ICINCO 2022 Abstracts

Area 1 - Industrial Informatics
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

Short Papers
Paper Nr: 20
Title:

Emerging Technologies in the Era of Digital Transformation: State of the Art in the Railway Sector
Authors:

Dietmar Möller, Lukas Iffländer, Michael Nord, Patrik Krause, Bernd Leppla, Kristin Mühl, Nikolai Lensik and Peter Czerkewski
Abstract: Emerging technologies and digital transformation are essential indicators in today’s industrial sectors. The railway and public transportation sectors are undergoing a substantial transformation through digitalization and emerging technologies. However, little is known about the manifold of applications in the industrial sectors and progress so far. In this study, we consider various emerging technologies and proposed use-cases. Next, using a two-step survey and a SWOTA analysis, we analyze both sector’s maturity levels regarding these technologies. The analysis indicates technologies currently permeating the analyzed sectors, shows discrepancies between technology application and knowledge, and multiple issues hamper their implementation.
Download
Paper Nr: 23
Title:

Towards Data-driven Production: Analysis of Data Models Describing Machinery Jobs in OPC UA
Authors:

Tonja Heinemann, Marwin Gihr, Oliver Riedel and Armin Lechler
Abstract: This work analyzes the Open Platform Communications Unified Architecture (OPC UA) specifications for flat glass, plastics and rubber, machine vision, ISA-95 and machine tools regarding their job descriptions. Common contents of job models in the domain of machinery are deducted. Using a structured qualitative content analysis, more than 70 functional elements used in OPC UA job models have been identified. While some of these functional elements are modeled similarly in multiple domains, major differences are identified for other functional elements. Especially those differences constitute impediments in the standardization of industrial communication. The results of this work harmonize the contents and the modeling techniques regarding machining jobs in OPC UA and provide a generally applicable method for the standardization of machine communication throughout different domains. With this method for standardization, this work contributes directly to the goal of OPC UA, to easily exchange data between platforms from multiple vendors.
Download
Paper Nr: 106
Title:

Feedrate Planning for a Delta Parallel Kinematics Numerically Controlled Machine using NURBS Toolpaths
Authors:

Gabriel Karasek and Krystian Erwinski
Abstract: This paper presents a concept of a computationally efficient feedrate planning algorithm for parallel kinematics machine in a linear delta configuration. Non-Uniform Rational B-Spline (NURBS) polynomial curve is used for toolpath definition which provides a smooth trajectory. The feedrate profile is defined as a jerk limited S-Curve which takes into account limitations stemming from the toolpath curvature and the machine kinematics. The article presents the outline of the method, structure of the experimental station including the real-time control system and preliminary experimental results. Further direction of the research is also described. The proposed method can provide a smooth motion trajectory obtained in real-time with small computational requirements.
Download
Paper Nr: 112
Title:

The Visual Inspection of Solder Balls in Semiconductor Encapsulation
Authors:

Conceição N. Silva, Neandra P. Ferreira, Sharlene S. Meireles, Mario Otani, Vandermi J. da Silva, Carlos A. O. de Freitas and Felipe G. Oliveira
Abstract: The growing demand for increasing memory storage capacity has required a high density of integration within the semiconductor encapsulation and, consequently, has made this process more complex and susceptible to failures during the production stage. In the semiconductor encapsulation area, the costs of materials and equipment are high and the profit margin is narrow, making it necessary to rigorously inspect the process steps to keep the productive activity viable. This work addresses the problem of quality control in silicon wafers soldering procedure, allowing error detection before the epoxy resin molding process, generating useful infor-mation for correcting equipment configurations and predicting failures from the raw materials and inputs used in the process. We propose an approach to classify solder balls, in the soldering process of silicon wafers on Ball Grid Array (BGA), contained in the Printed Circuit Board (PCB) substrates. The proposed methodology is composed of two main steps: i) Solder ball segmentation; and ii) Solder ball classification through deep learning. The proposed predictive model learns the relation between visual features and the different soldering conditions. Real and simulated experiments were carried out to validate the proposed approach. Results show the obtained accuracy of 99.4%, using Convolutional Neural Network (CNN) classification model. Furthermore, the proposed approach presents high accuracy even regarding noisy images, resulting in accuracy of 92.8% and 75.7% for a Salt and Pepper and Gaussian noise, respectively, in the worst scenario. Experiments demonstrate reliability and robustness, optimizing the manufacturing.
Download
Paper Nr: 82
Title:

Intelligent Thermal Accumulator Operation Control System based on Renewable Energy Sources
Authors:

N. M. Tasmurzayev, B. S. Amangeldy, Y. S. Nurakhov, D. Z. Akhmed-Zaki and Zh. E. Baigarayeva
Abstract: In this paper, we consider the software and hardware implementation of an intelligent control system for optimal use of solar thermal energy and geothermal energy accumulator for heating and hot water supply of residential areas, multi-storey buildings, greenhouses with the highest possible efficiency. To achieve maximum results, the system complies with the Industry 4.0 concept and uses a multi-level management and monitoring structure such as Web dispatching, local and global system management and monitoring, cloud and local data storage, cloud and local management and monitoring, emergency notification and changes in the system via web technologies.
Download

Area 2 - Intelligent Control Systems and Optimization

Full Papers
Paper Nr: 35
Title:

Shore based Control Center Architecture for Teleoperation of Highly Automated Inland Waterway Vessels in Urban Environments
Authors:

Arne Lamm, Janusz A. Piotrowski and Axel Hahn
Abstract: The following paper presents an SCC architecture that allows to take over the remote control of one or more ships from the shore side, especially in critical situations, in order to present a concrete solution of a remote control center as proposed in the MASS levels for autonomous navigation. Particular attention was paid to the technical and functional components and requirements specified by the regulations, and the practicability based on decision-making and action execution was investigated. In particular, the three levels of situational awareness were taken into account and the remote control center was finally implemented as a prototype. For the evaluation, the practicability based on the RTT was assessed and the completeness based on the design specifications of common INS was examined.
Download
Paper Nr: 38
Title:

Neuro-dynamic Control of an above Knee Prosthetic Leg
Authors:

Zunaed Kibria and Sesh Commuri
Abstract: The control of a prosthetic leg for above-knee amputees is fraught with several challenges. While the dynamics of the knee-ankle system are complex and unknown, the control problem is exacerbated by the lack of desired joint trajectories as they are dictated by the locomotion needs of the individual. Improper movement of the knee and ankle joints can have serious implications for the safety of the user. Further, dissimilarities in the gait of the amputated side and the intact side can result in gait abnormalities that result in increased metabolic energy consumption and musculo-skeletal pains in the short term, and cardiovascular and other health complications in the long term. In this paper, we propose a novel neuro-dynamic control strategy that can guarantee stable control of the prosthetic limb while minimizing the gait asymmetry between the intact and prosthetic limb. Further, the algorithm learns the unknown elements of the dynamics and adapts to the changing locomotion needs of the individual. The efficacy of the proposed approach is demonstrated through numerical simulations.
Download
Paper Nr: 55
Title:

Optimal Social Limitation Reduction under Vaccination and Booster Doses
Authors:

Paolo Di Giamberardino and Daniela Iacoviello
Abstract: In the paper an optimal control solution is provided for the containment of the number of infected individuals in COVID-19 pandemic under vaccination campaign. The possibility to dynamically change the cost of the controls according to the ongoing evolution within the design procedure allows to get great efforts in presence of very serious disease conditions, saving resources otherwise. The different contribution of vaccinated and unvaccinated individuals to the epidemic spread is investigated, optimising the controls which describe the individual contact restrictions separately for the two classes and showing that it would have been possible to reduce all the social limitations introduced by many governments for the vaccinated individuals since the beginning of the vaccination campaign.
Download
Paper Nr: 60
Title:

A Recommendation Mechanism of Selecting Machine Learning Models for Fault Diagnosis
Authors:

Wen-Lin Sun, Yu-Lun Huang and Kai-Wei Yeh
Abstract: Faults of a machine tool generally lead to a suspension of a production line when the defeated parts need a long lead time. The prevention of such suspension depends on the health condition of machine tools in a factory. Hence, monitoring the health conditions of machine tools with modern Machine Learning (ML) technologies is one of the highlights of industry evolution 4.0. Though researchers presented several methods and mechanisms to solve the fault detection and prediction of machine tools, the current works usually focus on deploying one ML algorithm to one specific machine tool and generating a well-trained model for fault diagnosis and detection for that machine tool, which are impractical since a factory typically runs a variety of machine tools. This paper presents an Automatic Fault Diagnosis Mechanism (AFDM), taking historical data provided by an administrator and then recommending a machine-learning algorithm for fault diagnosis. AFDM can handle different types of data, diagnose faults for different machine tools, and provide a friendly interface for a factory administrator to select a proper analytical model for the specified type of machine tools. We design a series of experiments to prove the diversity, feasibility, and stability of AFDM.
Download
Paper Nr: 77
Title:

Resilient Control of Interconnected Microgrids Under Attack by Robust Nonlinear MPC
Authors:

Sarah Braun, Sebastian Albrecht and Sergio Lucia
Abstract: With the growing share of renewable energy sources, the uncertainty in power supply is increasing, on the one hand because of fluctuations in the renewables, but on the other hand also due to the threat of deliberate malicious attacks, which may become more prevalent due to the growing number of distributed generation units. It is thus essential that local microgrids are controlled in a robust manner in order to ensure stability and supply security even in the event of disturbances. To this end, we introduce a mathematical model for interconnected, physically coupled microgrids with renewable generation that are exposed to the risk of attacks. For optimal energy management and control, we present a resilient framework that combines a model-based method to identify occurring attacks and a model predictive control scheme to compute robust control inputs. We demonstrate the efficiency of the method for microgrid control in numerical experiments.
Download
Paper Nr: 86
Title:

Design and Implementation of Non-prehensile Manipulation Strategies
Authors:

Pooja Bhat, Matthias Nieuwenhuisen and Dirk Schulz
Abstract: Grasping of objects is not always feasible for robot manipulators, e.g., due to their geometric properties. Non-prehensile manipulation strategies can enable manipulators to successfully move these objects around. We discuss strategies for non-prehensile manipulation and focus on the investigation of such manipulation strategies based on open- and closed-loop control based on force torque measurements. The design of grippers for moving objects is also an important factor that is evaluated. The strategies are implemented and evaluated in simulation and on a KUKA LWR4+ manipulator arm.
Download
Paper Nr: 102
Title:

Analysis of the Squat Exercise from Visual Data
Authors:

Fatma Youssef, Ahmed B. Zaky and Walid Gomaa
Abstract: Squats are one of the most frequent at-home fitness activities. If the squat is performed improperly for a long time, it might result in serious injuries. This study presents a multiclass, multi-label dataset for squat workout evaluation. The dataset collects the most typical faults that novices make when practicing squats without supervision. As a first step toward universal virtual coaching for indoor exercises, the main objective is to contribute to the creation of a virtual coach for the squat exercise. A 3d position estimation is used to extract critical points from a squatting subject, then placed them in a distance matrix as the input to a multilayer convolution neural network with residual blocks. The proposed approach uses the exact match ratio performance metric and is able to achieve 94% accuracy. The performance of transfer learning as a known machine learning technique is evaluated for the squat activity classification task. Transfer learning is essential when changing the setup and configuration of the data collection process to reduce the computational efforts and resources.
Download
Paper Nr: 111
Title:

Robust Gain-scheduling LPV Control for a Reconfigurable Robot
Authors:

R. Al Saidi and S. Alirezaee
Abstract: This paper develops a robust gain–scheduling linear parameter varying (LPV) control for a reconfigurable robot that combines as many properties of different open kinematic structures as possible and can be used for a variety of applications. The kinematic design parameters, i.e., the Denavit–Hartenberg (D–H) parameters, can be modified to satisfy any configuration required to meet a specific task. By varying the joint twist angle parameter (a configuration parameter), the presented model is reconfigurable to any desired open kinematic structure, such as ABB, FANUC and SCARA robotic systems. A robust LPV control is developed for on–line measured parameters of a perturbed LPV model of a Bosch Scara robot arm. This control achieves superior tracking performance in the presence of dynamic and parameter uncertainties.
Download
Short Papers
Paper Nr: 8
Title:

Human Action Recognition using Convolutional Neural Network: Case of Service Robot Interaction
Authors:

Souhila Kahlouche and Mahmoud Belhocine
Abstract: This paper proposes a Human Robot Interaction (HRI) framework for a service robot capable of understanding common interactive human activities. The human activity recognition (HAR) algorithm is based on end to end deep Convolutional Neutral Network architecture. It uses as an input a view invariant 3D data of the skeleton joints, which is recorded from a single Microsoft Kinect camera to create a specific dataset of six interactive activities. In addition, an analysis of the most informative joint is made in order to optimize the recognition process. The system framework is built on Robot Operating System (ROS), and the real-life activity interaction between our service robot and the user is conducted for demonstrating the effectiveness of the developed HRI system. The trained model is evaluated on an experimental dataset created for this work and also the publicly available datasets Cornell Activity Dataset (CAD-60), and KARD HAR datasets. The performance of the proposed algorithm is proved when compared to other approaches and the results confirm its efficiency.
Download
Paper Nr: 16
Title:

Robust Neural Network for Sim-to-Real Gap in End-to-End Autonomous Driving
Authors:

Stephan Pareigis and Fynn L. Maaß
Abstract: A neural network architecture for end-to-end autonomous driving is presented, which is robust against discrepancies in system dynamics during the training process and in application. The proposed network architecture presents a first step to alleviate the simulation to reality gap with respect to differences in system dynamics. A vehicle is trained to drive inside a given lane in the CARLA simulator. The data is used to train NVIDIA’s PilotNet. When an offset is given to the steering angle of the vehicle while the trained network is being applied, PilotNet will not keep the vehicle inside the lane as expected. A new architecture is proposed called PilotNet∆, which is robust against steering angle offsets. Experiments in the simulator show that the vehicle will stay in the lane, although the steering properties of the vehicle differ
Download
Paper Nr: 29
Title:

Efficient Verification of CPA Lyapunov Functions
Authors:

Sigurdur F. Hafstein
Abstract: Lyapunov functions can be used to characterize the stability and basins of attraction for dynamical systems, whose dynamics are defined by ordinary differential equations. Since the analytic generation of Lyapunov functions for nonlinear systems is a formidable task, one often resorts to numerical methods. In this paper we study the efficient verification of the conditions for a Lyapunov function using affine interpolation over a triangulation; the values of the Lyapunov function candidate at the vertices of the triangulation can be generated using various different formulas from converse theorems in the Lyapunov stability theory. Further, we give an implementation in C++ and demonstrate its efficiency and applicability.
Download
Paper Nr: 30
Title:

Solving Stable Generalized Lyapunov Equations for Hankel Singular Values Computation
Authors:

Vasile Sima
Abstract: Generalized Lyapunov equations are often encountered in systems theory, analysis and design of control systems, and in many applications, including balanced realization algorithms, procedures for reduced order models, or Newton methods for generalized algebraic Riccati equations. An important application is the computation of the Hankel singular values of a generalized dynamical system, whose behavior is defined by a regular matrix pencil. This application uses the controllability and observability Gramians of the system, given as the solutions of a pair of generalized Lyapunov equations. The left hand side of each of these equations follows from the other one by applying the (conjugate) transposition operator. If the system is stable, the solutions of both equations are non-negative definite, hence they can be obtained in a factorized form. But these theoretical results may not hold in numerical computations if the symmetry and non-negative definiteness are not preserved by a solver. The paper summarizes new related numerical algorithms for complex continuous- and discrete-time generalized systems. Such solvers are not yet available in the SLICOT Library or MATLAB. The developed solvers address the essential practical issues of reliability, accuracy, and efficiency.
Download
Paper Nr: 36
Title:

Open-loop Control of a Soft Arm in Throwing Tasks
Authors:

Diego Bianchi, Michele G. Antonelli, Cecilia Laschi and Egidio Falotico
Abstract: This paper presents the implementation of an open-loop controller that allows a soft arm to throw objects in target positions. This valuable ability enables the robotic arm to expand its working space by tossing the objects outside it. Soft robots are characterized by high compliance and flexibility, which is paid in terms of dynamics that is highly non-linear and therefore hard to be modelled. An artificial neural network is employed to approximate the relationship between the actuation set and the target landing position, i.e., the direct model of the task. An optimization problem is defined to find the actuation set necessary to throw in a desired target. The proposed methodology has been tested on a soft robotic simulator (Elastica). Results show that the open-loop controller allows throwing objects in a target position with an average error of 0.90 mm and a maximum error of 10.47 mm, which compared to the characteristic dimension of the work-space correspond respectively to 0.07 % and 0.83 %.
Download
Paper Nr: 43
Title:

Navigation of Concentric Tube Continuum Robots using Optimal Control
Authors:

Siva C. Dhanakoti, John H. Maddocks and Martin Weiser
Abstract: Recently developed Concentric Tube Continuum Robots (CTCRs) are widely exploited in, for example in minimally invasive surgeries which involve navigating inside narrow body cavities close to sensitive regions. These CTCRs can be controlled by extending and rotating the tubes one inside the other in order to reach a target point or perform some task. The robot must deviate as little as possible from this narrow space and avoid damaging neighbouring tissue. We consider open-loop optimal control of CTCRs parameterized over pseudo-time, primarily aiming at minimizing the robot’s working volume during its motion. External loads acting on the system like tip loads or contact with tissues are not considered here. We also discussed the inclusion of tip’s orientation in the optimal framework to perform some tasks. We recall a quaternion-based formulation of the robot configuration, discuss discretization, develop optimization objectives addressing different criteria, and investigate their impact on robot path planning for several numerical examples. This optimal control framework can be applied to any backbone based continuum robot.
Download
Paper Nr: 45
Title:

A Digital Twin Setup for Safety-aware Optimization of a Cyber-physical System
Authors:

Jalil Boudjadar and Martin Tomko
Abstract: Digital twin technology offers a sophisticated and flexible methodology to design high fidelity models of cyber-physical systems for simulation, optimization, formal verification and validation purposes. This has made such a technology a nascent process being currently adopted in many industries. This paper introduces a digital twin setup for safety-aware performance optimization of a cyber-physical system (Energy Buck converter EBC). This is achieved by designing a high fidelity digital twin model of the Buck converter through synchronization of the model with the physical system, namely calibration. The behavior model is originally built in MATLAB to identify potential runtime optimization patterns using a genetic algorithm. Such a model is translated to a Uppaal model to perform formal verification of the safety properties. The behavior patterns from optimization are provided as inputs to the verification engine for approval, where only valid and feasible patterns are pushed into the actual control loop of EBC. The proposed setup has led to maintain the system safety while optimizing the performance and reducing the output errors.
Download
Paper Nr: 47
Title:

In Situ Calibration Algorithm to Optimize Energy Consumption in an Automotive Stamping Factory Process
Authors:

Ivan Peinado-Asensi, N. Montes and E. García
Abstract: The world’s large factories in all sectors consume a great deal of resources, either raw materials or energy, to develop their products. Saving resources can have a positive impact on the sustainable development of the planet. Automotive manufacturers are a clear example of how to save by investing resources in improving technologies and optimizing processes. This article focuses on one of the most common processes in the automotive sector: the stamping process. For the optimization of this process, previous simulations are usually carried out in order to define the optimal parameters and which should only be applied for a correct operation. The real circumstances of the plant show there is a large discrepancy between the parameters obtained by simulation and the real process because of the difference in material properties, lubrication, press operation, etc. The solution is that the operators must adjust the parameters a posteriori and the only criterion to follow is obtaining the right quality of the part. In many cases, the parameters are well above the ideal. This article presents some algorithms used in order to perform an in situ calibration of the stamping presses to find the press parameters that, guaranteeing the quality of the part, allow to adjust the energy consumption to the minimum. At the end of this article the experimental results from this in-situ calibration process and the energy savings are shown.
Download
Paper Nr: 48
Title:

Comparison of Different Excitation Strategies for Fault Diagnosis of Belt Drives: Industrial Application Scenarios
Authors:

Moritz Fehsenfeld, Johannes Kühn, Zygimantas Ziaukas and Hans-Georg Jacob
Abstract: Machine learning (ML) has received a lot of attention in solving fault diagnosis (FD) tasks. As a result, more and more advanced machine learning algorithms have been developed to increase accuracy. But the system’s excitation has likewise a high impact on the diagnosis performance and applicability. For this purpose, we describe different industrial application scenarios and the related set trajectory. They are divided into passive FD, where normal operation data serves as the input, and active FD, where an optimized excitation is injected. All scenarios are investigated concerning achievable accuracy and data requirement based on comprehensive measurements. We demonstrate that in active scenarios a high accuracy of 97:6% combined with a small number of measurements are obtained by very basic algorithms like a one-nearest neighbor with Euclidean distance. In passive scenarios, where the FD task is generally harder, the demand for large datasets and more advanced ML methods increases. In this way, we illustrate how intelligent use of an optimized excitation strategy leads to feasible, reliable, and accurate fault diagnosis with a broad industrial application spectrum.
Download
Paper Nr: 50
Title:

Intersection Traffic State Estimation using Speed Transition Matrix and Fuzzy-based Systems
Authors:

Željko Majstorović, Leo Tišljarić, Edouard Ivanjko and Tonči Carić
Abstract: Urban traffic congestion is a significant problem for almost every city, affecting various aspects of life. Besides increasing travel time, congestion also affects air and life quality causing economic losses. The construction of infrastructure to solve congestion problems is not always feasible, and, at the end, attracts only additional traffic demand. Thus, a better approach for solving the problem of city congestion is by optimal management of the existing infrastructure. Timely detection of traffic congestion on the road level can prevent congestion formation and even improve road network capacity when used for appropriate traffic control actions. Detecting congestion is a complex process that depends on available traffic data. In this paper, for traffic state estimation, including congestion level, at the intersection level, a new method based on Speed Transition Matrix and Fuzzy-Based System is presented. The proposed method utilizes the Connected Vehicle environment. It is tested on a model of an isolated intersection made in SUMO simulation software based on real-world traffic data. The validation results confirm the successful detection of traffic state (congestion level) at intersections.
Download
Paper Nr: 57
Title:

Learning Optimal Robot Ball Catching Trajectories Directly from the Model-based Trajectory Loss
Authors:

Arne Hasselbring, Udo Frese and Thomas Röfer
Abstract: This paper is concerned with learning to compute optimal robot trajectories for a given parametrized task. We propose to train a neural network directly with the model-based loss function that defines the optimization goal for the trajectories. This is opposed to computing optimal trajectories and learning from that data and opposed to using reinforcement learning. As the resulting optimization problem is very ill-conditioned, we propose a preconditioner based on the inverse Hessian of the part of the loss related to the robot dynamics. We also propose how to integrate this into a commonly used dataflow-based auto-differentiation framework (TensorFlow). Thus it keeps the framework’s generality regarding the definition of losses, layers, and dataflow. We show a simulation case study of a robot arm catching a flying ball and keeping it in the torus shaped bat. The method can also optimize “voluntary task parameters”, here the starting configuration of the robot.
Download
Paper Nr: 59
Title:

Maximum Correntropy Criterion-based UKF for Tightly Coupling INS and UWB with non-Gaussian Uncertainty Noise
Authors:

Seong Y. Cho, Jae H. Lee and Chan G. Park
Abstract: In this paper, unscented Kalman filter (UKF) based on maximum correntropy criterion (MCC) instead of minimum mean square error (MMSE) criterion, and it is applied to tightly coupled integration of inertial navigation system (INS) and ultra wide-band (UWB). UWB can measure distance with an accuracy of less than 30cm in line-of-sight environment, but provides distance measurement with various types of non-Gaussian uncertainty noise in non-line-of-sight environment. In this case, if the INS/UWB system is configured with the existing MMSE-based filter, a large error occurs. To solve this problem, in this paper, UKF is designed based on MCC. Through simulation analysis, it is confirmed that the proposed filter has robust characteristics against UWB uncertainty and enables stable INS/UWB integration.
Download
Paper Nr: 61
Title:

Sensorless Condition Monitoring of Feed Axis Components in Production Systems by Applying Prony Analysis
Authors:

Chris Schöberlein, Johannes Quellmalz, Holger Schlegel and Martin Dix
Abstract: Condition monitoring of modern production systems has established itself as an independent area of research in recent years. Main goal is to achieve an increase in machine productivity by reducing downtime and maintenance costs. In particular, the installed electromechanical axes offer great potential for improvement. Besides an installation of additional sensors, modern drive systems also provide various signals suitable for superordinated monitoring systems. The paper presents a novel approach for monitoring of specific mechanical axis components based solely on internal control loop signals. Fundamental idea is to combine a parametric approach for vibration analysis, the so-called Prony analysis, with a drive-based setpoint generation and data aquisition. The method is verified by detecting emulated malfunctions on a single-axis test stand and a three-axis vertical milling machining center. Experimental investigations prove that the presented approach is capable of reliably detecting the artificially introduced defects on different axis components.
Download
Paper Nr: 64
Title:

A Planning Tool for COD Flow Optimisation to a Waste Water Treatment Plant
Authors:

Kirsten M. Nielsen and Tom S. Pedersen
Abstract: The waste water flows to a typical wastewater treatment plant (WWTP) is comprised from periodic domestic flows and more stochastic industrial flows. Especially variations in the flow of Chemical Oxygen Demand (COD) at the inlet to the WWTP are problematic due to the biological purification process and bio gas production. Traditionally the inlet is un-controlled. A way to reduce variations is to insert a buffer tank near the industrial areas and control the tank outlet according to a prediction of household COD flow. As a first step a planning tool for operator control of the buffer tank outlet 24 hours ahead is designed. The WWTP in the Danish town Fredericia is used as a case. At the moment the only on-line measurement is the inlet flow to the wastewater treatment plant and reliable measurements in the network are difficult to establish. A Model Predictive Control scheme is shown to be able to give considerable reduction in the COD flow variations. To do this two models are introduced; one describing the buffer tank and sewer network from the tank to the WWTP and one describing the daily variations in the household flow. Additionally prediction of the industrial outlet is included. The control scheme has been tested showing good results in a SWMM simulation environment (ProtectionAgency, 2016) based on network architecture and measurements in Fredericia.
Download
Paper Nr: 71
Title:

Nonlinear Set-based Model Predictive Control for Exploration: Application to Environmental Missions
Authors:

A. Anderson, J. G. Martin, N. Bouraqadi, L. Etienne, K. Langueh, L. Rajaoarisoa, G. Lozenguez, L. Fabresse, J. M. Maestre and E. Duviella
Abstract: Acquiring vast and reliable data of physicochemical parameters is critical to environment monitoring. In the context of water quality analysis, data collection solutions have to overcome challenges related to the scale of environments to be explored. Sites to monitor can be large or remote. These challenges can be approached by the use of Unmanned Vehicles (UVs). Robots provide both flexibility on intervention plans and technological methods for real-time data acquisition. Being autonomous, UVs can explore areas difficult to access or far from the shore. This paper presents a nonlinear Model Predictive Control (MPC) for UV-based exploration. The strategy aims to improve the data collection of physicochemical parameters with the use of an Unmanned Surface Vehicle (USV) targeting water quality analysis. We have performed simulations based on real field experiments with a SPYBOAT® on the Heron Lake in Villeneuve d’Ascq, France. Numerical results suggest that the proposed strategy outperforms the schedule of mission planning and exploration for large areas.
Download
Paper Nr: 85
Title:

An Information System for Air Quality Monitoring using Mobile Sensor Networks
Authors:

Pedro Mariano, Susana M. Almeida, Alexandre Almeida, Carolina Correia, Vânia Martins, José Moura, Tomás Brandão and Pedro Santana
Abstract: Engineering the information system that runs a heterogeneous mobile sensor network is a complex task. In this paper we present the solution that was developed in the context of the ExpoLIS project. The goal of this project is to deploy a network of mobile (low-cost) sensors in city buses. Besides the software that needs to transfer, process, and store sensor data, we also developed a mobile application to increase awareness on air pollution, and a tool that allows scientists to subscribe to sensor data. We present the engineering solutions that form the backbone of the information system, and the structure and design of developing supporting tools. We discuss our choices regarding how sensor data are processed in order to make these data available for the common citizen. We mention possible future directions for the software that we have developed.
Download
Paper Nr: 94
Title:

Finger Type Classification with Deep Convolution Neural Networks
Authors:

Yousif A. Al-Wajih, Waleed M. Hamanah, Mohammad A. Abido, Fouad Al-Sunni and Fakhraddin Alwajih
Abstract: The Automated Fingerprint Identification System (AFIS) is a biometric identification methodology that uses digital imaging technology to obtain, store, and analyse fingerprint information. There has been an increased interest in fingerprint-based security systems with the rise in demand for collecting demographic data through security applications. Reliable and highly secure, these systems are used to identify people using the unique biometric information of fingerprints. In this work, a learning-based method of identifying fingerprints was investigated. Using deep learning tools, the performance of the AFIS in terms of search time and speed of matching between fingerprint databases was successfully enhanced. A convolutional neural network (CNN) model was proposed and developed to classify fingerprints and predict fingerprint types. The proposed classification system is a novel approach that classifies fingerprints based on figure type. Two public datasets were used to train and evaluate the proposed CNN model. The proposed model achieved high validation accuracy with both databases, with an overall accuracy in predicting fingerprint types at around 94%.
Download
Paper Nr: 98
Title:

Space-filling Optimization of Excitation Signals for Nonlinear System Identification
Authors:

Volker Smits and Oliver Nelles
Abstract: The focus of this paper is on space-filling optimization of excitation signals for nonlinear dynamic multi-variate systems. Therefore, the study proposes an extension of the Genetic Optimized Time Amplitude Signal (GOATS) to multi-variate nonlinear dynamic systems, an incremental version of GOATS (iGOATS), a new space-filling loss function based on Monte Carlo Uniform Distribution Sampling Approximation (MCUDSA), and a compression algorithm to significantly speed up optimizations of space-filling loss functions. The results show that the GOATS and iGOATS significantly outperform the state-of-the-art excitation signals Amplitude Pseudo Random Binary Signal (APRBS), Optimized Nonlinear Input Signal (OMNIPUS), and Multi-Sine in the achievable model performances. This is demonstrated on a two-dimensional artificially created nonlinear dynamic system. Beside the good expectable model quality, the GOATS and iGOATS are suitable for the usage for stiff systems, supplementing existing data, and easy incorporation of constraints.
Download
Paper Nr: 100
Title:

Improving the Positional Accuracy of Industrial Robots by Forward Kinematic Calibration using Laser Tracker System
Authors:

Mojtaba A. Khanesar, Samanta Piano and David Branson
Abstract: Precision positioning of industrial robots is a vital requirement on the factory floor. Robot end effector positioning using joint angle readings from joint encoders and industrial robot forward kinematics (FKs) is a common practice. However, mechanical wear, manufacturing and assembly tolerances, and errors in robot dimension measurement result in parameter uncertainties in the robot FK model. Uncertainties in robot FK result in inaccurate position measurement. In this paper, we use a multi-output least squares support vector regression (MLS-SVR) method to improve the positioning accuracies of industrial robots using a highly accurate laser tracker system, Leica AT960-MR. This equipment is a non-contact metrology one capable of performing measurements with error of less than 3/ . To perform this task, industrial robot FK is formulated as a regression problem whose unknown parameters are tuned using laser tracker position data as target values. MLS-SVR algorithm is used to estimate the industrial robot FK parameters. It is observed that using the proposed approach, the accuracy of industrial robot FKs in terms of mean absolute errors of static and near-static motion in all three dimensions decreases from its measured value: from 71.9 to 20.9 (71% decrease).
Download
Paper Nr: 104
Title:

Explainable AI based Fault Detection and Diagnosis System for Air Handling Units
Authors:

Juri Belikov, Molika Meas, Ram Machlev, Ahmet Kose, Aleksei Tepljakov, Lauri Loo, Eduard Petlenkov and Yoash Levron
Abstract: Fault detection and diagnosis (FDD) methods are designed to determine whether the equipment in buildings is functioning under normal or faulty conditions and aim to identify the type or nature of a fault. Recent years have witnessed an increased interest in the application of machine learning algorithms to FDD problems. Nevertheless, a possible problem is that users may find it difficult to understand the prediction process made by a black-box system that lacks interpretability. This work presents a method that explains the outputs of an XGBoost-based classifier using an eXplainable Artificial Intelligence technique. The proposed approach is validated using real data collected from a commercial facility.
Download
Paper Nr: 4
Title:

LMI Stability Condition for NCS with Packet Delay and Event-triggered Control
Authors:

M. S. Fadali
Abstract: This paper presents a controller design for networked control systems (NCS) with packet delay and event-triggered control. The total network delay is assumed to be an integer multiple of a fixed sampling period so that the overall system is time-varying with each model depending on the number of time delays. The design methodology is applicable to an arbitrary number of packet delays, regardless of whether the delays are random or deterministic. The methodology is applied to a simple example and Monte Carlo simulation results show that the controller stabilizes the NCS and is robust with respect to random variations in the sampling period and to changes in the probability of packet delays.
Download
Paper Nr: 44
Title:

Input-Output Multiobjective Optimization Approach for Food-Energy-Water Nexus
Authors:

Isaac Okola
Abstract: Food, energy and water are essential for human survival. These resources consume each other thus enhancing security in one resource can reduce security in another resource. Multiobjective optimization approaches have been used to understand the complexity associated with the Food, Energy Water (FEW) Nexus. However most of these approaches focus on either maximizing resource production or minimizing resource consumption in the FEW Nexus but not addressing the two simultaneously. To achieve sustainability of the FEW Nexus sustainable consumption and production of the resources need to be emphasized. In this paper, the Input-Output theory is used to develop a multiobjective optimization approach that minimises resource intensities. Minimising resource intensities results into minimised consumption and maximised production of resources in the nexus. Using the developed approach simulations are carried out to demonstrate its applicability in FEW Nexus. The results show that the approach can be used to explore alternative ways of minimizing consumption and maximizing production simultaneously based on the concept of non-dominated solutions.
Download
Paper Nr: 49
Title:

A Novel Real-Time Wear Detection System for the Secondary Circuit of Resistance Welding Guns
Authors:

D. Ibáñez, E. García, J. Martos and J. Soret
Abstract: Currently, many resources are invested in high-production automotive factories to correct quality defects caused in the bodywork due to secondary circuit wear. In the same way, energy losses are generated due to the increase in resistance caused by secondary wear, thus reducing efficiency and increasing the final cost of the product. This happens because, at present, there is no method that allows the predictive detection of problems in the secondary and the arms of the welding gun. Consequently, a solution must be developed to carry out predictive maintenance applicable to the automotive industry to detect this defect. This research provides an answer by proposing a method to detect variations in the state of the secondary of the welding gun using existing data in the welding process, specifically, the evolution of the angle of degassing of the IGBTs of the welding control. To validate the relationship between the control shift angle and the increase in wear, an electronic simulation software was used to simulate the behaviour of the real welding control.
Download

Area 3 - Robotics and Automation

Full Papers
Paper Nr: 13
Title:

Path Planning Incorporating Semantic Information for Autonomous Robot Navigation
Authors:

Silya Achat, Julien Marzat and Julien Moras
Abstract: This paper presents an approach to take into account semantic information for autonomous robot tasks requiring path planning capabilities. A semantic pointcloud or map serves as input for generating a multi-layered map structure, which can then be exploited to address various navigation goals and constraints. Semantic-aware adaptations of A* , Transition-based RRT and a shortcut algorithm are derived in this framework, and evaluated numerically on an exploration and observation task using a reference dataset with multiple semantic classes as an illustrative test environment.
Download
Paper Nr: 24
Title:

Interval-based Robot Localization with Uncertainty Evaluation
Authors:

Yuehan Jiang, Aaronkumar Ehambram and Bernardo Wagner
Abstract: Being able to provide trustworthy localization for a robot in a map is essential for various tasks with safety-related requirements. In contrast to classical probabilistic approaches that represent the uncertainty as a Gaussian distribution, we use interval error bounds for the uncertainty estimation of a localization problem. To tackle and identify the limitations of probabilistic localization uncertainty estimation, we carry out comparison experiments between an interval-based method and a factor graph-based probabilistic method. Different measurement error models are propagated by the two methods to derive the robot pose uncertainty estimates. Results show that the probabilistic approach can provide very good pose uncertainty when there is no non-Gaussian systematic sensor error. However, if the measurements have unmodeled systematic errors, the interval approach is able to robustly contain the true poses whereas the probabilistic approach gives completely wrong results.
Download
Paper Nr: 37
Title:

Mutual Relative Localization in Heterogeneous Air-ground Robot Teams
Authors:

Samet Güler, İ. E. Yıldırım and H. H. Alabay
Abstract: Air and ground robots with distinct sensing characteristics can be combined in a team to accomplish demanding tasks robustly. A key challenge in such heterogeneous systems is the design of a local positioning methodology where each robot estimates its location with respect to its neighbors. We propose a filtering-based relative localization algorithm for air-ground teams composed of vertical-take-off-and-landing drones and unmanned aerial vehicles. The team members interact through a sensing/communication mechanism relying on onboard units, which results in a mutual connection between the air and ground components. Exploiting the supplementary features of omnidirectional distance sensors and monocular cameras, the framework can function in all environments without fixed infrastructures. Various simulation and experiment results verify the competency of our approach.
Download
Paper Nr: 56
Title:

Triangular Expansion Revisited: Which Triangulation Is The Best?
Authors:

Jan Mikula and Miroslav Kulich
Abstract: Visibility region is a classic structure in computational geometry that finds use in many agent planning prob-lems. Triangular expansion algorithm (TEA) is the state-of-the-art algorithm for computing visibility regions within polygons with holes in two dimensions. It has been shown that it is two orders of magnitude faster than the traditional rotation sweep algorithm for real-world scenarios. The algorithm triangulates the underlying polygon and recursively traverses the triangulation while keeping track of the visible region. Instead of the constraint Delaunay triangulation used by default, this paper introduces the idea of optimizing the triangulation to minimize the expected number of triangle edges expanded during the TEA’s traversal while assuming that every point of the input polygon is equally likely to be queried. The proposed triangulation is experimentally evaluated and shown to improve TEA’s mean query time in practice. Furthermore, the TEA is modified to consider limited visibility range of real-life sensors. Combined with the proposed triangulation, this adjustment significantly speeds up the computation in scenarios with limited visibility. We provide an efficient open-source implementation called TriVis which, besides the mentioned, includes determining visibility between two points and other useful visibility-related operations.
Download
Paper Nr: 63
Title:

Mechanical Design of an Assistive Robotic System for Bilateral Elbow Tendinopathy Rehabilitation
Authors:

Andres Guatibonza, Carlos Zabala, Leonardo Solaque, Alexandra Velasco and Lina Peñuela
Abstract: Diseases related to upper limb mobility are increasingly common among the actual population. For this reason, robotic physical assistive systems have been proposed to support therapy processes and improve the functional capabilities of people. However, there are still open issues related to mechanical design, such as joint coupling and bidirectional configurations. In this work, we present a novel design of a 7 DoF robotic assistive system with anthropometric adjustment, arm change configuration for elbow tendinopathies rehabilitation to use it in both arms. The design is supported by the analysis of the upper limb pathophysiology and the exercises required to treat elbow tendinopathies.
Download
Paper Nr: 84
Title:

Practical Formation Acquisition Mechanism for Nonholonomic Leader-follower Networks
Authors:

Kader M. Kabore and Samet Güler
Abstract: A grand challenge lying ahead of the realization of multi-robot systems is the lack of an adequate coordination mechanism with reliable localization solutions. In some workspaces, external infrastructure needed for precise localization may not be always available to the MRS, e.g., GPS-denied environments, and the robots may need to rely on their onboard resources without explicit communication. We address the practical formation control of nonholonomic ground robots where external localization aids are not available. We propose a systematic framework for the formation maintenance problem that is composed of a localization module and a control module. The onboard localization module relies on heterogeneity in sensing modality comprised of ultrawideband, 2D LIDAR, and camera sensors. Particularly, we apply deep learning-based object detection algorithm to detect the bearing between robots and fuse the outcome with ultrawideband distance measurements for precise relative localization. Integration of the localization outcome into a distributed formation acquisition controller yields high performance. Furthermore, the proposed framework can eliminate the magnetometer sensor which is known to produce unreliable heading readings in some environments. We conduct several realistic simulations and real world experiments whose results validate the competency of the proposed solution.
Download
Paper Nr: 108
Title:

Robot Collision Avoidance based on Artificial Potential Field with Local Attractors
Authors:

Matteo Melchiorre, Leonardo S. Scimmi, Laura Salamina, Stefano Mauro and Stefano Pastorelli
Abstract: This paper presents a novel collision avoidance technique that allows the robot to reach a desired position by avoiding obstacles passing through preferred regions. The method combines the classical elements of the artificial potential fields in an original manner by handling local attractors and repulsors. The exact solution, which is given in a closed form, allows to sculpt a potential field so that local minima related to the local attractors are prevented and the global minimum is unperturbed. The results show the algorithm applied to mobile robot navigation and prove the capability of local attractors to influence the robot path.
Download
Short Papers
Paper Nr: 1
Title:

Design and Locomotion Control of a Myliobatid-inspired Robot Actuated by Passively-flexing Pectoral Fins
Authors:

Songzi Guo, Zhiyin Li and Jinhua Zhang
Abstract: This article proposes the mechanical design of a myliobatid-inspired robot (XJRoman-I) based on oscillatory swimming mechanism for both stability and agile manoeuvrability. Inspired by anatomical studies, a pair of passively bending pectoral fins are developed to generate propulsive force for the prototype. An elevator is adopted to adjust its pitch attitude. Primary experimental research on the effect of fin’s spanwise stiffness on swimming performance is performed to improve its swimming performance. By embedding a stiff rod into the fin’s leading edge, the thrust and lateral force generated by the fins are improved significantly. Finally, a CPG-based control method is introduced to make the prototype achieve different locomotion patterns including cruising by flapping pectoral fins and turning by modulating phase relation of pectoral fins. This paper mainly focuses on propulsive capability of stability and agility for the prototype, and expects to propose an excellent underwater vehicle covering wide range of applications.
Download
Paper Nr: 9
Title:

Safe Robotized Polishing of Plastic Optical Fibers for Plasmonic Sensors
Authors:

Francesco Arcadio, Marco Costanzo, Giulio Luongo, Luigi Pellegrino, Nunzio Cennamo and Ciro Natale
Abstract: Plastic optical fibers (POFs) biosensors are getting widespread in a number of application fields owing to their low cost, high performance, and for their extreme flexibility in terms of detection ability of a large number of specific substances in different matrices. A specific category of such sensors are those based on the surface plasmon resonance (SPR) phenomenon, which can be made very specific by suitable integration with a biological or chemical molecular recognition element (MRE), specifically designed for binding with the desired substance (the analyte). Despite the flexibility of the SPR-POF sensors, their production is still difficult to automate on a large scale because of the special polishing process of the plastic optical fiber. Such a process is currently performed by a human trained operator who rubs the surface of a short fiber segment against a sandpaper sheet by following an 8-shaped path while exerting a specific force in the direction normal to the contact surface. The present paper proposes the adoption of a collaborative robot programmed to perform the same task based on the data acquired from the human operator. To ensure the safe use of the robotic cell by operators who share the same workspace of the robot, the system is endowed with a workspace monitoring system that ensures the polishing task execution while minimizing the possible occurrence of collisions with human operators by suitable exploiting the kinematic redundancy of the robot.
Download
Paper Nr: 10
Title:

Nonholonomic Robot Navigation of Mazes using Reinforcement Learning
Authors:

Daniel Gleason and Michael Jenkin
Abstract: Developing a navigation function for an unknown environment is a difficult task, made even more challenging when the environment has complex structure and the robot imposes nonholonomic constraints on the problem. Here we pose the problem of navigating an unknown environment as a reinforcement learning task for an Ackermann vehicle. We model environmental complexity using a standard characterization of mazes, and we show that training on complex maze architectures with loops (braid and partial braid mazes) results in an effective policy, but that for a more efficient policy, training on mazes without loops (perfect mazes) is to be preferred. Experimental results obtained in simulation are validated on a real robot operating both indoors and outdoors, assuming good localization and a 2D LIDAR to recover the local structure of the environment.
Download
Paper Nr: 17
Title:

Calibration of a 2D Scanning Radar and a 3D Lidar
Authors:

Jan M. Rotter and Bernardo Wagner
Abstract: In search and rescue applications, mobile robots have to be equipped with robust sensors that provide data under rough environmental conditions. One such sensor technology is radar which is robust against low-visibility conditions. As a single sensor modality, radar data is hard to interpret which is why other modalities such as lidar or cameras are used to get a more detailed representation of the environment. A key to successful sensor fusion is an extrinsically and intrinsically calibrated sensor setup. In this paper, a target-less calibration method for scanning radar and lidar using geometric features in the environment is presented. It is shown that this method is well-suited for in-field use in a search and rescue application. The method is evaluated in a variety of use-case relevant test scenarios and it is demonstrated that the calibration results are accurate enough for the target application. To validate the results, the proposed method is compared to a target-based state-of-the-art calibration method showing equivalent performance without the need for specially designed targets.
Download
Paper Nr: 34
Title:

Toward Autonomous Mobile Robot Navigation in Early-Stage Crop Growth
Authors:

Luis Emmi, Jesus Herrera-Diaz and Pablo Gonzalez-de-Santos
Abstract: This paper presents a general procedure for enabling autonomous row following in crops during early-stage growth, without relying on absolute localization systems. A model based on deep learning techniques (object detection for wide-row crops and segmentation for narrow-row crops) was applied to accurately detect both types of crops. Tests were performed using a manually operated mobile platform equipped with an RGB and a time-of-flight (ToF) cameras. Data were acquired during different time periods and weather conditions, in maize and wheat fields. The results showed the success on crop detection and enables the future development of a fully autonomous navigation system in cultivated fields during early stage of crop growth.
Download
Paper Nr: 39
Title:

Learning Human-like Driving Policies from Real Interactive Driving Scenes
Authors:

Yann Koeberle, Stefano Sabatini, Dzmitry Tsishkou and Christophe Sabourin
Abstract: Traffic simulation has gained a lot of interest for autonomous driving companies for qualitative safety evaluation of self driving vehicles. In order to improve self driving systems from synthetic simulated experiences, traffic agents need to adapt to various situations while behaving as a human driver would do. However, simulating realistic traffic agents is still challenging because human driving style cannot easily be encoded in a driving policy. Adversarial Imitation learning (AIL) already proved that realistic driving policies could be learnt from demonstration but mainly on highways (NGSIM Dataset). Nevertheless, traffic interactions are very restricted on straight lanes and practical use cases of traffic simulation requires driving agents that can handle more various road topologies like roundabouts, complex intersections or merging. In this work, we analyse how to learn realistic driving policies on real and highly interactive driving scenes of Interaction Dataset based on AIL algorithms. We introduce a new driving policy architecture built upon the Lanelet2 map format which combines a path planner and an action space in curvilinear coordinates to reduce exploration complexity during learning. We leverage benefits of reward engineering and variational information bottleneck to propose an algorithm that outperforms all AIL baselines. We show that our learning agent is not only able to imitate humane like drivers but can also adapts safely to situations unseen during training.
Download
Paper Nr: 46
Title:

Proportional Integral Derivative Decentralized Control vs Linear Quadratic Tracking Regulator in Vehicle Overtaking within a Platoon
Authors:

Alessandro Bozzi, Roberto Sacile and Enrico Zero
Abstract: This paper introduces a comparison between a decentralized Proportional Integral Derivative (PID) controller and a centralized Linear Quadratic Tracking (LQT) controller to automatise the exchange of two inner vehicles inside a platoon moving on a straight path. Lomonossoff’s model is used to represent vehicle’s longitudinal dynamics. A case study is presented to demonstrate the effectiveness of both controllers respectively on nonlinear and linearized model.
Download
Paper Nr: 52
Title:

Generation and Quality Evaluation of a 360-degree View from Dual Fisheye Images
Authors:

María Flores, David Valiente, Juan J. Cabrera, Oscar Reinoso and Luis Payá
Abstract: 360-degree views are beneficial in robotic tasks because they provide a compact view of the whole scenario. Among the different vision systems to generate this image, we use a back-to-back pair of fisheye lens cameras by Garmin (VIRB 360). The objectives of this work are twofold: generating a high-quality 360-degree view using different algorithms and performing an analytic evaluation. To provide a consistent evaluation and comparison of algorithms, we propose an automatic method that determines the similarity of the overlapping area of the generated views as regards a reference image, in terms of a global descriptor. These descriptors are obtained from one of the Convolutional Neural Network layers. As a result, the study reveals that an accurate stitching process can be achieved when a high number of feature points are detected and uniformly distributed in the overlapping area. In this case, the 360-degree view generated by the algorithm which employs the camera model provides more efficient stitching than the algorithm which considers the angular fisheye projection. This outcome demonstrates the wrong effects of the fisheye projection, which presents high distortion in the top and bottom parts. Likewise, both algorithms have been also compared with the view generated by the camera.
Download
Paper Nr: 53
Title:

Autonomous Loading of a Washing Machine with a Single-arm Robot
Authors:

Hassan Shehawy, Andrea M. Zanchettin and Paolo Rocco
Abstract: The perception and autonomous manipulation of clothes by robots is an ongoing research topic that is attracting a lot of contributions. We consider the application of handling garments for laundry in this work. A framework for loading a washing machine with clothes placed initially inside a box is presented. Our framework is created in a modular way to account for the sub-problems associated with the full process. We extend our grasping point estimation algorithm by finding multiple grasping points and defining a score to select one. Active contours segmentation is added to the algorithm as well for more robust clustering of the image. Model of the washing machine is used to create a motion plan for the robot to place the clothes inside the drum. A new module is added for detection of items fallen outside the drum so to plan corresponding corrective action. We use ROS, depth and 2D cameras and the Doosan A0509 robot for experiments.
Download
Paper Nr: 54
Title:

Telerobotic Radiation Protection Tasks in the Super Proton Synchrotron using Mobile Robots
Authors:

David Forkel, Enric Cervera, Raúl Marín, Eloise Matheson and Mario Di Castro
Abstract: In this paper a complete robotic solution is presented, which allows the teleoperation of the radiation survey in the Super Proton Synchrotron (SPS) accelerator at CERN. Firstly, an introduction to radiation protection is given. Subsequently, the execution of the radiation survey in person is described and the potential of robotic solutions for such missions is outlined. After providing a brief state of the art on the subject, the development of the robot base, as well as its component selection and design is shown. Hereafter, the software implementation is explained. The test procedure of this project includes the most important requirements for a correct execution of the survey, as well as the operational steps and data treatment in detail. The results underline the correct execution of the mission, and show the advantages of the teleoperated robotic solution, such as the improved and unified measurement conditions. Thus, this robotic system will allow to significantly reduce the radiation dose of the radiation protection staff. For further development, the automation of this task is planned, which presupposes the gradual autonomization of the robotic system from assisting the user to the self-reliant execution of the survey.
Download
Paper Nr: 65
Title:

World State-dependent Action Graph: A Representation of Action Possibility and Its Variations in Real Space based on World State
Authors:

Yosuke Kawasaki and Masaki Takahashi
Abstract: For intelligent systems, it is important to understand the action possibility for agent in real space. As the action possibility varies with the subsystem configuration of the agent and its states, the possibilities should be understood based on the world state comprising the agent’s state as well as the environmental state. However, most conventional methods consider only the environmental state. Therefore, this study proposes a world state-dependent action graph based on knowledge representation using scene graphs which allows the capturing of the action possibility of agents, which implies the feasible actions and their positions in real space, and their recursive variations depending on the world state. Moreover, the effectiveness of the proposed method was verified with simulations, assuming a coffee shop environment.
Download
Paper Nr: 74
Title:

Task and Motion Planning Methods: Applications and Limitations
Authors:

Kai Zhang, Eric Lucet, Julien D. Sandretto, Selma Kchir and David Filliat
Abstract: Robots are required to perform more and more complicated tasks, which raises the requirement of more intelligent planning algorithms. As a domain having been explored for decades, task and motion planning (TAMP) methods have achieved significant results, but several challenges remain to be solved. This paper summarizes the development of TAMP from solving objectives, simulation environments, methods and remaining limitations. In particular, it compares different simulation environments and methods used in different tasks aiming to provide a practical guide and overview for the beginners.
Download
Paper Nr: 81
Title:

Design of a Switched Control Lyapunov Function for Mobile Robots Aggregation
Authors:

Chrystian P. Yuca Huanca, Gian Paolo Incremona, Roderich Groß and Patrizio Colaneri
Abstract: This paper proposes a novel aggregation strategy for a network of mobile wheeled robots with constrained dynamics. The strategy assumes a centralized control architecture, which collects all the robot positions and generates the control signals sent to the robots in the network. To do this a control Lyapunov function (CLF) based approach is designed relying on a switched formulation of the robot models. Such a formulation is in fact made possible by constraining the robot motion only to rotation and roto-translation in the plane. Moreover, a collision avoidance objective is taken into account in the design of the CLF. The approach is analyzed, and simulations as well as experiments with six robots show its effectiveness and practical applicability.
Download
Paper Nr: 83
Title:

External Force Adaptive Compensator for Serial Manipulators
Authors:

Albert Demian and Alexander Klimchik
Abstract: We propose a preliminary design concept for the external force compensator. An arrangement of lever-wheel arrangement with a group of springs producing counter torque to compensate for external force. The springs are fixed on adjustable pivot points to allow compensation of a range of payloads. We introduce the use of self-locking worm gears to ensure the compensator’s torque is purely applied on either the wheel or the lever. We investigated the compensator design with a 2-DOF manipulator which consists of two orthogonal rotational joints. We present a design methodology to the compensator together with a selection of spring coefficients to match a certain range of payloads. Results of the simulation show complete compensation of external force is possible as compensation of certain components of the force vectors.
Download
Paper Nr: 88
Title:

Challenges of Autonomous In-field Fruit Harvesting and Concept of a Robotic Solution
Authors:

Tim Tiedemann, Florian Cordes, Matthis Keppner and Heiner Peters
Abstract: Since the beginning of humans cultivating plants in fields, agriculture underwent a continuous shift from purely manual labor over simple machinery to more and more automated processes. Autonomous driving with navigation and self localization in the field is state of the art. Also, automated machines for fruit processing are available as well. In cases where the fruit is damageable and varies in size and shape, automated processing is challenging. One example of such damageable fruits are strawberries. Size, weight, and shape at the optimal ripeness can vary a lot. Additionally, a change from ripe to overripe occurs relatively quick and is sometimes hard to recognize. A further challenge when harvesting strawberries is a dense leafage that can cover the fruits partly or completely. In this paper, a concept of an autonomous in-field strawberry harvesting robot for non-elevated but ground-raised strawberry plants, with or without a tunnel, is presented. The robot is supposed to use multi-spectral imaging and machine learning based ripeness classification. Besides the overall concept, first data of this early-stage project is shown, too.
Download
Paper Nr: 105
Title:

A Novel Constrained Trajectory Planner for Safe Human-robot Collaboration
Authors:

Matteo Melchiorre, Leonardo S. Scimmi, Stefano Mauro and Stefano Pastorelli
Abstract: This paper presents a novel collision avoidance algorithm for collaborative robotics that can influence the collision-free trajectory of the robot according to preferred directions with respect to the human posture. The aim is to avoid the human body parts in a controlled manner so that the robot trajectory is predictable. The algorithm is based on closed loop inverse kinematics and uses velocity commands to modify the robot trajectory in real-time. The existing human tracking devices allow to measure the human posture in three dimensions. The idea is to combine the human posture estimation with repulsive volumes, i.e. regions that approximate the human size and that produce repulsive velocities on the robot, and to add attractive surfaces made of cylindrical sectors to condition the avoidance manoeuvre in a chosen direction. The algorithm is tested in a simulation environment built with the model of a collaborative robot and a mock-up of the human, whose motion is generated from real data acquired by 3d vision sensors. The results show the effectiveness of the proposed method during a pick and place task in common scenarios, where the human intersects the robot planned path with different body parts.
Download
Paper Nr: 107
Title:

Smart Autonomous Part Displacement System based on Point Cloud Segmentation
Authors:

Eber S. Gouveia, Rupal Srivastava, Maulshree Singh, Sean Lyons, Eddie Armstrong and Declan Devine
Abstract: Robotic arms are widely used in manufacturing lines to automate the manipulation of products, providing many advantages, such as increasing production and minimizing labour costs. However, most robotic arms operate in a controlled environment, executing predefined movements. Such a feature prevents the robot arm from working in an environment where multiple product types are in different placements. In this way, this concept paper describes the development of a smart robotic system capable of performing an autonomous pick-and-place task of injected moulded parts from the first conveyor belt to the next, based on its spatial data obtained from a 3D scanner. After obtaining the digital point cloud from the moulded part, the PointNet deep learning model was used to segment and then extract the spatial position of its sprue, which is one of the common structures of any moulded part. Finally, the robotic arm combined with its end-effector can pick up these parts regardless of their shape, orientation, and size. The system proposed is composed of three components, i.e., the IRB 1200 robotic arm from ABB, the PhoXi 3D Scanner from Photoneo, and the two-finger gripper PB-0013 from Gimatic. Moreover, all system components were interconnected using Robot Operating System as middleware. This concept paper discusses the setup and plan for the same.
Download
Paper Nr: 109
Title:

Vision-based Sliding Mode Control with Exponential Reaching Law for Uncooperative Ground Target Searching and Tracking by Quadcopter
Authors:

Hamza Bouzerzour, Mohamed Guiatni, Mustapha Hamerlain and Ahmed Allam
Abstract: This paper propose a robust approach based on vision and sliding mode controller for searching and tracking an uncooperative and unidentified mobile ground target using a quadcopter UAV (QUAV). The proposed strategy is an Image-Based Visual Servoing (IBVS) approach using target’s visual data projected in a virtual camera combined with the information provided by the QUAV’s internal sensors. For an effective visual target searching, a circular search trajectory is followed, with a high altitude using the Camera Coverage Area (CCA). A Sliding Mode Controller (SMC) based on Exponential Reaching Law (ERL) is used to ensure the QUAV control in the presence of external disturbances and measurement uncertainties. Simulation results are presented to assess the proposer strategy considering different scenarios.
Download
Paper Nr: 18
Title:

A Novel Connection Mechanism for Dynamically Reconfigurable Modular Robots
Authors:

James White, Mark A. Post and Andy M. Tyrrell
Abstract: This paper describes a novel hermaphroditic, single sided disconnect physical connector for heterogeneous modular robots built using eight permanent magnets arranged in rotating pairs. The connector has 4 rotational degrees of symmetry and incorporates power and data sharing. The connector has been designed as part of a project creating 10 cm cubic heterogeneous modules but could be easily scaled to different sizes for other applications. The paper begins with an introduction to connection mechanisms in modular robots, followed by a detailed description of the design of the connector. A description of the simulation environment created to test systems of interconnected modular robots is given, followed by the implementation and testing of the connection system created.
Download
Paper Nr: 19
Title:

Segmenting Maps by Analyzing Free and Occupied Regions with Voronoi Diagrams
Authors:

Alicia Mora, Adrián Prados and Ramón Barber
Abstract: Traditional mapping techniques rely on metric properties, which represent indoor information with specific geometric characteristics. This fact highly differs from the way in which people interpret their surroundings. By geometrically segmenting occupancy grid maps into rooms, robots are brought closer to the way in which we understand indoor environments. In this work, Voronoi diagrams are proposed as the main tool to locate map partitions. As a novelty, they are extracted from free and occupied spaces to analyze their shape. This allows to locate narrow passages on free zones which coincide with protruding parts on occupied zones, indicating a nearby door. An additional advantage is the use of a varying threshold that depends on the map structure. This dynamic value can adjust to multiple scenarios, avoiding the use of a fixed threshold that cannot be generalized. Experiments have been conducted in multiple maps, showing the potential of the propose method.
Download
Paper Nr: 31
Title:

Comparative Study of a Vacuum Powered Upper Limb Exoskeleton
Authors:

Dimitar Chakarov, Ivanka Veneva and Pavel Venev
Abstract: In the present work, an exoskeleton of upper limb intended for rehabilitation and training is studied. The aim of the work is to find and evaluate an appropriate design solution that provides performance on the one hand and transparency and natural safety on the other. Therefore, a pneumatic drive is proposed and transparency of the exoskeleton is investigated, where positive pressure actuation is compared with vacuum pressure actuation. To assess transparency, the interaction force between the patient and the exoskeleton in passive mode is examined. Simulations and estimates of the interaction force between the patient and the exoskeleton as a result of exoskeleton gravity and the elastic forces from the pneumatic actuation are performed. In this case, the forces in the closed chambers of the pneumatic actuators are used to compensate for the gravitational forces. Simulations are performed with harmonic motion imposed by the patient at one joint of the exoskeleton. The interaction force at the end effector is evaluated in two cases of pneumatic actuation: at pressures higher than atmospheric pressure and at vacuum pressure. The simulation results are shown graphically. A discussion is presented as well as conclusions and directions for future work.
Download
Paper Nr: 68
Title:

Prospects for the Use of Unmanned Ground Vehicles in Artillery Survey
Authors:

Jan Ivan, Michal Sustr, Ondřej Pekar and Ladislav Potuzak
Abstract: The article deals with the currently realized research of a new survey vehicle of the Czech field artillery, which task will be support of the activity of autonomous and non-autonomous artillery weapon systems. The article describes the basic aspects of artillery survey together with the current progress of the project. Baseline for the article is description of current status of Czech artillery survey and the way it supports the artillery operations. The individual chapters then present the identified variants of the functionality of the gun navigation system and the resulting requirements for the capability of the unmanned artillery survey vehicle. Main focus of the article is to present specific approach which Czech armed forces have in terms of artillery use under degraded and GPS denied operations. All these proposals are presented according to current status of Czech artillery which transitions from non-autonomous 152mm howitzers to the new, NATO standard 155mm autonomous weapon systems.
Download
Paper Nr: 79
Title:

Motion Planning for Mobile Robots using the Human Tracking Velocity Obstacles Method
Authors:

Zoltán Gyenes, Ilshat Mamaev, Dongxu Yang, Emese G. Szádeczky-Kardoss and Björn Hein
Abstract: Human-robot interaction is playing an increasingly important role in everyday life and we can expect an even bigger explosion in the use of robots in the future. One such use is where a mobile robot needs to follow the human. The main objective of this paper is to introduce a novel motion planning algorithm for mobile robots, which can be used to enable the robot to follow a human while maintaining a given distance. The motion planning algorithm has to take into account obstacles in the workspace of the robot at each sampling time and to generate a collision-free motion for the agent.
Download
Paper Nr: 91
Title:

Simulation Study on Robot Calibration Approaches
Authors:

Pavel Kozlov and Alexandr Klimchik
Abstract: The paper compares elastostatic calibration approaches for serial industrial robots. Specifically, this paper compares identification strategies based on the different measurement point locations and data fusion algorithms. The paper analyzes several robot calibration hypotheses based on different robot models. All the hypotheses were tested in a simulation study with 1000 data sets. The results showed that “4-6DoF after 6+3DoF” and “3+6DoF comb” methods demonstrated the best results for the considered methods. Strategies were at least 1.86 times more accurate for the resulting deviation metric than the classical “6DoF” identification.
Download
Paper Nr: 95
Title:

Coupled PID-SDRE Controller of a Quadrotor: Positioning and Stabilization of UAV Flight
Authors:

Marcin Chodnicki, Wojciech Stecz, Wojciech Giernacki and Sławomir Stępień
Abstract: This work presents a coupled Proportional-Integral-Derivative and State-Dependent Riccati Equation (PID-SDRE) controller. PID angular position controller coupled to nonlinear infinite-time SDRE controller for speed stabilization is proposed. For the quadrotor modelling a full 6 degree of freedom (DoF) model is considered and described by nonlinear state-space approach. Also, a stable state-dependent parameterization (SDP) necessary for solution of the SDRE control problem is proposed. Solution of the SDRE control problem with adequate defined weighting matrices in the performance index shows the possibility of fast and precise quadrotor positioning with optimal stabilization of speeds. Two methods of optimal SDRE-based stabilization are proposed, tested, and compared.
Download
Paper Nr: 103
Title:

A Single Motor Driving and Steering Mechanism for a Transformable Bicycle
Authors:

Kazuki Sekine and Ikuo Mizuuchi
Abstract: This research aims to propose a bicycle capable of transforming into a stable form suitable for autonomous driving, and achieving both driving and steering with a single motor, using differential drive method. A novel mechanism of one-motor differential drive using bevel gears and one-way clutches was devised. Then, a prototype without transforming mechanism was fabricated. An experiment was conducted to demonstrate that differential drive with a singke motor is possible. In the experiment, the prototype was capable of running in straight lines and curves with small meandering. Next, to formulate the deceleration of the non-drive side wheel in the proposed drive mechanism, another series of experiments was conducted. The equation for the change in wheel rotating speed derived from the results enables accurate estimation of the future position of the prototype, allowing it to run autonomously in further research.
Download
Paper Nr: 110
Title:

Gear Wheels based Simulation of Crawlers for Mobile Robot Servosila Engineer
Authors:

Ruslan Gabdrahmanov, Tatyana Tsoy, Yang Bai, Mikhail Svinin and Evgeni Magid
Abstract: In a process of research, it is beneficial to test new theories and early stage developments in virtual worlds of an adequate realistic simulation before starting real world experiments. While modelling of wheeled mobile robots is well-studied and typically does not imply significant difficulties, a realistic modelling of a crawler robot is a complicated task. This paper discusses several existing approaches for a crawler robot modelling in Gazebo simulator and presents a new approach, which approximates each crawler with a set of gear wheels. We compared several approaches for Servosila Engineer crawler robot modelling in Gazebo by their climbing capabilities, velocity, acceleration and real time factor parameters with regard to the real robot. The comparison results demonstrated that the new approach is feasible in terms of CPU load and provides a better approximation to the real robot performance. Moreover, it successfully eliminated an issue of a crawler seizure while climbing sharp edges of obstacles, which is typical for pseudo-wheels based approaches.
Download

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

Full Papers
Paper Nr: 11
Title:

Optimal Prediction of Tessarine Signals from Multi-sensor Uncertain Observations under Tk-Properness Conditions
Authors:

José D. Jiménez-López, Rosa M. Fernández-Alcalá, Jesús Navarro-Moreno and Juan C. Ruiz-Molina
Abstract: In this paper, the optimal one-stage prediction problem of tessarine signals from multi-sensor uncertain observations is approached. At each instant of time, there exists a non-null probability that the observation tessarine component coming from each sensor, contains the corresponding signal component, or only noise. To model the uncertainty, multiplicative noises modeled by Bernoulli random variables are included in the observation equations. Under correlation hypotheses between the signal and observation additive noises, a recursive algorithm to calculate the optimal least-squares linear predictor of the signal and its mean-squared error is proposed, derived by using an innovation approach. The theoretical results are illustrated by means of a numerical simulation example, in which the performance of the proposed estimator is evaluated under different uncertainty probabilities.
Download
Paper Nr: 21
Title:

Importance Order Ranking for Texture Extraction: A More Efficient Pooling Operator Than Max Pooling?
Authors:

Vincent Vigneron, S. V. Ibarra, Hichem Maaref and J.-Ph. Conge
Abstract: Much of convolutional neural network (CNN)’s success lies in translation invariance. The other part resides in the fact that thanks to a judicious choice of architecture, the network is able to make decisions taking into account the whole image. This work provides an alternative way to extend the pooling function, we named rank-order pooling, capable of extracting texture descriptors from images. The rank-order pooling layers are non parametric, independent of the geometric arrangement or sizes of the image regions, and can therefore better tolerate rotations. Rank-order pooling functions produce images capable of emphasizing low/high frequencies, contours, etc. We shows rank-order pooling leads to CNN models which can optimally exploit information from their receptive field.
Download
Paper Nr: 25
Title:

Persistent Homology based Classification of Chaotic Multi-variate Time Series with Application to EEG Data
Authors:

Martina Flammer and Knut Hüper
Abstract: An application of persistent homology for detection of epileptic events in EEG data is presented. Given point cloud data, persistent homology is a tool from topological data analysis to describe the structure of the underlying space on which the data was sampled by utilizing topological invariants and tracking their behavior on several spatial scales. As a preprocessing step, a novel method called Dynamical Component Analysis is used that reduces the dimension of a multi-variate time series by incorporating information about the dynamics of the system. The results show that our proposed method is appropriate to detect the occurence of petit-mal epileptic seizures in EEG signals.
Download
Paper Nr: 28
Title:

Control-relevant Model Selection for Multiple-mass Systems
Authors:

Mathias Tantau, Torben Jonsky, Zygimantas Ziaukas and Hans-Georg Jacob
Abstract: Physically motivated parametric models are the basis of several techniques related to control design. Industrial model-based controller tuning methods include pole placement, symmetric optimum and damping optimum. The challenge is that the resulting model-based controller is satisfactory only if the underlying model is appropriate. Typically, a set of potential models is known a priori, but it is not known, which model should be used. So, the critical question in model-based controller tuning is that of model selection. Existing approaches for model selection are mostly based on maximizing accuracy, but there is no reason why the most accurate model should also be the optimal model for control design. Given the overall aim to design a high-performance controller, in this paper the best model is considered as the one that has the potential to give a model-based controller the highest performance. The proposed method identifies parametric candidate models for control design. Then, a nonparametric model is used to predict the actual performance of the various controllers on the real system. A validation with two industry-like testbeds shows success of the method.
Download
Paper Nr: 66
Title:

Optimal Resource Allocation for Fast Epidemic Monitoring in Networked Populations
Authors:

Paolo Di Giamberardino, Daniela Iacoviello and Federico Papa
Abstract: The COVID-19 pandemic highlighted the fragility of the world in addressing a global health threat. The available resources of the pre-pandemic national health systems were inadequate to cope with the huge number of infected subjects needing health care and with the rapidity of the infection spread characterizing the COVID-19 outbreak. Indeed, an adequate allocation of the resources could produce in principle a strong reduction of the infection spread and of the hospital burden, preventing the collapse of the health system. In this work, taking inspiration from the COVID-19 and the difficulties in facing the emergency, an optimal problem of resource allocation is formulated on the basis of an ODE multi-group model composed by a network of SEIR-like submodels. The multi-group structure allows to differentiate the epidemic response of different populations or of various subgroups in the same population. In fact, an epidemic does not affect all populations in the same way, and even within the same population there can be epidemiological differences, like the susceptibility to the virus, the level of infectivity of the infectious subjects and the recovery from the disease. The subgroups are selected within the total population based on some peculiar characteristics, like for instance age, work, social condition, geographical position, etc., and they are connected by a network of contacts that allows the virus circulation within and among the groups. The proposed optimal control problem aims at defining a suitable monitoring campaign that is able to optimally allocate the number of swab tests between the subgroups of the population in order to reduce the number of infected patients (especially the most fragile ones) so reducing the epidemic impact on the health system. The proposed monitoring strategy can be applied both during the most critical phases of the emergency and in endemic conditions, when an active surveillance could be crucial for preventing the contagion rise.
Download
Short Papers
Paper Nr: 27
Title:

Design and Validation of an Adaptive Force Control Algorithm with Parameter Estimation Unit for Electromechanical Feed Axis
Authors:

André Sewohl, Manuel Norberger, Stefan Sigg, Holger Schlegel and Martin Dix
Abstract: Production technology is characterized by the use of electromechanical feed axes, for which the concept of cascade control has become established. The concept is based on linear control engineering. It is not suitable for the control of process forces, which is associated with nonlinearities. Here, adaptive control algorithms from the field of higher control engineering represent a promising approach for improvements of manufacturing strategies and processes in terms of stability, quality, and efficiency. This can also ensure in reducing the number of parts rejected due to bad quality and thus aiding as a significant economic benefit. In this paper, the development of an adaptive control concept that automatically reacts to different and changing environmental conditions during the process is presented. The digital, parameter-adaptive controller consists of a recursive online parameter estimation unit, the controller design procedure, which is based on the setting rule for the symmetric optimum, and the control algorithm. The functionality of the adaptive control concept is demonstrated in simulation and validated by means of experiments on a test setup. It is real-time capable and implemented directly on the machine control together with all calculation algorithms.
Download
Paper Nr: 73
Title:

Electric Power System Operation: A Technique to Modelling, Monitoring and Control via Petri Nets
Authors:

Milton Bastos de Souza, Evangivaldo A. Lima and Jès F. Cerqueira
Abstract: Petri nets have been widely used as a tool to model, monitor and control several kind of systems. In this paper, Petri nets are used to model, monitor and control Electrical Power Systems (EPS). The electric power model will be expanded through a linear transformation. The restrictions imposed for that expansion specialize the new places with attributes that allow to monitor or control the dynamics of the original Petri net.
Download
Paper Nr: 78
Title:

Comparative Study between EKF, SVSF, Combined SVSF-EKF, and ASVSF Approaches based Scale Estimation of Monocular SLAM
Authors:

Elhaouari Kobzili, Ahmed Allam and Cherif Larbes
Abstract: This paper presents a comparative study of scale recovering in monocular simultaneous localization and mapping (Mono-SLAM) by adopting and adapting four estimators into a multi-rate fusion mechanism and considering the scale as an element of the state vector. These estimators are: extended Kalman filter (EKF), smooth variable structure filter (SVSF), combined SVSF-EKF, and particularly adaptive smooth variable structure filter (ASVSF). The use of the ASVSF estimator represents the novelty of this paper because it provides a robust estimation of the trajectory scale as well as the covariance matrix at each iteration. This later represents the estimation incertitude. A second sensor is involved (inertial measurement unit (IMU)) as a reference to align the up to scale trajectory provided by the Mono-SLAM box. The designed system allows finding the scale factor with a rate not further than the IMU frequency and avoids complex synchronization. In order to outline the limitation of each estimator used for scale recovering, a deep analysis of the proposed approaches in terms of robustness, stability, accuracy, and real-time constraint was carried out.
Download
Paper Nr: 93
Title:

Contribution to Robot System Identification: Noise Reduction using a State Observer
Authors:

Bilal Tout, Jason Chevrie, Laurent Vermeiren and Antoine Dequidt
Abstract: Conventional identification approach based on the inverse dynamic identification model using least-squares and direct and inverse dynamic identification techniques has been effectively used to identify inertial and friction parameters of robots. However these methods require a well-tuned filtering of the observation matrix and the measured torque to avoid bias in identification results. Meanwhile, the cutoff frequency of the low-pass filter fc must be well chosen, which is not always easy to do. In this paper, we propose to use a Kalman filter to reduce the noise of the observation matrix and the output torque signal of the PID controller.
Download
Paper Nr: 97
Title:

Adaptive Fault Detection and Isolation for DC Motor Input and Sensors
Authors:

Nikita Kolesnik, Alexey Margun, Artem Kremlev and Andrei Zhivitskii
Abstract: The paper is devoted to the development of an adaptive approach to the fault detection and isolation of input and sensor failures of armature-controlled direct current motors. The proposed detection method is based on the full state Luenberger observer. Isolation scheme uses the directional residual set and relationships between fault directions and residual vector. Adaptability is provided by dynamic regressor extension and mixing approach for online estimation of parameters. Proposed scheme allows to isolate following faults: unaccounted load acting on the rotor, input voltage disturbance, failures of velocity and current sensors. Simulation results confirm performance of the proposed approach.
Download
Paper Nr: 101
Title:

Finite-time Stability Analysis for Nonlinear Descriptor Systems
Authors:

N. Shopa, D. Konovalov, A. Kremlev and K. Zimenko
Abstract: Sufficient conditions of finite-time stability are presented for the class of nonlinear descriptor systems. Both, explicit and implicit Lyapunov function methods, are extended for finite-time stability analysis of descriptor systems and the corresponding settling time estimates are obtained. The theoretical results are supported by numerical examples.
Download
Paper Nr: 51
Title:

Calibration of the Nonlinear Wheel Odometry Model with an Improved Genetic Algorithm Architecture
Authors:

Máté Fazekas, Balázs Németh and Péter Gáspár
Abstract: To guarantee the required motion estimation accuracy for an autonomous vehicle, the integration of the wheel encoder measurements is an adequate choice besides the generally applied GNSS, inertial and visual-odometry methods. Wheel odometry is a robust and cost-effective technique, but the required calibration of the nonlinear odometry model in the presence of noise remains an open problem in the context of autonomous vehicles. The core problem is that due to the nonlinear behavior of the model, the identified parameters will be biased even with Gaussian-type measurement noises. The presented method operates with genetic algorithms and utilizes two novel improvements: compensation of the state initialization of the model inside the estimation process, and equilibration of the parameter estimation by an adaptive weighting technique. With these innovations the distortion effects are mitigated and unbiased model calibration can be obtained even when several local minimums exist. The performance of the developed algorithm and the accuracy of parameter estimation are demonstrated with detailed validation and test with a real vehicle.
Download
Paper Nr: 76
Title:

Power System Operation Modeling, Monitoring and Control using Petri Nets
Authors:

Milton Bastos de Souza, Evangivaldo A. Lima and Jès F. Cerqueira
Abstract: Petri nets have been widely used as a tool for the model of Dynamics Discrete Event System (DDES). In this paper, Petri nets are used to model, monitor and control Electrical Power Operation (EPO). For that, it will be used a linear transformation to expand the original Petri net. The expansion will change the original rules of firing transitions. Its changes impose restrictions on the system’s operation. As result, a simplified equation is presented and it is used for monitoring and controlling EPO.
Download
Paper Nr: 80
Title:

Web based User Interface Solution for Remote Robotic Control and Monitoring Autonomous System
Authors:

Hugo Perier, Eloise Matheson and Mario Di Castro
Abstract: The area of robotic control and monitoring, or automated systems, covers a wide range of applications. The operating system, the kind of control, and the size of the screen used to present information to the user all vary in different robotic or industrial systems. This article proposes a system based on a user interface for real-time robotic control or monitoring of autonomous systems using web technologies laid on open-source projects. The purpose of this software is to be highly scalable over time and easily pluggable into different types of robotic solutions and projects. It must offer a high user experience and an appealing modern UI design, allowing technicians not expert in robot operation to perform interventions or maintenance tasks. The web environment provides an ideal platform to ensure the portability of the application so that it can be released on a multitude of devices, including laptops, smartphones, and tablets. This article introduces and describes the module, features, and advantages of the Neutron Framework. It presents how the users can interact with it and how to integrate this solution inside the CERN’s Mechatronic Robotic and Operation solution.
Download
Paper Nr: 89
Title:

A Geometric Approach for Partial Liquids’ Pouring from a Regular Container by a Robotic Manipulator
Authors:

Jeeangh J. Reyes-Montiel, Antonio Marin-Hernandez and Sergio Hernandez-Mendez
Abstract: Partial liquid pouring is a very useful task in many environments; however, it is still a very challenging task for autonomous mobile robots. In this work, is presented a geometric approach to accurately partial pouring by autonomous robots. While diverse approaches propose to deal with this problem measuring liquid’s volume at destination container, in this work is analyzed the geometry and initial volume of liquid at pouring container, i.e., liquid’s volume and container characteristics are known. Then based on the transversal sections volumes’ is proposed to control pouring. Proposed approach computes the cross-section areas formed by liquid in the container when this is tilted an angle q. The geometric analysis shows that an angle-based linear control does not guarantee a regular flow to perform an accurate liquid control, since cross-sectional volumes have not linear relation with the angle q when tilted. As it is show in this work, these volumes increase and decrease according to the tilted angle and the container characteristics. To effectively obtain a regular flow those volumes should be considered in the control phase as here is proposed.
Download