Hybrid Reinforcement Learning and Its Application to Biped Robot Control

A learning system composed of linear control modules, reinforcement learning modules and selection modules (a hybrid reinforcement learning system) is proposed for the fast learning of real-world control problems. The selection modules choose one appropriate control module dependent on the state. This hybrid learning system was applied to the control of a stilt-type biped robot. It learned the control on a sloped floor more quickly than the usual reinforcement learning because it did not need to learn the control on a flat floor, where the linear control module can control the robot. When it was trained by a 2-step learning (during the first learning step, the selection module was trained by a training procedure controlled only by the linear controller), it learned the control more quickly. The average number of trials (about 50) is so small that the learning system is applicable to real robot control.