Statistical Learning by Imitation of Competing Constraints in Joint Space and Task Space

Sylvain Calinon, Aude Billard
2009 Advanced Robotics  
We present a probabilistic architecture for solving generically the problem of extracting the task constraints through a Programming by Demonstration (PbD) framework and for generalizing the acquired knowledge to various situations. In previous work, we proposed an approach based on Gaussian Mixture Regression (GMR) to find a controller for the robot reproducing the statistical characteristics of a movement in joint space and in task space through Lagrange optimization. In this paper, we
more » ... an alternative procedure to handle simultaneously constraints in joint space and in task space by combining directly the probabilistic representation of the task constraints with a solution to Jacobian-based inverse kinematics. The method is validated in manipulation tasks with two 5 DOFs Katana robotic arms displacing a set of objects. Keywords: Robot programming by demonstration, learning by imitation, kinesthetic teaching, Gaussian mixture regression, inverse kinematics [1]; and (2) a geometric inverse kinematics approach for a 4 DOFs humanoid arm by representing the
doi:10.1163/016918609x12529294461843 fatcat:ktodcepvd5bthe6giizkriie2e