Combined Reinforcement Learning and CPG Algorithm to Generate Terrain-Adaptive Gait of Hexapod Robots

Terrain adaptation research can significantly improve the motion performance of hexapod robots. In this paper, we propose a method that combines reinforcement learning with central pattern generator (CPG) to enhance terrain robots in terms gait planning. The robot’s complex task presents high-dimensional observation and action space, which makes it challenging directly apply robot control. Therefore, utilize CPG algorithm generate rhythmic while compressing space dimension agent. Additionally, proposed requires less internal sensor information, exhibits strong applicability. Finally, conduct experiments deploy framework simulation environment. results show policy trained our enables move more smoothly efficiently on rugged compared traditional method.