Lifelong learning for autonomous robots

3L4AR is a researcher mobility project between the Chronorobotics Lab at the Czech Technical University in Prague (CTU) in Czechia and the CIAD Lab at the University of Technology of Belfort-Montbéliard (UTBM) in France. The project aims to improve acceptance of mobile robots through the integration of lifelong learning methods into the modules that constitute the core of their navigation systems. We argue that the robustness of the navigation systems in dynamic environments and changing conditions can be improved during the robot routine operation through machine learning methods. The incorporation of lifelong learning would allow the robots to gradually improve their ability to perceive, understand and exploit the environmental dynamics caused by natural processes and human activities. Understanding of the environment dynamics will allow the robots to anticipate and to react more intelligently to the everyday situations, which will improve their acceptance by humans.