Terrain classification and adaptive locomotion for a hexapod robot Qingzhui

Abstract Legged robots have potential advantages in mobility compared with wheeled outdoor environments. The knowledge of various ground properties and adaptive locomotion based on different surface materials plays an important role improving the stability legged robots. A terrain classification method for a hexapod robot named Qingzhui is proposed this paper. First, force-based suggested. Ground contact force calculated by collecting joint torques inertial measurement unit information. substrates are classified feature vector extracted from collected data using support machine algorithm. Then, proposed. dynamic alternating tripod trotting gait developed to control robot, parameters active compliance change terrain. Finally, integrated tested real experiments. Our shown effective walk concrete, wood, grass, foam. strategies experimental results can be valuable reference other applied