载人六足机器人驾驶决策

This paper proposes an algorithm to improve the smooth passing ability of the manned hexapod robot in complex terrain such as transition from slope to plane and obstacle terrain. The algorithm was established based on a discrete foot-end model by using the leg configuration of the hexapod robot, and then provided real-time stability evaluation index to make effective driving decisions by analyzing stable movement space at the foot. For the transition from slope to plane, a control algorithm based on cooperative game which could feedback the optimal foot and body motion sequence to the driver in the form of command information was designed. For obstacle terrain, a control algorithm based on Bayesian decision was established to adjust the body position according to the foot state, and helped the driver make the best decision. In addition, the algorithm introduced weighting factors to control the adaptive switching of two algorithms in different terrains. The weighting factors worked by adjusting the kinematics margin and the stability margin based on the information fed back by the sensor.The simulation experiments show that the algorithm assists the driver to make driving decisions effectively. For complex terrain, the stability of the hexapod robot was improved, body tilt and foot impact were reduced, and the handling efficiency was promoted.