Research On Control Method Of Inverted Balance And Leap Between Bars Of Underactuated Gymnastic Robot

A gymnastics robot is a typical underactuated system in which the number of its own drivers is less than the number of degrees of freedom.This system has complex strong coupling internal characteristics,which leads to non-holonomic constraints that are difficult to perform feedback linearization.This system is an ideal model for verifying nonlinear control methods.Underactuated systems are widely used in production and life,such as space robots,mobile robots,quadrotors,and so on.This paper mainly studies the inverted balance of the gymnastics robot and the leap between horizontal bars.The inverted balance is that the gymnastic robot swings up to the vertical upward unstable balance point and maintains balance;the bar changing motion is the process of the robot leaping between two parallel horizontal bars.The gymnastic robot swings on one of the horizontal bars to the appropriate initial state.After that,the hand is released,and after the posture change in the air,another horizontal bar parallel to it is grasped in an appropriate posture.Focusing on the two main actions of the gymnastics robot,the main work of this paper is as follows:Firstly,the movement of the gymnastic robot can be divided into two parts movement when the gripper grabs the bar and movement when the gripper is off the bar,and the two movements are mathematically modeled.Among them,the Lagrangian equation is used to simulate the dynamics of the grab bar motion.In the process of the leaping movement,the gymnastic robot is not subjected to the external moment in space,and the angular momentum conservation equation is derived.The above model was verified in MATLAB/Simulink.Secondly,the segmentation swing-up control strategy of the gymnastic robot with partial feedback linearization and Lyapunov controller is proposed.The inverted balance controller based on LQR algorithm is designed,and the motion planning of the shifting process is carried out.The segmentation controller utilizes partial feedback linearization in the initial stage to make the mechanical energy of the robot accumulate rapidly.When the threshold value is reached,it switches to the Lyapunov controller,so that the gymnastic robot obtains the desired state required for inverted balance and bar-leaping control.On this basis,the inverted balance control of the gymnastics robot is realized by LQR algorithm.At the same time,the initial conditions of the bar-changing action were obtained so that the gymnastics robot's leap motion was realized.The algorithm was jointly simulated by MATLAB/Adams to verify the control and planning effects.Thirdly,the ideal dynamic equation of the gymnastic robot is modified.The friction term is added and the parameter identification is carried out by the least squares method.This method improves the parameter accuracy.Finally,the software framework of the developed gymnastics robot was built.Apply the aforementioned control method to the physical platform.The experimental results show that the designed control method can achieve swing and inverted balance and has better robustness.