F. Mondada - Academia.edu (original) (raw)

Papers by F. Mondada

IEEE Access, 2021

This work involved human subjects in its research. The authors confirm that all human subject res... more This work involved human subjects in its research. The authors confirm that all human subject research procedures and protocols were exempt from review board approval.

Advances in the Evolutionary Synthesis of Intelligent Agents, 2001

We investigate the dynamics of competitive co-evolution in the framework of two miniature mobile ... more We investigate the dynamics of competitive co-evolution in the framework of two miniature mobile robots, a predator with a vision system and a faster prey with proximity sensors. Both types of robots are controlled by evolutionary neural networks. A variety of efficient chase-escape behaviors emerge in few generations. These results are analyzed in terms of variable fitness landscapes and selection criteria. A new vision of artificial evolution as generation and maintainance of adaptivity is suggested and contrasted with the theory and practice of mainstream evolutionary computation. In a second stage, different types of ontogenetic changes applied to the robot controllers are compared and the results are analyzed in the context of competitive co-evolution. It is shown that predators benefit from forms of directional changes whereas prey attempt to exploit unpredictable behaviors. These results and their effect on coevolutionary dynamics are then considered in relation to open-ended evolution in unpredictably changing environments.

Swarm robotics in real world requires a large number of robots and thus enough room for experimen... more Swarm robotics in real world requires a large number of robots and thus enough room for experimentation. Therefore, to implement such experiments with limited budget, robots should be compact and low cost, which entails the use of microcontroller-based miniature robots. In this context, developing behaviour is challenging, because microcontrollers are not powerful enough to support common high-level development environments such as Java. Furthermore, the development tools for microcontrollers are not able to monitor and debug groups of robots online. In this paper, we present a new event-based control architecture: aseba. It solves the problem of developing and testing collective behaviours by running script inside a lightweight virtual machine on each microcontroller and by providing an integrated development environment to program and monitor the whole group of robots from a single application running on any desktop computer. We have validated aseba by implementing a dangerous-area avoidance experiment using the epuck robot. Experiments of this type are common in swarm robotics, but porting them to real robots is often challenging. By easing the development of complex behaviours on real robots, aseba both exposes collective robotics programming to a large community and opens new research perspectives for swarm robotics.

Mobile robots have the potential to become the ideal tool to teach a broad range of engineering d... more Mobile robots have the potential to become the ideal tool to teach a broad range of engineering disciplines. Indeed, mobile robots are getting increasingly complex and accessible. They embed elements from diverse fields such as mechanics, digital electronics, automatic control, signal processing, embedded programming, and energy management. Moreover, they are attractive for students which increases their motivation to learn. However, the requirements of an effective education tool bring new constraints to robotics. This article presents the e-puck robot design, which specifically targets engineering education at university level. Thanks to its particular design, the e-puck can be used in a large spectrum of teaching activities, not strictly related to robotics. Through a systematic evaluation by the students, we show that the epuck fits this purpose and is appreciated by 90 percent of a large sample of students.

2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2010

To design, to study, and to control mixed animalsrobots societies is a challenging field of scien... more To design, to study, and to control mixed animalsrobots societies is a challenging field of scientific exploration that can bring new frameworks to study individual and collective behaviors in animal and mixed robot-animal societies. In the Chicken Robot project we aim at developing a mobile robot, able to collaborate with a group of chicks and to control certain group behaviors. The first research step is to build formal models of relevant animal behaviors by performing ethological experiments. Hence, one of the principal tasks is to design a setup equipped with appropriate monitoring tools. In this paper, we present a toolset for running chick-robot experiments and analyzing results. It includes an autonomous PoulBot robot and an experimental setup, able to autonomously record experimental video and audio data, to detect displacements of chicks and robots, to detect their calling activity and to provide robots with these data. We also present a visual data analysis system to extract behavioral features of individual chicks using the variational Bayesian Gaussian mixture model classification with a particle filters based prediction of future positions of chicks. We show how these tools are currently used to carry out chick-robot experiments, to collect behavioral data and to extract animal behavioral features that allow us to build behavioral models bound to be implemented in the robot.

Lecture Notes in Computer Science, 2014

The Thymio II robot was designed to be used by teachers in their classrooms for a wide range of a... more The Thymio II robot was designed to be used by teachers in their classrooms for a wide range of activities and at all levels of the curriculum, from very young children to the end of high school. Although the educationally oriented design of this innovative robot was successful and made it possible to distribute more than 800 Thymio robots in schools with a large majority in the French-speaking part of Switzerland, it was not sufficient to significantly raise the number of teachers using robot technology in their teaching after three years of commercialization. After an introduction and a first section on the design of this educational robot, this paper presents some results of a sociological analysis of the benefits and blockages identified by teachers in using robots, or not, with their pupils.

2014 IEEE Eighth International Conference on Self-Adaptive and Self-Organizing Systems Workshops, 2014

IEEE Transactions on Systems, Man and Cybernetics, Part B (Cybernetics), 1996

Abstract| In this paper we describe the evolution of a discrete-time recurrent neural network to ... more Abstract| In this paper we describe the evolution of a discrete-time recurrent neural network to control a real mobile robot. In all our experiments the evolutionary procedure is carried out entirely on the physical robot without human intervention. We s h o w that the autonomous development of a set of behaviors for locating a battery charger and periodically returning to it can be achieved by lifting constraints in the design of the robot/environment i n teractions that were employed in a preliminary experiment. The emergent homing behavior is based on the autonomous development o f a n i n ternal neural topographic map (which i s not pre-designed) that allows the robot to choose the appropriate trajectory as function of location and remaining energy.

ABSTRACT The project "Hands-On Mechatronics" aims to develop a problem-based le... more ABSTRACT The project "Hands-On Mechatronics" aims to develop a problem-based learning environment for mechatronics. This environment is based on four specific aspects: a mobile robot competition as motivation factor, a WEB environment as framework for all student, the SmartEASE robot as experimental platform and a good software environment for design and fast prototyping. The actual status of this concept is applied at the Swiss Federal Institute of Technology in Lausanne (EPFL) with participants coming from several schools of different levels. We present the actual implementation of the concept "Hands-On Mechatronics", the hardware and software tools used and some problems found in our approach.

The MagneBike robot is a magnetic wheeled robot designed for the inspection of ferromagnetic stru... more The MagneBike robot is a magnetic wheeled robot designed for the inspection of ferromagnetic structures in power plants, especially steam chests. This video first presents the robot's locomotion concept, i.e. two aligned magnetic wheels integrating lateral lever arms, that allow the robot to pass over complex combinations of obstacles. Laboratory and field experiments show the high mobility of the robot. This video also describes the localization and mapping strategy that consists in combining 3D odometry with 3D scanning and scan registration. An animation of the 3D reconstruction of the environment shows that the localization procedure allows to provide the necessary 3D visual feedback for the remote user or for inspection mission planning.

Lecture Notes in Computer Science, 2014

IEEE Access, 2021

This work involved human subjects in its research. The authors confirm that all human subject res... more This work involved human subjects in its research. The authors confirm that all human subject research procedures and protocols were exempt from review board approval.

Advances in the Evolutionary Synthesis of Intelligent Agents, 2001

We investigate the dynamics of competitive co-evolution in the framework of two miniature mobile ... more We investigate the dynamics of competitive co-evolution in the framework of two miniature mobile robots, a predator with a vision system and a faster prey with proximity sensors. Both types of robots are controlled by evolutionary neural networks. A variety of efficient chase-escape behaviors emerge in few generations. These results are analyzed in terms of variable fitness landscapes and selection criteria. A new vision of artificial evolution as generation and maintainance of adaptivity is suggested and contrasted with the theory and practice of mainstream evolutionary computation. In a second stage, different types of ontogenetic changes applied to the robot controllers are compared and the results are analyzed in the context of competitive co-evolution. It is shown that predators benefit from forms of directional changes whereas prey attempt to exploit unpredictable behaviors. These results and their effect on coevolutionary dynamics are then considered in relation to open-ended evolution in unpredictably changing environments.

Swarm robotics in real world requires a large number of robots and thus enough room for experimen... more Swarm robotics in real world requires a large number of robots and thus enough room for experimentation. Therefore, to implement such experiments with limited budget, robots should be compact and low cost, which entails the use of microcontroller-based miniature robots. In this context, developing behaviour is challenging, because microcontrollers are not powerful enough to support common high-level development environments such as Java. Furthermore, the development tools for microcontrollers are not able to monitor and debug groups of robots online. In this paper, we present a new event-based control architecture: aseba. It solves the problem of developing and testing collective behaviours by running script inside a lightweight virtual machine on each microcontroller and by providing an integrated development environment to program and monitor the whole group of robots from a single application running on any desktop computer. We have validated aseba by implementing a dangerous-area avoidance experiment using the epuck robot. Experiments of this type are common in swarm robotics, but porting them to real robots is often challenging. By easing the development of complex behaviours on real robots, aseba both exposes collective robotics programming to a large community and opens new research perspectives for swarm robotics.

Mobile robots have the potential to become the ideal tool to teach a broad range of engineering d... more Mobile robots have the potential to become the ideal tool to teach a broad range of engineering disciplines. Indeed, mobile robots are getting increasingly complex and accessible. They embed elements from diverse fields such as mechanics, digital electronics, automatic control, signal processing, embedded programming, and energy management. Moreover, they are attractive for students which increases their motivation to learn. However, the requirements of an effective education tool bring new constraints to robotics. This article presents the e-puck robot design, which specifically targets engineering education at university level. Thanks to its particular design, the e-puck can be used in a large spectrum of teaching activities, not strictly related to robotics. Through a systematic evaluation by the students, we show that the epuck fits this purpose and is appreciated by 90 percent of a large sample of students.

2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2010

To design, to study, and to control mixed animalsrobots societies is a challenging field of scien... more To design, to study, and to control mixed animalsrobots societies is a challenging field of scientific exploration that can bring new frameworks to study individual and collective behaviors in animal and mixed robot-animal societies. In the Chicken Robot project we aim at developing a mobile robot, able to collaborate with a group of chicks and to control certain group behaviors. The first research step is to build formal models of relevant animal behaviors by performing ethological experiments. Hence, one of the principal tasks is to design a setup equipped with appropriate monitoring tools. In this paper, we present a toolset for running chick-robot experiments and analyzing results. It includes an autonomous PoulBot robot and an experimental setup, able to autonomously record experimental video and audio data, to detect displacements of chicks and robots, to detect their calling activity and to provide robots with these data. We also present a visual data analysis system to extract behavioral features of individual chicks using the variational Bayesian Gaussian mixture model classification with a particle filters based prediction of future positions of chicks. We show how these tools are currently used to carry out chick-robot experiments, to collect behavioral data and to extract animal behavioral features that allow us to build behavioral models bound to be implemented in the robot.

Lecture Notes in Computer Science, 2014

The Thymio II robot was designed to be used by teachers in their classrooms for a wide range of a... more The Thymio II robot was designed to be used by teachers in their classrooms for a wide range of activities and at all levels of the curriculum, from very young children to the end of high school. Although the educationally oriented design of this innovative robot was successful and made it possible to distribute more than 800 Thymio robots in schools with a large majority in the French-speaking part of Switzerland, it was not sufficient to significantly raise the number of teachers using robot technology in their teaching after three years of commercialization. After an introduction and a first section on the design of this educational robot, this paper presents some results of a sociological analysis of the benefits and blockages identified by teachers in using robots, or not, with their pupils.

2014 IEEE Eighth International Conference on Self-Adaptive and Self-Organizing Systems Workshops, 2014

IEEE Transactions on Systems, Man and Cybernetics, Part B (Cybernetics), 1996

Abstract| In this paper we describe the evolution of a discrete-time recurrent neural network to ... more Abstract| In this paper we describe the evolution of a discrete-time recurrent neural network to control a real mobile robot. In all our experiments the evolutionary procedure is carried out entirely on the physical robot without human intervention. We s h o w that the autonomous development of a set of behaviors for locating a battery charger and periodically returning to it can be achieved by lifting constraints in the design of the robot/environment i n teractions that were employed in a preliminary experiment. The emergent homing behavior is based on the autonomous development o f a n i n ternal neural topographic map (which i s not pre-designed) that allows the robot to choose the appropriate trajectory as function of location and remaining energy.

ABSTRACT The project "Hands-On Mechatronics" aims to develop a problem-based le... more ABSTRACT The project "Hands-On Mechatronics" aims to develop a problem-based learning environment for mechatronics. This environment is based on four specific aspects: a mobile robot competition as motivation factor, a WEB environment as framework for all student, the SmartEASE robot as experimental platform and a good software environment for design and fast prototyping. The actual status of this concept is applied at the Swiss Federal Institute of Technology in Lausanne (EPFL) with participants coming from several schools of different levels. We present the actual implementation of the concept "Hands-On Mechatronics", the hardware and software tools used and some problems found in our approach.

The MagneBike robot is a magnetic wheeled robot designed for the inspection of ferromagnetic stru... more The MagneBike robot is a magnetic wheeled robot designed for the inspection of ferromagnetic structures in power plants, especially steam chests. This video first presents the robot's locomotion concept, i.e. two aligned magnetic wheels integrating lateral lever arms, that allow the robot to pass over complex combinations of obstacles. Laboratory and field experiments show the high mobility of the robot. This video also describes the localization and mapping strategy that consists in combining 3D odometry with 3D scanning and scan registration. An animation of the 3D reconstruction of the environment shows that the localization procedure allows to provide the necessary 3D visual feedback for the remote user or for inspection mission planning.

Lecture Notes in Computer Science, 2014