面向未知地形的四足机器人足端轨迹优化

Foot Trajectory Optimization for Quadruped Robot Walking on Unknown Terrain

在未知地形行走时,由于地形突变,采用常见步态算法的四足机器人容易受到冲击,导致失稳,为此,提出一种改进的四足机器人足端轨迹规划的算法。将足端运动轨迹分段优化,减小机器人在水平地面、上坡和下坡地形的足端力矩变化,增加四足机器人运动的稳定性。实验选用斯坦福四足机器人,记录俯仰角pitch、横滚角roll及足端力矩在机器人通过不同地形时的变化。实验结果表明,足端轨迹优化后的机器人在未知地形中行走的稳定性得到有效提升。

Quadruped robots are susceptible to suffer from impulse and lose balance in unknown terrain using common feet trajectory.To deal with this problem,a method for planning feet trajectory is proposed.Feet trajectory is optimized by means of segmentation to reduce the range of feet moment on horizontal,uphill and downhill terrain and increases the stability of quadruped robot in motion.Experiments were implemented on Stanford robot.The pitch angle,roll angle and feet moment of robot going through different terrains were recorded and compared.The results show that the stability of quadruped robot on unknown terrain is gain improvement after the feet trajectory was optimized.