This project aims to develop a new methodology for tracking people by mobile robots based on AI.The learning and repetition of trajectories in the field of robotics have developed over the ten last years. This enables wider adoption of robots by society and facilitates human-robot collaboraon.
Learning and repeating trajectories is one of the most relevant abilities for robots autonomous. It is the capacity of a robot to follow autonomously (repetition phase) a trajectory which was previously followed by another or the same robot but under human supervision (learning phase). Potential applications include, for example, surveillance robots, which must follow trajectories predefined with little variation.
The goal of this project is to develop algorithms to control robots in phase learn in a more natural way for humans. More precisely, the human will walk by following a path that the robot will have to repeat later. During the teaching phase during which the human will walk, the robot will follow its movement in order to estimate and learn its trajectory.