Research On Leg Compliance Control And Pose Optimization Of A Hexapod Robot

With the development of human society,the scope of people’s activities is more and more extensive.In some complex or dangerous environments,the difficulty of operation is also higher and higher.It is impossible to meet the requirements of operation only by relying on human.Legged robots have strong adaptability to complex operating environments,especially hexapod robots,which play an irreplaceable role in complex operating environments.Therefore,research on hexapod robots is of great significance.In order to improve the motion stability of hexapod robot in complex environment,this paper focuses on the study of leg flexibility control and pose optimization of hexapod robot.Based on the biomimetic configuration of heavy-duty hexapod robot developed by our research group,the kinematics model of hexapod robot is established,and the coordinate system is established based on the D-H parameter method to obtain the leg kinematic equation of hexapod robot,and the kinematic equation of the hexapod robot torso is derived based on the kinematic analysis method of parallel mechanism.On this basis,the influence of the hexapod robot’s initial pose on the driving force of the hexapod robot’s trunk workspace and leg hydraulic system is analyzed.The multiobjective optimization of the initial pose of the hexapod robot is carried out by using genetic algorithm,aiming at the maximum volume of the trunk workspace and the minimum driving force of the leg hydraulic system.Finally,the optimal initial pose of the hexapod robot is obtained according to the optimization results.When hexapod robots walk in complex environment,due to the unknown environment,the legs of hexapod robots are easy to have unexpected contact or collision with the terrain environment,resulting in damage or even destruction of the hexapod robot itself.In order to solve the above problems,this paper adopts positionbased impedance control as the basic control strategy to achieve leg compliance control.In order to reduce the steady-state errors of force tracking under various expected forces of impedance control strategy,an adaptive compensation algorithm is introduced on the basis of impedance control strategy to achieve force tracking under various expected forces of foot.The stability and steady-state errors of the adaptive compensation algorithm were analyzed under various expected force signals.Finally,in order to improve the adaptive ability of the adaptive compensation algorithm to different terrain environments,fuzzy algorithm was used to adjust the impedance control parameters.The pose of hexapod robot directly affects the motion performance of hexapod robot,so the choice of pose is of great significance for the movement of hexapod robot in complex environment.Based on the above reasons,this paper completed the position and pose optimization of hexapod robot under slope terrain.First,the terrain perception strategy is adopted,that is,based on the biological ontology perception principle and combined with the IMU,displacement sensor,force sensor and kinematic relationship of the hexapod robot,on the premise of external environment sensing devices such as visual sensors and lidar,the current position of the hexapod robot is estimated.Then based on the perceived terrain,combined with the constraints of body safety size,the limit position of driving hydraulic cylinder,and the stability of slope terrain,the numerical solutions of walking energy consumption under each body pose were obtained.The body pose state under the minimum energy consumption value was taken as the body pose under the current terrain environment.In order to verify the validity and correctness of the above theories a simulation platform was built and a prototype experiment of a hexapod robot was carried out.The effectiveness of the leg compliance control strategy of a hexapod robot was verified by simulation and experiments under different "stiffness" terrain environments.The walking simulation and experiment of a hexapod robot under different slope environment and different trunk pose were completed to verify the correctness of pose optimization.