基于模糊控制的仿人机器人步态稳定性研究

The Study of Humanoid Robot Gait Stability Based on Fuzzy Control Method

摘要: 仿人机器人在步行过程中会遇到不可预见的干扰而导致机器人不能稳定的行走。因此，本文提出一种模糊控制的方法调整机器人走路姿态以增加步行的抗干扰性。该方法以MF (Metal Fight)仿人机器人为对象，用参考轨迹的方法对机器人进行步态规划，得到保证机器人稳定的步行关节参数；针对步行过程中出现的干扰使机器人稳定裕度降低的情况，先根据干扰对机器人稳定裕度的影响程度进行分级，然后通过模糊逻辑调整机器人双足的关节参数，最终使稳定裕度回归期望值，从而实现对步态稳定性的控制。实验结果表明该方法能有效的克服环境中未知干扰对机器人步行的影响。

Abstract: The unforeseen disturbance will cause the instability during the process of humanoid robot’s walking. So, a gait stability control method based on fuzzy control is proposed, and the purpose is to enhance the anti-interference performance of humanoid robot. Implemented in MF (Metal Fight) humanoid robot, the proposed method accomplishes the gait planning along the reference trajectory and through the method, we obtain the walking joint parameters which are used to control the humanoid robot. In more detail, firstly, to classify ranks for the disturbances according to the disturbance severity which may reduce the stability margin of the robot. Secondly, the joint parameters can be adjusted through fuzzy logic control. Thirdly, to control the gait stability by adjusting the stability margin within the excepted value. The experiment results show that this method can effectively get over the unknown environment impact of interference on the robot.