COBOTA 2025 Abstracts

Area 1 - Bridging the Gap in COllaborative roBOtics: from Theory to real Applications

Full Papers

Paper Nr:	8
Title:	Preliminary Design and Control of an Operator-Assistance System Integrated into a Cobot, for Anatomical Meat-Cutting Process
Authors:	Alexis Babut, Chedli Bouzgarrou, Laurent Sabourin and Nicolas Bouton
Abstract:	This paper presents the preliminary design and control of a collaborative robotic cell for operator assistance in tasks involving soft material manipulation, such as meat cutting. The system integrates force/torque sensors and employs a Cartesian admittance controller to enable compliant, intuitive physical interaction. The mechanical design of the end-effector, the control architecture, and the communication strategy are described. Initial experiments validate the system’s ability to provide stable and responsive assistance in a physical Hu-man–Robot Interaction (pHRI) context.
Download

Paper Nr:	13
Title:	Design and Validation of Sensorized Tools for Deformable Object Manipulation in Meat Cutting and Doll Demoulding
Authors:	Saltanat Seitzhan, Dionisio Cartagena González, Alexis Babut, Daniel Sánchez-Martínez, Juan Antonio Micó, Chedli Bouzgarrou and Juan Antonio Corrales Ramón
Abstract:	While robotic arms are extensively deployed in mass production environments, their application in tasks involving deformable object manipulation remains limited due to the complex dynamics of soft materials. Addressing this challenge requires task-specific end-effector tools capable of replicating manual operations with precision and adaptability. Standard human tools are often incompatible with robotic systems, especially in domains such as meat processing and doll manufacturing. This study presents the design and experimental validation of sensor-integrated end-effectors tailored for deformable object handling: a knife tool for roboticassisted meat cutting and a plier tool for analyzing the demoulding process in doll production. Both tools incorporate multimodal sensing, including force/torque sensors and inertial measurement units, and are synchronized via ROS to capture manipulation data under realistic conditions. While full cobotic manipulation and Learning from Demonstration (LfD) are reserved for future work, the results demonstrate the feasibility of embedding sensing into manual and robotic tools to support future automation in soft material handling.
Download

Short Papers

Paper Nr:	9
Title:	Soft Robotics for Advanced Handling of Delicate Fruit Products
Authors:	Luan Lang, Rodrigo Antunes, Martim Lima de Aguiar, Nuno Pereira, Thiago Assis Dutra, Yebo Lu and Pedro Dinis Gaspar
Abstract:	Soft robotic grippers can be an effective tool for handling sensitive and irregularly shaped objects, such as horticultural products. This study evaluates three soft gripper designs-Straight, Constant Curve, and Beak-fabricated using Thermoplastic Polyurethane (TPU) with shore hardness 60A and 95A. The grippers were produced using a 3D printer and tested on a universal testing machine to assess mechanical performance. Practical tests revealed that the Beak gripper made with TPU 60A exhibited superior performance, achieving a peak force of 10.59 N at a displacement of 21.65 mm, making it suitable for delicate tasks like handling fruits, without causing damage. In contrast, grippers made with TPU 95A, while possessing higher force capacities, were excessively rigid and risked damaging delicate items. The study shows the importance of material selection and gripper design in optimizing performance for specific applications. The findings validate simulation data and indicate that TPU 60A is more appropriate for applications requiring gentle handling. Future work includes testing with objects of varying shapes, conducting fatigue tests, and exploring multi-material gripper designs with embedded sensing capabilities to enhance adaptability and control during use.
Download

Paper Nr:	10
Title:	Automated Computational Workflow for the Parametric Design and Optimization of a 3D-Printed Fin Ray Effect Soft Robotic Finger
Authors:	Rodrigo Antunes, Luan Lang, Martim Lima de Aguiar, Nuno José Matos Pereira, Thiago Assis Dutra, Yebo Lu and Pedro Dinis Gaspar
Abstract:	The design of soft grippers is challenged by the complex, non-linear coupling of material properties, geometry, and control, rendering traditional design methods inefficient. To address this, this paper presents an automated computational workflow for the parametric design and optimization of a 3D-printed Fin Ray Effect soft robotic gripper finger. The tool establishes a closed-loop digital thread, connecting a web-based parametric design interface using FreeCAD to a finite element analysis backend driven by PyAnsys. A parametric study was conducted, varying the number of internal crossbeams from 1 to 16, to analyse the gripper's performance using an experimentally validated hyperelastic model for TPU 60A. The results show a trade-off between contact pressure and pressure distribution, with an optimal configuration of 14-16 crossbeams identified for applications requiring a gentle grip with low-pressure concentrations. The developed workflow proved to be an effective method for rapidly iterating through designs and identifying an optimal solution, showcasing the power of automated simulation in the Design for Additive Manufacturing (DfAM) process.
Download

Paper Nr:	11
Title:	Differential Kinematics Control Using Circles as Bivectors of Conformal Geometric Algebra
Authors:	Julio Zamora-Esquivel, Alberto Jaimes Pita, Edgar Macias-Garcia, Javier Felip-Leon, David Gonzalez-Aguirre and Eduardo Bayro-Corrochano
Abstract:	In this paper, we propose a modern mathematical framework to model a robotic arm and compute the differential kinematics of its end effector, which is represented as a circle in a three-dimensional space. This circle is described using a bi-vector within the context of conformal geometric algebra. By utilizing a circle to characterize the grasping pose on the object and the pose of the end-effector, we develop a differential kinematics-based control law that guides the end-effector to minimize the error between both circles. The circle representation offers three degrees of freedom for the center, two degrees for orientation, and one degree for the radius, allowing us to effectively describe the end-effector pose using a single geometric primitive. Our approach allows for simultaneous adjustment of both the position and orientation of the end effector.
Download