Robot Learning by Demonstration

In this report, two systems have been developed for robot behavior acquisition using kinesthetic demonstrations. The first enables a humanoid robot to imitate constrained reaching gestures directed towards a target using a learning algorithm based on Gaussian Mixture Regression. The imitation trajectory can be reshaped in order to satisfy the constraints of the task and it can adapt to changes in the initial conditions and to target displacements occurring during the movement execution. The second is focused on behavior learning and walk-gait optimization by simulation using Swarm Intelligence. The fitness of each swarm particle is evaluated using a simulator until the expected behavior is reproduced and then tested on the real robot. The potential of these methods is evaluated using experiments involving Aldebaran’s Nao humanoid robot and Fawkes, an open source robot software by the KBSG at RWTH University.