonçalves, Silvia Silva Costa D. Botelho, Emanuel Silva Diaz D. Estrada, Danubia Bueno Espindola, Eduardo Nunes Borges, Werner Luft Botelho, Bruno Machado Lobell and Lucas Silva Marca
Abstract: Databooks are the main documents for monitoring and validating a work in the oil industry. However, the analysis of these Databooks requires many man-hours of professionals, and will still be subject to failures during these analyzes. An approach to optimize this task would be to use an intelligent search and validation system for this documentation. Once the documents are scanned, algorithms could assist the professional to analyze these Databooks, based on automatic indexing and checks. A problem that arises from this approach is the difficulty to develop a valid conceptual model for all Databooks found and all types of components and structures subject to inspection procedures. As this modeling is done manually by professionals, this task becomes difficult, slow and often inefficient. To mitigate this problem, this work proposes a system to automatically generate a conceptual model from textual descriptions of the inspection elements, making this task simpler. This is done from the development of an ontology that describes the standardized knowledge on inspection of different types of components of the structure of works for oil platforms for the validation of inspection reports and quality certificates regarding criteria of completeness and compliance.
Download
Paper Nr: 109
Title:

Empirical Analysis of the Impact of Additional Padding on the Collaborative Robot Velocity Behavior in Transient Contact Cases
Authors:

Christopher Schneider, Maximilian M. Seizmeir, Thomas Suchanek, Martina Hutter-Mironovová, Mohamad Bdiwi and Matthias Putz
Abstract: In this paper, a suitable measurement setup is presented and applied to conduct force and pressure measurements for transient contact cases with the shoulder at the example of lathe machine tending. Empirical measurements were executed on a selected collaborative robot’s behavior regarding allowable operating speeds under consideration of sensor sensitivity, robot collision geometry, and damping materials. Comparisons between the theoretic calculations proposed in ISO/TS 15066 and the practical measurement results present a basis for future research. With the created database, preliminary risk assessment and economic assessment procedures of collaborative machine tending cells can be facilitated.
Download
Paper Nr: 122
Title:

Agent-based Intelligent KPIs Optimization of Public Transit Control System
Authors:

Nabil Morri, Sameh Hadouaj and Lamjed Ben Said
Abstract: Public transit has a wide variety of resources. There is an infrastructure including stations and routes with multiple trips provided by different modes of transportation (metro, subway, bus). These resources must be well exploited to ensure good quality of service to passengers and especially against perturbations that may occur during the day. The contribution of this work is to model and implement a transit control system that detects perturbations and finds, through optimization, the best regulation action while respecting the constraints of the traffic situation. This system combines various measures of Key Performance Indicators (KPIs) into a single performance value, covering several dimensions depending on the type of service quality to be guaranteed. To take into account the complex and dynamic nature of transportation systems, a multi-agent approach is adopted in the modelling of our system. The validation is based on real traffic data. The results show better performance of our system compared to the current resolution.
Download
Paper Nr: 125
Title:

Control System Design via Constraint Satisfaction using Convolutional Neural Networks and Black Hole Optimization
Authors:

Saber Yaghoobi and M. S. Fadali
Abstract: This paper proposes a new approach to control system design through solving a Constraint Satisfaction Problem (CSP) using artificial intelligence, first using a genetic algorithm then using a Convolutional Neural Network (CNN). The genetic algorithm determines the feasible controller parameters by minimizing a cost function subject to inequality design constraints. The CNN-finds the parameters by designing a deep neural network. It is shown that the evolutionary optimization algorithm converges almost surely to the optimal solution. To demonstrate the methodologies, they are applied to the design of PID controllers for linear and nonlinear systems. Two examples are presented, an armature-controlled DC motor and Bouc-Wen nonlinear hysteresis model. Simulations results show that the proposed methods yield solutions that satisfy design specifications.
Download
Paper Nr: 7
Title:

Revisiting Johann Bernoulli's Method for the Brachistochrone Problem
Authors:

Ido Braun and Joseph Z. Ben-Asher
Abstract: This paper reviews Johann Bernoulli's solution to the Brachistochrone problem, using an analogy to the movement of light and Fermat's principle of least time. Bernoulli's method is later used to derive solutions to some generalizations of the Brachistochrone problem. The problems solved using Bernoulli's method are the classical flat gravity Brachistochrone, spherical gravity outside the earth, and spherical gravity inside the earth ('gravity train').
Download
Paper Nr: 23
Title:

Model Predictive Control: A Survey of Dynamic Energy Management
Authors:

Nsilulu T. Mbungu, Raj M. Naidoo, Ramesh C. Bansal and Mukwanga W. Siti
Abstract: This paper presents the structure of the model predictive control (MPC), its development and application through optimal energy system. The MPC is one of the algorithms that are used in a computer controlled environment to predict the future behaviour of an explicit process model. It is devised by computing and adjusting the next sequence of the input variables at each control interval. The MPC is an algorithm in which the challenge is to optimize the behaviour of a future plant. The optimization sequence starts by sending the first input into the plant and then at each subsequent control interval the entire computation is repeated to reach the performance index function to follow. MPC offers a variety of applications in a wide range of industries. This is due to its robustness in the optimal control design of a process. MPC is also widely used in aerospace, automotive, chemical and food processing applications. This study describes the implementation of the energy management scheme through the use of MPC design.
Download
Paper Nr: 101
Title:

Searching & Generating Discrete-Event Systems
Authors:

T. J. Helliwell, B. Morgan and M. Mahfouf
Abstract: In this paper we introduce a new method for the automatic generation and computer experimentation of Discrete-Event Systems (DES). We introduce the concept that DES descriptions may be used to define a searchable configuration space. Configuration instances in this space a represented as a permutation encoding which shows the number-of and types-of resources in a given configuration. Each instance is checked that the number of resources does not exceed a maximum and whether a fixed set of processes (a decomposed goal of uncertain time intervals drawn from a Gaussian distribution) can be logically completed on the given set of resources. If the permutation instance satisfies these constraints, it is subsequently constructed as a simulation model to quantify completion through global features of makespan and total processing time. We claim this is the basis for a powerful tool in high-level informed design of these types of systems that have hitherto avoided autonomous description or have been previously individually designed using time consuming manually defined programs.
Download

Area 3 - Robotics and Automation

Full Papers
Paper Nr: 6
Title:

Edge and Corner Detection in Unorganized Point Clouds for Robotic Pick and Place Applications
Authors:

Mohit Vohra, Ravi Prakash and Laxmidhar Behera
Abstract: In this paper, we propose a novel edge and corner detection algorithm for an unorganized point cloud. Our edge detection method classifies a query point as an edge point by evaluating the distribution of local neighboring points around the query point. The proposed technique has been tested on generic items such as dragons, bunnies, and coffee cups from the Stanford 3D scanning repository. The proposed technique can be directly applied to real and unprocessed point cloud data of random clutter of objects. To demonstrate the proposed technique’s efficacy, we compare it to the other solutions for 3D edge extractions in an unorganized point cloud data. We observed that the proposed method could handle the raw and noisy data with little variations in parameters compared to other methods. We also extend the algorithm to estimate the 6D pose of known objects in the presence of dense clutter while handling multiple instances of the object. The overall approach is tested for a warehouse application, where an actual UR5 robot manipulator is used for robotic pick and place operations in an autonomous mode.
Download
Paper Nr: 8
Title:

Stiffness Modeling of Compliant Serial Manipulators based on Tensegrity Mechanism under External Loading
Authors:

Wanda Zhao, Anatol Pashkevich and Damien Chablat
Abstract: The paper focuses on the stiffness modeling of a new type of compliant manipulator and its non-linear behavior under external loading. The manipulator under study is a serial mechanical structure composed of dual-triangle segments. The main attention is paid to the possible equilibriums and the manipulator stiffness behavior under the loading for the initial non-straight configuration. It was demonstrated that there is a quasi-buckling phenomenon for this manipulator while the external loading increasing. In the neighborhood of these configurations, the manipulator behavior was analyzed using the enhanced Virtual Joint Method (VJM). Relevant simulation study confirmed the obtained theoretical results.
Download
Paper Nr: 48
Title:

Optimization-based Trajectory Prediction Enhanced with Goal Evaluation for Omnidirectional Mobile Robots
Authors:

Wei Luo and Peter Eberhard
Abstract: In this paper, an optimization-based trajectory prediction enhanced with goal evaluation for omnidirectional mobile robots is proposed. The proposed approach tries to predict the mobile platform’s trajectory based on its previous positions. A two-stage strategy is introduced. At the first stage, the likely goal of the robot in the scenario is evaluated based on an improved Bayesian framework, which also predicts the possible waypoints in a discrete roadmap based on Monte-Carlo sampling in the future. Then, based on the predicted waypoints, an optimization problem is formulated based on the complementary progress constraints, the system dynamics, and the model constraints. After solving the proposed optimization problem, a more reasonable predicted trajectory can be generated. At the end, an experimental scenario is set up, and it is verified with the experimental data, whether the trajectories can be predicted well.
Download
Paper Nr: 55
Title:

Fractional Order Tracking Control of Unmanned Aerial Vehicle in Presence of Model Uncertainties and Disturbances
Authors:

Heera L. Maurya, Padmini Singh, Subhash C. Yogi, Laxmidhar Behera and Nishchal K. Verma
Abstract: An unmanned Aerial Vehicle (UAV) is a highly non-linear unstable system. In this work using fractional order calculus, a novel fractional order dynamics of UAV is proposed. The concept of fractional order depicts the more realistic behavior of UAVs. For proposed fractional order model, a fractional order sliding mode controller (SMC) is designed such that the desired path can be achieved by the UAV in finite-time. In addition to this model uncertainty and disturbance is considered in the system which is handled by the proposed robust SMC. Stability analysis is given for the fractional order SMC using fractional Lyapunov method. Simulations have been done for position and attitude tracking of UAV to demonstrate the efficacy of the proposed method.
Download
Paper Nr: 59
Title:

The Influence of the Gear Reduction Ratio on the Free-floating Space Manipulator’s Dynamics
Authors:

Mateusz Wojtunik and Karol Seweryn
Abstract: Utilisation of space manipulator mounted on the satellite is one the main methods for the proposed Active Debris Removal and On-Orbit Servicing missions. Precise numerical models of the manipulator’s joint are very important as its dynamics has a strong effect on the behaviour of the system including the base where it is mounted. One of aspects that can be considered is the extension of manipulator’s dynamical equations with gear kinematic constraints. To achieve this goal, dynamical equations of motion for planar 3DoF free-floating manipulator with gear kinematic constraints are presented in this paper. Open-loop analysis is performed to form conclusions concerning the influence of the gear reduction ratio on space manipulator’s dynamics. Torques required to perform end-effector straight line trajectory are evaluated using inverse dynamics path planning algorithm and then utilised as motor driving torques for different gear reduction ratios. It appears that the gear reduction ratio influences the system mass matrix nonlinearly causing the end-effector trajectory to deviate from the straight line. These deviations are already observed for relatively low gear reduction ratios.
Download
Paper Nr: 65
Title:

Path Planning in Unstructured Urban Environments for Self-driving Cars
Authors:

Anderson Mozart, Gabriel Moraes, Rânik Guidolini, Vinicius B. Cardoso, Thiago Oliveira-Santos, Alberto F. De Souza and Claudine Badue
Abstract: We present a path planner for unstructured urban environments (PPUE) for self-driving cars. PPUE receives initial and goal poses as input, as well as maps of the environment. It employs a hybrid A* algorithm with two heuristics for generating paths, which are smoothed using Conjugate Gradient optimization. Different from previous works, PPUE uses: (i) an obstacle distance grid-map, instead of an occupancy grid-map, for representing the environment; and (ii) an accurate but simple collision model of the car. We have examined PPUE’s performance experimentally in simulated and real world scenarios. Our results show that PPUE computes smooth and safe paths, which follow the kinematic constraints of the vehicle, fast enough for suitable real world operation.
Download
Paper Nr: 83
Title:

A Robust CNN Training Approach to Address Hierarchical Localization with Omnidirectional Images
Authors:

Juan J. Cabrera, Sergio Cebollada, Luis Payá, María Flores and Oscar Reinoso
Abstract: This paper reports and evaluates the training optimization process of a Convolutional Neural Network (CNN) with the aim of addressing the localization of a mobile robot. The proposed method addresses the localization problem by means of a hierarchical approach by using a visual sensor that provides omnidirectional images. In this sense, AlexNet is adapted and re-trained with a twofold purpose. First, the rough localization step consists of a room retrieval task. Second, the fine localization step within the retrieved room is carried out by means of a nearest neighbour search by comparing a holistic descriptor obtained from the CNN with the visual model of the retrieved room. The novelty of the present work lies in the use of a CNN and holistic descriptors obtained from raw omnidirectional images as captured by the vision system, with no panoramic conversion. In addition, this work proposes the use of a data augmentation technique and a Bayesian optimization to address the training process of the CNN. These approaches constitute an efficient and robust solution to the localization problem, as shown in the experimental section even in presence of substantial changes of the lighting conditions.
Download
Paper Nr: 95
Title:

Energy-based Control for Soft Manipulators using Cosserat-beam Models
Authors:

Brandon Caasenbrood, Alexander Pogromsky and Henk Nijmeijer
Abstract: In this work, we describe an energy-based control method for under-actuated soft manipulators. The continuous dynamics of the soft robot are modeled by the differential geometry of Cosserat beams. Through a finite-dimensional truncation, a reduced port-Hamiltonian model is obtained that preserves desirable passivity conditions. Exploiting the passivity, we propose a stabilizing energy-shaping controller that ensures the potential energy is minimal at the desired end-effector configuration. Finally, the effectiveness of the energy-based controller is demonstrated through simulations of a soft manipulator inspired by the tentacle of an octopus.
Download
Paper Nr: 105
Title:

Design Optimization of a Manipulator for CERN’s Future Circular Collider (FCC)
Authors:

Hannes Gamper, Hubert Gattringer, Andreas Müller and Mario Di Castro
Abstract: CERN is often confronted with very specialized automation problems in hazardous, radioactive and semi-structured environments for its particle accelerators, test rigs or other experiments. These frequently lead to specific requirements that do not allow the usage of common industrial robots. Thus, a design problem with almost no restrictions on the actual robot topology, but very hard requirements concerning workspace, allowed robot space, payload, robot weight and accuracy (due to elasticity/error propagation) has to be solved. This paper reports an approach to this problem, which was applied to find an optimal robotic design for inspection and maintenance tasks in CERN’s Future Circular Collider (FCC).
Download
Paper Nr: 119
Title:

Domain Adaptation in LiDAR Semantic Segmentation by Aligning Class Distributions
Authors:

Iñigo Alonso, Luis Riazuelo, Luis Montesano and Ana C. Murillo
Abstract: LiDAR semantic segmentation provides 3D semantic information about the environment, an essential cue for intelligent systems, such as autonomous vehicles, during their decision making processes. Unfortunately, the annotation process for this task is very expensive. To overcome this, it is key to find models that generalize well or adapt to additional domains where labeled data is limited. This work addresses the problem of unsupervised domain adaptation for LiDAR semantic segmentation models. We propose simple but effective strategies to reduce the domain shift by aligning the data distribution on the input space. Besides, we present a learning-based module to align the distribution of the semantic classes of the target domain to the source domain. Our approach achieves new state-of-the-art results on three different public datasets, which showcase adaptation to three different domains.
Download
Paper Nr: 127
Title:

Singularity Avoidance of Task-redundant Robots in Pointing Tasks: On Nullspace Projection and Cardan Angles as Orientation Coordinates
Authors:

Moritz Schappler and Tobias Ortmaier
Abstract: Robot manipulators are often deployed in tool-symmetric tasks, which only requires defining end effector position and pointing direction. In this case six-axis serial industrial robots and full-mobility (spatial) parallel robots have one degree of task redundancy. Using Cardan angles as orientation coordinates, a unified formulation of the position-level and second-order inverse kinematics problem is set up for both robot types. An efficient scheme for difference-quotient approximation of gradients of performance criteria for projection into the task redundancy’s nullspace is presented. The simulation example of a hexapod robot shows that avoiding and exiting parallel robot singularities of type II is possible with the nullspace of all joints. The nullspace controller scheme can be used in offline trajectory optimization and in online motion generation.
Download
Short Papers
Paper Nr: 2
Title:

Flying Wing Drones based on Cricket Antennas
Authors:

Walid Hassairi and Mohamed Abid
Abstract: Drones represent an important part of the shipsets’ domain. There are different application areas and depending on the field of application, problems of stability, trajectory correction and autonomy arise. The flying wings drones are one of the drones’ categories inspired from birds flying technique. This category of drones has several problems quite different from the classical drones. Among these problems we can identify the drone hunter issue. To solve this problem, we propose a solution inspired from the wood crickets. In fact, the crickets are extremely fast as they can process information locally. They have a kind of “back brains” that process the information locally and control the movement of their legs. Unlike human who strictly send all information to the main brain that treat them and make a reaction, the cricket has several brains inside the body, so that it can send the information about the airflow to small brains behind its legs. These little back brains not only process the information about the airflow that comes from the crests and their multiple hairs, but also controls the movement of the rear legs. This unusual performance of the crickets’ crests hair was the origin of our research contribution. We therefore propose a biometric flow camera based on several electronic hairs connected together. We have selected REMANTA as a winged drone to implement our proposed solution. We will integrate our micro-sensors in this 10 cm dimensions drone to solve three problems: trajectory correction, stability, and enemy avoidance.
Download
Paper Nr: 15
Title:

A Genetic Algorithm for HMI Test Infrastructure Fine Tuning
Authors:

Lukas Rosenbauer, Anthony Stein and Jörg Hähner
Abstract: Human machine interfaces (HMI) have become a part of our daily lives. They are an essential part of a variety of products ranging from computers over smart phones to home appliances. Customer’s requirements for HMIs are rising and so does the complexity of the devices. Several years ago, many products had a rather simple HMI such as mere buttons. Nowadays lots of devices have screens that display complex text messages and a variety of objects such as icons. This leads to new challenges in testing, the goal of which it is to ensure quality and to find errors. We combine a genetic algorithm with computer vision techniques in order to solve two testing use cases located in the automated verification of displays. Our method has a low runtime and can be used on low budget equipment such as Raspberry Pi which reduces the operational cost in practice.
Download
Paper Nr: 39
Title:

Towards Real Time Predictive System for Mechanical Stamping Presses to Assure Correct Slide Parallelism
Authors:

Ivan Peinado-Asensi, N. Montes and E. García
Abstract: Automotive companies are going through a rough time due to the decrease in the car sales market, therefore OEMs trend is cost reduction in the next years over improving efficiency increasing digitalization, implementing new industry 4.0 technologies to turn their facilities in smart factories. Within car manufacturing processes, stamping present many possibilities for development, in this paper an approach to bring stamping plants closer to smart factories is presented. The most common problems in stamping are unexpected breakdowns in equipment and poor quality parts produced, to avoid these problems corrective and predictive maintenance tasks are carried out to improve presses and tools performance. One of the critical maintenance tasks in press machines are parallelism, a malfunction in the kinetic transmission can lead to high cost and duration breakdowns. To monitor machine working parameters a novel method is presented using IIoT techniques, having access to machine working parameters in Real-Time to predict machine malfunction in order to reduce the number of breakdowns.
Download
Paper Nr: 40
Title:

Wearable MIMUs for the Identification of Upper Limbs Motion in an Industrial Context of Human-Robot Interaction
Authors:

Mattia Antonelli, Elisa Digo, Stefano Pastorelli and Laura Gastaldi
Abstract: The automation of human gestures is gaining increasing importance in manufacturing. Indeed, robots support operators by simplifying their tasks in a shared workspace. However, human-robot collaboration can be improved by identifying human actions and then developing adaptive control algorithms for the robot. Accordingly, the aim of this study was to classify industrial tasks based on accelerations signals of human upper limbs. Two magnetic inertial measurement units (MIMUs) on the upper limb of ten healthy young subjects acquired pick and place gestures at three different heights. Peaks were detected from MIMUs accelerations and were adopted to classify gestures through a Linear Discriminant Analysis. The method was applied firstly including two MIMUs and then one at a time. Results demonstrated that the placement of at least one MIMU on the upper arm or forearm is suitable to achieve good recognition performances. Overall, features extracted from MIMUs signals can be used to define and train a prediction algorithm reliable for the context of collaborative robotics.
Download
Paper Nr: 47
Title:

Real-time Robust Trajectory Control for Vehicle Platoons: A Linear Matrix Inequality-based Approach
Authors:

Alessandro Bozzi, Enrico Zero, Roberto Sacile and Chiara Bersani
Abstract: This paper proposes a solution to dynamically adjust vehicle platoon trajectories. The goal of the control algorithm is to keep the optimal interdistance between adjacent vehicles proceeding at cruising speed on a straight road. After a proposal of the interdistance required between neighboring vehicles, a robust decentralized controller based on a linear control law provides the speed profile for each component of the platoon. Its objective is to minimize the divergence in space in respect to the planned trajectories while assuring a safe span between adjacent members of the platoon. The results on a limited instance demonstrate the effectiveness of the proposed approach.
Download
Paper Nr: 51
Title:

Automatic 3D Object Recognition and Localization for Robotic Grasping
Authors:

Bruno Santo, Liliana Antão and Gil Gonçalves
Abstract: With the emergence of Industry 4.0 and its highly re-configurable manufacturing context, the typical fixed-position grasping systems are no longer usable. This reality underlined the necessity for fully automatic and adaptable robotic grasping systems. With that in mind, the primary purpose of this paper is to join Machine Learning models for detection and pose estimation into an automatic system to be used in a grasping environment. The developed system uses Mask-RCNN and Densefusion models for the recognition and pose estimation of objects, respectively. The grasping is executed, taking into consideration both the pose and the object’s ID, as well as allowing for user and application adaptability through an initial configuration. The system was tested both on a validation dataset and in a real-world environment. The main results show that the system has more difficulty with complex objects; however, it shows promising results for simpler objects, even with training on a reduced dataset. It is also able to generalize to objects slightly different than the ones seen in training. There is an 80% success rate in the best cases for simple grasping attempts.
Download
Paper Nr: 53
Title:

Context-aware Social Robot Navigation
Authors:

Frederik Haarslev, William K. Juel, Avgi Kollakidou, Norbert Krüger and Leon Bodenhagen
Abstract: With the emergence of robots being deployed in unstructured environments outside the industrial domain, the importance of robots behaving appropriately in the vicinity of people is becoming more clear. These behaviours are hard to model as they depend on the social context. This context includes among other things where the robot is deployed, how crowded that place is, as well as who are residing in that place. In this paper we extend social space theory with the social context, making them adaptable to the current situation. We implement the social spaces as costmaps used in the standard ROS navigation stack. Our method – Context-Aware Social robot Navigation (CASN) – is tested in the context of people avoidance in social navigation. We compare CASN with the social navigation layer package, which also implements costs based on detected people. We show that by using CASN a mobile robot complies with social conventions in four different navigation scenarios.
Download
Paper Nr: 56
Title:

Evaluating the Influence of Feature Matching on the Performance of Visual Localization with Fisheye Images
Authors:

María Flores, David Valiente, Sergio Cebollada, Oscar Reinoso and Luis Payá
Abstract: Solving the localization problem is a crucial task in order to achieve autonomous navigation for a mobile robot. In this paper, the localization is solved using the Adaptive Probability-Oriented Feature Matching (APOFM) method, which produces robust matching data that permit obtaining the relative pose of the robot from a pair of images. The main characteristic of this method is that the environment is dynamically modelled by a 3D grid that estimates the probability of feature existence. The spatial probabilities obtained by this model are projected on the second image. These data are used to filter feature points in the second image by proximity to relevant areas in terms of probability. This approach improves the outlier rejection. This work aims to study the performance of this method using different types of local features to extract the visual information from the images provided by a fisheye camera. The results obtained with the APOFM method are evaluated and compared with the results obtained using a standard visual odometry process. The results determine that combining the APOFM method with ORB as local features provides the most efficient solution both to estimate relative orientation and translation, in contrast to SURF, KAZE and FAST feature detectors.
Download
Paper Nr: 87
Title:

Design and Development of a Dexterous Master Glove for Nuclear Waste Telemanipulation
Authors:

Florian Gosselin, Mathieu Grossard, Djibril Diallo, Benoit Perochon and Pascal Chambaud
Abstract: The rise of the nuclear industry in middle of the last century required the development of remotely controlled robotic solutions. Researches on radioactivity and its applications were initially performed in gloveboxes and hot cells with which operators can efficiently and safely access dangerous materials at distance using telemanipulators. Owing to the relatively limited variety of the objects used in such environments, and to the fact that they can usually be adapted for remote manipulation, it was possible to efficiently grasp them using purely mechanical or robotic 6 degrees of freedom (DoF) master-slave systems equipped with bi-digital grippers on the slave side and simple handles on the master side. Such solutions, which were perfectly adapted for handling a limited quantity and variety of radioactive material, are however no more sufficient when processing huge quantities of nuclear waste accumulated over time and/or produced at the occasion of dismantling operations occurring decades later at the end of the nuclear power plants lifecycle. The quantity and diversity of nuclear waste require more efficient and versatile systems. To answer this challenge and increase the operators’ productivity, we developed a novel dexterous master-slave system composed of a tri-digital master glove and a remotely controlled three fingers dexterous gripper. This paper presents the design and development of this master hand device. We first introduce its design rationale, then we present its electro-mechanical design, with details on the kinematics, actuators, sensors and controller, and finally its integration in a master-slave system which is used to validate its ability to perform dexterous telemanipulation.
Download
Paper Nr: 94
Title:

Automated Lane Change Decision Making in Highway using a Hybrid Approach
Authors:

Ozan Çaldıran, Engin Baglayici, Morteza Dousti, Eren Mungan, Enes E. Bulut, M. D. Demir and Furkan Koçyiğit
Abstract: This study proposes a decision-making model for lane changing and lane keeping decisions in highway autonomous driving. In order to perform a safe and efficient lane change, it is crucial to decide whether a lane change is needed, the desired lane is more suitable, and making a lane change maneuver is safe. In this work, we propose a model that is capable of assessing these considerations and suggest appropriate lane-change maneuvers. The model uses probabilistic utility functions and a deterministic but conservative gap selection method that considers not only the gaps in the target lane but also the vehicles in the driving lane. In addition to simulation tests, we integrated our model into a SUV vehicle that has 360-degree perception and motion control capabilities and performed autonomous highway driving to test real-life performance.
Download
Paper Nr: 98
Title:

Altitude Correction of an UAV Assisted by Point Cloud Registration of LiDAR Scans
Authors:

Marcus D. Forte, Polycarpo S. Neto, George P. Thé and Fabricio G. Nogueira
Abstract: This paper presents an online localization estimation of an Unmanned Aerial Vehicle (UAV) by fusing data provided by the on-board flight controller and a LiDAR (Light Detection and Ranging) carried by the UAV. Pose estimations solely obtained by the UAV are often corrupted by noise or instrumentation limitation, which may lead to erroneous mapping of the environment. To correct potential estimation errors, the LiDAR scans are assembled into a local point cloud history and matched against a partial map of the environment using a proposed point cloud registration method, similar to a Simultaneous Localization and Mapping (SLAM) approach. The resulting correction is incorporated into the estimation of the UAV using an asynchronous Kalman Filter implementation. For this work, only the altitude errors are corrected by the registration. We conducted tests on a local thermal power plant which contained three large coal stockpiles. We chose one of them as our Region of Interest (ROI).
Download
Paper Nr: 100
Title:

Concept of a Robotic System for Autonomous Coarse Waste Recycling
Authors:

Tim Tiedemann, Matthis Keppner, Tom Runge, Thomas Vögele, Martin Wittmaier and Sebastian Wolff
Abstract: The recycling of coarse waste such as construction and demolition waste (CDW), bulky waste, etc., is a process that is currently performed mechanically and manually. Unlike packaging waste, commercial waste and the like, which is usually cut or shredded into small pieces and then automatically separated and sorted on conveyor belt-based systems, coarse waste is separated by specialized personnel using wheel loaders, cranes or excavators. This paper presents the concept of a robotic system designed to autonomously separate recyclable coarse materials from bulky waste, demolition and construction waste, etc. The proposed solution explicitly uses existing heavy equipment (e.g., an excavator currently in use on-site) rather than developing a robot from scratch. A particular focus is set on the sensory system options used to identify and classify waste objects.
Download
Paper Nr: 106
Title:

Deep vs. Deep Bayesian: Faster Reinforcement Learning on a Multi-robot Competitive Experiment
Authors:

Jingyi Huang, Fabio Giardina and Andre Rosendo
Abstract: Deep Learning experiments commonly require hundreds of trials to properly train neural networks, often labeled as Big Data, while Bayesian learning leverages scarce data points to infer next iterations, also known as Micro Data. Deep Bayesian Learning combines the complexity from multi-layered neural networks to probabilistic inferences, and it allows a robot to learn good policies within few trials in the real world. In here we propose, for the first time, an application of Deep Bayesian Reinforcement Learning (RL) on a real-world multi-robot confrontation game, and compare the algorithm with a model-free Deep RL algorithm, Deep Q-Learning. Our experiments show that DBRL significantly outperforms DRL in learning efficiency and scalability. The results of this work point to the advantages of Deep Bayesian approaches in bypassing the Reality Gap and sim-to-real implementations, as the time taken for real-world learning can quickly outperform data-intensive Deep alternatives.
Download
Paper Nr: 117
Title:

Data Sharing and Assimilation in Multi-Robot Systems for Environment Mapping
Authors:

Abdul Z. Yousaf and Gianni A. Di Caro
Abstract: We consider scenarios where a mobile multi-robot system is used for mapping a spatial field. Gaussian processes are a widely employed regression model for this type of tasks. For the sake of generality, scalability, and robustness, we assume that planning and control are fully distributed and that robots can only communicate via range-limited channels. In such scenarios, one core challenge is how to let the robots efficiently coordinate in order to maintain a shared view of the mapping process, and, accordingly, make plans minimizing overlaps and optimizing joint information gain from obtained measurements. A simple approach of sharing and utilizing all the sampled data would not scale to large teams, neither for computation nor for communication (assuming a general ad hoc robot network). Building on previous work where robots adaptively plan where to sample data by selecting convex containment regions, we propose a data sharing and assimilation strategy which aims to minimize the impact on communication and computation while minimizing the loss on accuracy in map estimation. The strategy exploits convexity of the regions to create compact meta-data that are locally shared. Submodularity of information processes and properties of GPs are used by the robots to create highly informative summaries of the sampled regions, that are shared on-demand based on the meta-data. In turn, a received summary is assimilated by a robot into its local GP only if/when needed. We perform a number of studies in simulation using real data from bathymetric maps to show the efficacy of the strategy for supporting scalability of computations and communications while guaranteeing learning accurate maps.
Download
Paper Nr: 118
Title:

Multi-robot Decentralized Exploration using Weighted Random Selection
Authors:

Abhijith N. Balan and Asokan Thondiyath
Abstract: The exploration tasks using multi-robot systems require efficient coordination and information sharing between robots. The map creation is usually done through allocating frontiers,i.e., the boundary between explored and unknown regions of the map, to each robot. This paper introduces an efficient frontier allocation method based on weighted random selection for a decentralized multi-robot system. The weights are calculated based on the size of the frontiers and the distance between a robot and the frontiers. In this strategy, each robot identifies the available frontiers in a shared map and select the goal for exploration through a random selection. Even though a robot is randomly picking a frontier without coordination with other robots, a collective intelligence is developed at the swarm level. The proposed method is computationally efficient and uses minimal communication between the robots in a decentralized multi-robot system. As a result, the robots get allocated to frontier according to the calculated weight. A comparison with the nearest frontier exploration approach in computerized simulation demonstrates the efficiency of the proposed algorithm.
Download
Paper Nr: 5
Title:

A Control Engineering Framework for Quadrotors: An Application for the Crazyflie 2.X
Authors:

Rafael Socas, Raquel Dormido, María Guinaldo and Sebastián Dormido
Abstract: In this work, a new framework to develop and investigate new control algorithms in quadrotors is presented. The proposed system includes two main elements: 1) a simulation laboratory and 2) a real environment based on the Crazyflie 2.X quadcopter. Both elements offer all the necessary tools to design and implement a wide variety of controllers in the real system. The phases to carry out the controller’s designs and how to apply them to the real quadrotor are explained in detail. A practical application to check the proposed framework and some additional experiments have been investigated in depth. The simulation and experimental results corroborate the validity of the proposed framework.
Download
Paper Nr: 17
Title:

Time Delay Investigation in Telerobotic Surgery
Authors:

Vivian Chai and Dan-Sorin Necsulescu
Abstract: Telerobotic surgery is a medical technology that allows a surgeon to operate on a patient from a distance, via a communication network. As with all telerobotic procedures, factors such as time delay or limited bandwidth may affect the performance of the robotics. In addition, surgeries require contact with soft tissues of the body, resulting in unaccounted external forces on the robotic manipulator. The paper investigates the effects of time delay on the desired trajectory of a telerobotic arm that undergoes forces due to contact with skin tissue. Two behaviours will be explored through simulation, under different time delays, with the robotic arm’s end effector sliding across the surface of the skin, as well as compressing the skin perpendicularly. Simulation results are compared with the expected behaviour of existing telerobotic performance. It was found that tangential forces across the skin’s surface do not significantly impact telerobotic performance but when subject to normal contact, the robotic arm fails under shorter time delays than expected. Further experimentation with trajectories, not limited to parallel or perpendicular motion of the effector on the skin, and trials involving real tissues and manipulators will provide greater insight for understanding telerobotic performance in surgery.
Download
Paper Nr: 19
Title:

Over Two Years of Challenging Environmental Conditions for Localization: The IPLT Dataset
Authors:

Youssef Bouaziz, Eric Royer, Guillaume Bresson and Michel Dhome
Abstract: This paper presents a new challenging dataset for autonomous driving applications: Institut Pascal Long-Term — IPLT — Dataset which was collected over two years and it contains, at the moment, 127 sequences and it still growing. This dataset has been captured in a parking lot where our experimental vehicle has followed the same path with slight lateral and angular deviations while we made sure to incorporate various environmental conditions caused by luminance, weather, seasonal changes.
Download
Paper Nr: 26
Title:

From 2D to 3D Mixed Reality Human-Robot Interface in Hazardous Robotic Interventions with the Use of Redundant Mobile Manipulator
Authors:

Krzysztof A. Szczurek, Raul M. Prades, Eloise Matheson, Hugo Perier, Luca R. Buonocore and Mario Di Castro
Abstract: 3D Mixed Reality (MR) Human-Robot Interfaces (HRI) show promise for robotic operators to complete tasks more quickly, safely and with less training. The objective of this study is to assess the use of 3D MR HRI environment in comparison with a standard 2D Graphical User Interface (GUI) in order to control a redundant mobile manipulator. The experimental data was taken during operation with a 9 DOF manipulator mounted in a robotized train, CERN Train Inspection Monorail (TIM), used for the Beam Loss Monitor robotic measurement task in a complex hazardous intervention scenario at CERN. The efficiency and workload of an operator were compared with the use of both types of interfaces with NASA TLX method. The usage of heart rate and Galvanic Skin Response parameters for operator condition and stress monitoring was tested. The results show that teleoperation with 3D MR HRI mitigates cognitive fatigue and stress by improving the operators understanding of both the robot’s pose and the surrounding environment or scene.
Download
Paper Nr: 71
Title:

Evaluation of the Capabilities of a Hybrid Driven Exoskeleton in Passive Mode of Interaction
Authors:

Dimitar Chakarov, Ivanka Veneva, Pavel Venev and Mihail Tsveov
Abstract: A new construction of upper limbs exoskeleton with hybrid drive was studied in this work. The paper presents mechanical structure and actuation of exoskeleton with hybrid drive including pneumatic cylinders and electric motors. In order to evaluate the transparency and safety, the capabilities of the exoskeleton in passive mode of interaction was estimated. A model of the interaction force between the patient and the exoskeleton arm in passive mode was built. The model was based on harmonious movements imposed in one joint of the exoskeleton arm. Experiments and simulations were performed to assess the interaction force between the patient and the exoskeleton because of the mechanical impedance of the device. The force of interaction was obtained from passive forces, such as inertial, frictional and gravitational forces, as well as from the elasticity of the pneumatics. The patient-initiated harmonic motion was studied in two cases-without pressure in the chambers and with pressure for gravity compensation. The results where demonstrated graphically. Conclusions where made about the behavior of the exoskeleton in the passive mode of interaction.
Download
Paper Nr: 72
Title:

CAD-based Grasp and Motion Planning for Process Automation in Fused Deposition Modelling
Authors:

Andreas Wiedholz, Michael Heider, Richard Nordsieck, Andreas Angerer, Simon Dietrich and Jörg Hähner
Abstract: Planning the right grasp pose and motion into it has been a problem in the robotic community for more than 20 years. This paper presents a model-based approach for a Pick action of a robot that increases the automation of FDM based additive manufacturing by removing a produced object from the build plate. We treat grasp pose planning, motion planning and simulation-based verification as separate components to allow a high exchangeability. When testing a variety of different object geometries, feasible grasps and motions were obtained for all objects. We also found that the computation time is highly dependent on the random seed, leading us to employ a system of budgeted runs for which we report the estimated success probability and expected running time. Within the budget, some objects never found feasible picks. Thus, we rotated these objects by 90º which lead to a substantial improvement in success probabilities.
Download
Paper Nr: 93
Title:

Design of a Rehabilitation Exoskeleton with Impedance Control: First Experiments
Authors:

Gaëtan Courtois, Jason Chevrie, Antoine Dequidt, Xavier Bonnet and Philippe Pudlo
Abstract: In this paper, we disclose the design strategy, control design and preliminary works leading to the development of a post stroke gait rehabilitation exoskeleton. The strategy is established based on the conventional gait rehabilitation currently used in rehabilitation centers and defines the exoskeleton as an interface between the therapist and the patient. The final purpose of this interface is to complete the conventional rehabilitation by intensifying the work of the patient while relieving the physical burden on the therapist. As the conventional rehabilitation is based on successive exercises the control is designed to have several operating modes triggered depending on the currently processing exercise. A test bench was realised to evaluate quantitatively as well as qualitatively these operating modes. Preliminary results of quantitative experiments on the transparent operation mode are then presented. These results validate the control design and comfort us on our development method.
Download
Paper Nr: 116
Title:

Visual Navigation Datasets for Event-based Vision: 2014-2021
Authors:

Andrejs Zujevs and Agris Nikitenko
Abstract: Visual navigation is becoming the primary approach to the way unmanned vehicles such as mobile robots and drones navigate in their operational environment. A novel type of visual sensor named dynamic visual sensor or event-based camera has significant advantages over conventional digital colour or grey-scale cameras. It is an asynchronous sensor with high temporal resolution and high dynamic range. Thus, it is particularly promising for the visual navigation of mobile robots and drones. Due to the novelty of this sensor, publicly available datasets are scarce. In this paper, a total of nine datasets aimed at event-based visual navigation are reviewed and their most important properties and features are pointed out. Major aspects for choosing an appropriate dataset for visual navigation tasks are also discussed.
Download
Paper Nr: 121
Title:

Dynamics of a Four Wheeled Wall Climbing Robot
Authors:

Anokhee Chokshi and Jaina Mehta
Abstract: In this paper, a mathematical model of a four wheeled independently driven Wall Climbing Robot (WCR) is developed. The consideration of only the kinematic model for a WCR may reduce its performance during sudden changes in acceleration and turning. To address this issue, a dynamic model that includes the wall/wheel interactions i.e., lateral and longitudinal frictional forces, is proposed. The effect of wheel slip is considered for a more realistic dynamic model. The models that are typically developed for the vertical contact forces, an important parameter affecting the frictional forces, assume equal weight distribution on the wheels. However, to accommodate the load shift due to the variation in acceleration along with the distribution of adhesion force, lateral and longitudinal acceleration components are also taken into account. The major components of this WCR model consist of the wheel dynamics, the wall/wheel interactions, the kinematics and the dynamics. Simulations are performed to demonstrate and verify the model. The suggested model in the future can be applied in the development of control algorithms for wheeled WCRs.
Download
Paper Nr: 123
Title:

Strawberry Disease Detection in Precision Agriculture
Authors:

Aguirre Santiago, Leonardo Solaque and Alexandra Velasco
Abstract: Crop disease detection in precision agriculture has an important impact on farming, improving production, and reducing economic losses. This is why some efforts have been done in this direction. This paper compares 4 object detection algorithms based on deep learning to detect diseases in strawberry crops. Here, we present a step towards detecting the most common diseases to prevent economical losses. The main purpose is to detect mainly three diseases of the strawberry crops, i.e. Botrytis cinerea, Leaf scorch, and Powdery mildew, to take further actions if the crops are unhealthy. We have chosen these three diseases because these are frequent and unpredictable issues, and the risk of infection is high. For this, we trained four algorithms, two based on Single Shot MultiBox Detector and two based on EfficientDet algorithm. We focus the analysis on the two best results based on the mean average precision. We have used Google colab for training, then a Core i5 host computer and an Nvidia Jetson nano were used for testing. We have achieved a detection network with a mean average precision of 81% in the best case, in detecting the three proposed classes. While using an NVIDIA Jetson nano, the accuracy increases up to 86% due to the dedicated GPU that processes Convolutional Neural Networks(CNN).
Download

Area 4 - Signal Processing, Sensors, Systems Modelling and Control

Full Papers
Paper Nr: 24
Title:

Uniformly Regular Triangulations for Parameterizing Lyapunov Functions
Authors:

Peter Giesl and Sigurdur Hafstein
Abstract: The computation of Lyapunov functions to determine the basins of attraction of equilibria in dynamical systems can be achieved using linear programming. In particular, we consider a CPA (continuous piecewise affine) Lyapunov function, which can be fully described by its values at the vertices of a given triangulation. The method is guaranteed to find a CPA Lyapunov function, if a sequence of finer and finer triangulations with a bound on their degeneracy is considered. Hence, the notion of (h,d)-bounded triangulations was introduced, where h is a bound on the diameter of each simplex and d a bound on the degeneracy, expressed by the so-called shape-matrices of the simplices. However, the shape-matrix, and thus the degeneracy, depends on the ordering of the vertices in each simplex. In this paper, we first remove the rather unnatural dependency of the degeneracy on the ordering of the vertices and show that an (h,d)-bounded triangulation, of which the ordering of the vertices is changed, is still (h,d∗)-bounded, where d∗ is a function of d, h, and the dimension of the system. Furthermore, we express the degeneracy in terms of the condition number, which is a well-studied quantity.
Download
Paper Nr: 32
Title:

An Effective Method for Data Processing of Inertial Measurement Units Applied to Embedded Systems
Authors:

Christoph Kolhoff and Markus Kemper
Abstract: Autonomous functions for navigation and localization have piqued the attention and interest in many fields of science and engineering as automotive, aviation and robotics. Desiring high quantity of autonomous products, the used components are requested to be cheap. This often lead engineers or developers to apply micro-electromechanical systems that exhibit large errors. To use these sensors anyway, the acquired data must be processed online for error reduction. Hence there is a need for an algorithm that is easy to compute. The aim of this research is to develop a generic algorithm based on a Gauss-Markov process representing the drifting bias that can be parametrized easily and performs well on real-time systems. Therefore, the error model imitates the sensor’s output and removing the errors afterwards. Finally, a validation of the suggested algorithm is performed by comparing processed data of the micro-controller to data processed a posteriori on a high-performance computer.
Download
Paper Nr: 60
Title:

Improved Output Feedback Control of Constrained Linear Systems using Invariant Sets
Authors:

Ana O. Mancini, Tiago A. Almeida and Carlos T. Dórea
Abstract: We propose an improved design method for output feedback control of discrete-time linear systems subject to state and control constraints, additive disturbances and measurement noise. Output Feedback Controlled-Invariant polyhedral sets are used to ensure that state and input constraints are satisfied all time. The control strategy seeks to enforce the set of states consistent with the measured output into a closed ball around the origin. The control input is computed through the solution of Linear Programming (LP) problems, whose goal is to minimize the size of the ball one step ahead. Then, we use the optimization results to reduce the set of admissible states, steering the state to a smaller ball around the origin. The improvement provided by the proposed strategy is illustrated by numerical examples.
Download
Paper Nr: 62
Title:

Output-feedback MPC for Robotic Systems under Bounded Noise
Authors:

Lenka Kuklišová Pavelková and Květoslav Belda
Abstract: The paper presents an output-feedback model predictive control applied to the motion control of a dynamic model of a parallel kinematic machine. The controlled system is described by a stochastic linear discrete-time model with bounded disturbances. An approximate uniform Bayesian filter provides set state estimates. The choice of the specific point estimate from this set is a part of the optimization. The cost function includes penalties on the tracking error and the actuation effort respecting increments. Illustrative examples show the effectiveness of the proposed approach and provide a comparison with previous results.
Download
Paper Nr: 70
Title:

Upper Limb Anthropometric Parameter Estimation through Convolutional Neural Network Systems and Image Processing
Authors:

Andres Guatibonza, Leonardo Solaque, Alexandra Velasco and Lina Peñuela
Abstract: Anthropometry is a versatile tool for evaluating the human body proportions. This tool allows the orientation of public health policies and clinical decisions. But in order to optimize the obtaining of anthropometric measurements, different methods have been developed to determine anthropometry automatically using artificial intelligence. In this work, we apply a convolutional neural network to estimate the upper limb’s anthropometric parameters. With this aim, we use the OpenPose estimator system and image processing for segmentation with U-NET from a complete uncalibrated body image. The parameter estimation system is performed with total body images from 4 different volunteers. The system accuracy is evaluated through a global average percentage of 71% from the comparison between measured values and estimated values. A fine-tuning of algorithm hyper-parameters will be used in future works to improve the estimation.
Download
Paper Nr: 80
Title:

Robustness of Contraction Metrics Computed by Radial Basis Functions
Authors:

Peter Giesl, Sigurdur Hafstein and Iman Mehrabinezhad
Abstract: We study contraction metrics computed for dynamical systems with periodic orbits using generalized interpolation with radial basis functions. The robustness of the metric with respect to perturbations of the system is proved and demonstrated for two examples from the literature.
Download
Paper Nr: 84
Title:

Prediction of Multimodal Poisson Variable using Discretization of Gaussian Data
Authors:

Evženie Uglickich, Ivan Nagy and Matej Petrouš
Abstract: The paper deals with predicting a discrete target variable described by the Poisson distribution based on the discretized Gaussian explanatory data under condition of the multimodality of a system observed. The discretization is performed using the recursive mixture-based clustering algorithms under Bayesian methodology. The proposed approach allows to estimate the Gaussian and Poisson models existing for each discretization interval of explanatory data and use them for the prediction. The main contributions of the approach include: (i) modeling the Poisson variable based on the cluster analysis of explanatory continuous data, (ii) the discretization approach based on recursive mixture estimation theory, (iii) the online prediction of the Poisson variable based on available Gaussian data discretized in real time. Results of illustrative experiments and comparison with the Poisson regression is demonstrated.
Download
Paper Nr: 89
Title:

Vaccination and Time Limited Immunization for SARS-CoV-2 Infection
Authors:

Paolo Di Giamberardino and Daniela Iacoviello
Abstract: The paper aims at a discussion of the effects of the containment measures against COVID-19 through the analysis of the reproduction number. Starting from a mathematical model in which several controls are considered, including the vaccination, and introducing also an hypotised limited duration of the immunity acquired both from vaccine and from healing from the illness, the steady state behaviour, both in the uncontrolled and in the controled cases is studied. The expressions for the basic reproduction number and the actual reproduction number under control actions are computed by means of the next generation matrix approach. This function is numerically investigated, showing some graphs which illustrate, qualitatively and quantitatively, in an intuitive way the positive effects of the controls and the negative contribution of the absence of a lifetime immunization from virus.
Download
Paper Nr: 90
Title:

BioDeep: A Deep Learning System for IMU-based Human Biometrics Recognition
Authors:

Abeer Mostafa, Samir A. Elsagheer and Walid Gomaa
Abstract: Human biometrics recognition has been of wide interest recently due to its benefits in various applications such as health care and recommender systems. The rise of deep learning development, together with the massive data acquisition systems, made it feasible to reuse models trained on one task for solving another similar task. In this work, we present a novel approach for age and gender recognition based on gait data acquired from Inertial Measurement Unit (IMU). BioDeep design is composed of two phases, first of which is applying a statistical method for feature modelling, the autocorrelation function, then building a Convolutional Neural Network (CNN) for age regression and gender classification. We also use random forest as a baseline model to compare the results achieved by both methods. We validate our models using four publicly available datasets. The second phase is doing transfer learning over these diverse datasets. We train a CNN on one dataset and reuse its feature maps over the other datasets for solving both age and gender recognition problems. Our experimental evaluation over the four datasets separately shows very promising results. Furthermore, transfer learning achieved 20 − 30x speedup in the training time in addition to keeping the acceptable prediction accuracy.
Download
Paper Nr: 108
Title:

A Study on Model-based Optimization of Vaccination Strategies against Epidemic Virus Spread
Authors:

Zonglin Liu, Muhammed Omayrat and Olaf Stursberg
Abstract: This paper aims at applying optimal control to investigate different vaccination strategies against the epidemic spread of viral diseases. Background of the study is the situation in the first half of 2021, when many countries started their vaccination procedures against the COVID-19 disease, but suffered from shortages of vaccines, such that the efficient distribution of the available amount of vaccine turned out to be crucial to mitigate the pandemic. The paper first suggests an extended version of a known model of virus spread in order include the vaccination process. Based on this model, the formulation and solution of optimization problems is used to determine how available vaccine should be distributed over different age-groups of the population to minimize virus spread. Effectiveness of the obtained strategies compared to standard ones is demonstrated in simulations.
Download
Short Papers
Paper Nr: 10
Title:

Bayesian Mixture Estimation without Tears
Authors:

Šárka Jozová, Evženie Uglickich and Ivan Nagy
Abstract: This paper aims at presenting the on-line non-iterative form of Bayesian mixture estimation. The model used is composed of a set of sub-models (components) and an estimated pointer variable that currently indicates the active component. The estimation is built on an approximated Bayes rule using weighted measured data. The weights are derived from the so called proximity of measured data entries to individual components. The basis for the generation of the weights are integrated likelihood functions with the inserted point estimates of the component parameters. One of the main advantages of the presented data analysis method is a possibility of a simple incorporation of the available prior knowledge. Simple examples with a programming code as well as results of experiments with real data are demonstrated. The main goal of this paper is to provide clear description of the Bayesian estimation method based on the approximated likelihood functions, called proximities.
Download
Paper Nr: 14
Title:

Fractional Order PV/T Model Design and Estimation using the Fractional Observer
Authors:

Amer Aziz, Muwahida Liaquat, Aamer I. Bhatti and Lahoucine Ouhsaine
Abstract: In the last decade, the demand for renewable energy sources has been increased due to factors which include the rising fuel price and pollution, and consequently research on solar energy sources has been increased to improve their efficiency. Photovoltaic Thermal (PV/T) system provides electrical power and heat simultaneously, which is the promising technology. This research paper illustrates the simulation of the comprehensive thermal-based mathematical model of the PV/T system with the joint estimation of the system states: the temperature at each node, and disturbance using Fractional-High Order Sliding Mode Observer (HOSMO). A fractional-order differential equation describes the PV/T system because of its characterization in heterogeneous media and its multilayers structure. Fractional–HOSMO is a robust observer that can be used further for the reduced-sensor control of PVT, which can be comparatively cheap. The parameter values are derived from the thermal configuration of the layers and the properties of constituents.
Download
Paper Nr: 35
Title:

Ground Speed Measuring System for Autonomous Vehicles
Authors:

Yasmine S. Antille, Etienne Gubler and Juan-Mario Gruber
Abstract: In this paper a Ground Speed Measuring System which can measure the ground speed over the ground in three dimensions is proposed. The system uses two Kalman filters to compute the final ground speed based on the readings from its various sensors. The proposed solution combines state of the art techniques from different fields of sensor technology and will be incorporated into the high-performance driverless vehicle after completion of this project. The findings and learnings of developing this system are discussed and an evaluation of the module is presented. In the end, the system can accurately estimate a test vehicle’s ground speed during system field tests.
Download
Paper Nr: 57
Title:

Modeling, Analysis and Control of COVID-19 in Italy: Study of Scenarios
Authors:

Paolo Di Giamberardino, Rita Caldarella and Daniela Iacoviello
Abstract: Since the beginning of 2020 in few weeks all the world has been interested by the pandemic due to SARS-CoV 2, causing more than 3 millions of dead people and more than 146 millions of infected patients. The virus moves with people and the most effective containment measure appears to be the severe lockdown; on the other hand, for obvious social and economic reasons, it can not be applied for long periods. Moreover, the increasing knwoledge on the virus and on its trasmission modes suggested various strategies, such as the use of masks, social distancing, disinfection and the fast identification of infected patients, up to the recent vaccination campaign. In this paper, the COVID-19 spread is studied referring to the Italian situation; the control actions introduced during 2020-2021 are identified in terms of their actual effects, allowing to study possible intervention scenarios.
Download
Paper Nr: 92
Title:

Hybrid Impedance and Nonlinear Adaptive Control for a 7-DoF Upper Limb Rehabilitation Robot: Formulation and Stability Analysis
Authors:

Andres Guatibonza, Leonardo Solaque, Alexandra Velasco and Lina Peñuela
Abstract: Physical rehabilitation aims to improve the condition of people with any musculoskeletal disorder. Different assistive technologies have been developed to provide support to this process. In this context, human-machine interaction has progressively improved to avoid abrupt movements and vibrations, to obtain a more natural interaction, where control strategies play a key role. In this work, a control technique based on the combination of nonlinear adaptive theory with a hybrid impedance control applied to a 7-DoF upper limb assistance robotic device is proposed. Additionally, we include the stability analysis using Lyapunov functions. Then, we validate the strategies through simulations for one rehabilitation routine test. The articular and cartesian obtained results demonstrate the effectiveness of the control to follow trajectories. The control stabilizes the trajectories in 0.9 seconds even when the initial conditions start far from the desired trajectories, without producing vibrations or overshoots, which is the desired behavior in rehabilitation applications like the one we propose.
Download
Paper Nr: 97
Title:

Location Determination of On-body Inertial Sensors
Authors:

Hisham Madcor, Osama Adel and Walid Gomaa
Abstract: Human Activity Recognition has gained tremendous drive in recent years. This is due to the increasing ubiquity of all types of sensors in commodity devices such as smartphones, smart watches, tablets, etc. This has made available to the normal user a continuous stream of data including visual data, inertial motion data, audio, etc. In this paper we focus on data streamed from inertial motion units (IMUs). Such units are currently embedded on almost all wearable devices including smart watches, wrist bands, etc. In many research works, as well as in many real applications, different specialized IMU units are mounted on different body parts. In the current work, we try to answer the following question: given the streamed inertial signals of a gait pattern, as well as some other activities, determine which sensor location on the subject’s body generated this signal. We validate our work on several datasets that contain multi-dimensional measurements from a multitude of sensors mounted on different body parts. The main sensors used are the accelerometer and gyroscope. We use the Random Forest Classifier over the raw data without any prior feature extraction. This has proven yet very effective as evidenced by the results using different metrics including accuracy, precision, recall, F1-score, etc. An important application of such research can be in data augmentation of timeseries inertial data. This can be used as well for healthcare applications, for example, in treatment assessment for people with motion disabilities.
Download
Paper Nr: 120
Title:

Analysis and Application of Multispectral Image Processing Techniques Applied to Soybean Crops from Drones Vision System
Authors:

Evelio González, Cristhian Núñez, José Salinas, Jorge Rodas, Mariela Rodas, Enrique Paiva, Yassine Kali, Maarouf Saad, Fernando Lesme, Jose Lesme, Luis Gonzalez, Belen Maldonado and José Rodríguez-Piñeiro
Abstract: Drones are important in precision agriculture applications since they represent a new tool that can increase crop production. In this context, the digital processing of the images obtained from multispectral cameras integrated into the drones makes it possible to analyze the stress state of the crops, their vigor, a burned area, among others. The latter are usually obtained through proprietary applications with very high subscription costs. For this reason, this article presents the step-by-step implementation process of the different methods or algorithms to be applied to multispectral images using the open-source Python programming language. We use a soybean crop as an example of the application, and the results obtained from applying the digital image processing algorithms are presented.
Download
Paper Nr: 30
Title:

Estimating the Frequency of the Sinusoidal Signal using the Parameterization based on the Delay Operators
Authors:

Tung N. Khac, Sergey Vlasov and Radda Iureva
Abstract: The article presents an algorithm for estimating the frequency of an offset sinusoidal signal. Delay operators are applied to the measured signal, and a linear regression model is constructed containing the measured signals and the constant vector depending on unknown frequency. For the vector regression model, the method cascade reduction is used. A reduction procedure is proposed that allows the original model to be reduced to a reduced one containing a smaller number of unknown parameters. Finally, using the classical gradient method was used to compare the efficiency of the proposed method.
Download
Paper Nr: 43
Title:

A Novel Method for the Real-time Force Losses Detection in Servo Welding Guns
Authors:

D. Ibáñez, E. García, J. Martos and J. Soret
Abstract: Nowadays, real-time detection methods are increasingly necessary for predictive maintenance in production processes. Specifically, in the metal joining production processes that use the resistance welding process, the optimization of maintenance programs is sought to improve both the quality of the body and its manufacturing cost. In this novel paper a new method is presented for the sensorless detection of pressure losses in welding lines. The proposed system bases its operation on the measurement of the existing variables in the resistance welding process carried out using servo guns. This paper also shows the proposed system for the data acquisition, data sending and visualization in real-time of the health of the welding gun. This results in a system with a low installation cost but with great performance in reducing problems associated with force losses in welding guns.
Download
Paper Nr: 110
Title:

Analysis of a Nonlinear Control Law with Cubic Nonlinearity
Authors:

Melnikov Vitaly, Melnikov Gennady and Dudarenko Natalia
Abstract: The paper is considered the problem of improving the stabilization of nonlinear controlled systems. It is proposed to solve the problem by introducing cubic components into the control law. Using the method of polynomial transformation, a comparative analysis of the influence of cubic components on the dynamics of a controlled systems is presented. As a result, some conclusions about the choice of the structure and parameters of the nonlinear control law are presented and recommendations are given.
Download