The Research On Balance Control Of Cube Robot Based On Inertial Wheel Group

The self-balance control problem is a classical problem in nonlinear dynamic system control and it has a long history of research.As a hot research topic in the field of control,self-balance device is recognized as an ideal typical experimental equipment in automatic control theory.At present,the research on balance control mainly focuses on linear inverted pendulum or uniaxial inertial wheel inverted pendulum.There is little research on three-dimensional spatial inverted pendulum,especially the research on its control method is still in the primary stage,which needs to be further explored and studied.Cube robot is a kind of three-dimensional inverted pendulum based on inertial wheel group.Based on the reaction moment produced by inertial wheel group,it can be balanced upright on the edge and finally on the corner.This paper takes the cube robot as the research object,establishes the dynamics model of the cube robot system and designs the control algorithm of the system.Then,with the help of the LabVIEW software and the NI myRIO controller,it carries on the exploratory research to build,debug and realize the balance function of the experimental platform.First,this paper introduces the mechanical structure of the cube robot.The edge balance control of cube robot is abstracted into one-sided inertial wheel inverted pendulum control problem,and then the state equation of inertial wheel inverted pendulum is established.According to the Lagrange equation,this paper establishes the dynamic model of the cube robot system and verifies the effectiveness of the system model by MATLAB simulation.Then the stability and controllability of the system are analyzed.The result shows that the cube robot is a fully controllable unstable system.Secondly,the optimized LQR controller based on the catastrophic genetic algorithm is designed to complete the edge balance control of the cube robot,and the numerical simulation results in the MATLAB/Simulink environment show that the control performance of the LQR controller is improved after the optimization of the catastrophic genetic algorithm.`Thirdly,the adaptive integral backstepping control method is designed to control the corner balance of the cube robot,and the virtual control quantity is introduced to solve the control decoupling problem of this multivariable system.Through the joint simulation of the cube robot system by Adams and MATLAB/Simulink,the simulation result verifies the effectiveness of the designed control algorithm for the corner balance control of the cube robot system.Finally,with regard to the problem of acquiring real-time attitude information of cube robot,a complementary filtering algorithm based on quaternion is designed.By means of NI myRIO and corresponding sensors,the designed attitude fusion algorithm is coded in Lab VIEW software to complete the acquisition of attitude data of cube robot.After building the experimental platform of cube robot system and writing the code of the corresponding control algorithm in the LabVIEW software,the completed code is downloaded to the NI myRIO,and the actual balance control experiment is carried out.By adjusting the relevant parameters,the balance control on the edge and corner of cube robot is realized.