Improving biped walk stability with complementary corrective demonstration

We contribute a method for complementing an existing algorithmic solution to performing a low level skill with corrective human demonstration to improve skill execution performance. We apply the proposed method to biped walking problem, which is a good example of a complex low level skill due to the complicated dynamics of the walk process in a high dimensional state and action space. We introduce an incremental learning approach to improve the Nao humanoid robot’s stability during walking. First, we identify, extract, and record a complete walk cycle from the motion of the robot while it is executing a given walk algorithm as a black box. Second, we apply offline advice operators for improving the stability of the learned openloop walk cycle. Finally, we present an algorithm to directly modify the recorded walk cycle using real time corrective human demonstration. The demonstration is delivered using a commercially available wireless game controller without touching the robot. Through the proposed algorithm, the robot learns a closed-loop correction policy for the openloop walk by mapping the received corrective demonstraThe first author is supported by The Scientific and Technological Research Council of Turkey Programme 2214 and the Turkish State Planning Organization (DPT) under the TAM Project, number 2007K120610. Ç. Meriçli( ) Computer Science Dept., Dept. of Computer Engineering, Carnegie Mellon University, USA Boğaziçi University, Turkey E-mail: [email protected] M. Veloso Computer Science Department, Carnegie Mellon University, USA E-mail: [email protected] H. L. Akın Department of Computer Engineering Boğaziçi University, Turkey E-mail: [email protected] tion to the sensory readings while walking autonomously using the refined open-loop walk cycle. Experiment results demonstrate a significant improvement in the walk stability.