Research On Gait Planning Of Hexapod Robot And Impedance Control In Uncertain Environment

The multi-legged robot is a very representative bionic robot in recent years.It is a foot-type robot with locally connected legs and globally parallel legs.The robot has the characteristics of overall high symmetry and multiple degrees of freedom of the leg joints.It can complete a variety of gait walking switching and stable walking movements,so it has flexible and varied movement methods and good environmental adaptability.In response to this goal,in the traditional study of multi-legged robots,this article adds a vacuum suction device to the foot end of the multi-legged robot,so that the robot can achieve a transitional gait for climbing,and also for the movement under uncertainty The control has been studied so that the robot can better adapt to the environment in unexpected situations.Firstly,we set out from the research background of this subjec.we introduced the domestic and foreign research status of multi-legged robots and some major application scenarios,described the development history of multi-legged robots in detail,and outlined the motion planning and control methods.It provided ideas for our following research.Then,we used the D-H method to model the kinematics of the hexapod robot experimental platform,and based on this,we gave the inverse kinematics solving equations.Then we conducted a detailed analysis of the gait planning of the hexapod robot,introduced some conventional gait planning,and introduced an introduction to wall climbing gait,so that the robot has the ability to complete more complex tasks.Finally,in order to enhance the robot's ability to adapt to unknown disturbances during walking,we studied its motion control and first introduced the theory of dynamic movement primitives.This is a very widely used trajectory generation method in theory.In theory,It can generate trajectories of almost any shape,and then on the basis of reinforcement learning,we adopted a parameter update strategy based on path integration,and performed trajectory learning under the condition of applying a random force field to the robot legs,and then,the experimental results were verified through simulation experiments.Under the interference of different force fields,the trajectory learning curve of the robot's single leg was obtained.The results showed that the robot finally maintained the movement trend of the desired trajectory,which made the robot have a certain anti-interference ability.