Research On Gait Predictive Control Of Three-legged Robot

The walking gait of the three-legged robot is a hotspot in the research field of the three-legged robot.In the gait experiment of a three-legged robot,the unknown obstacles on the ground are often not taken into account.In the practical application of complex environment,the adaptability of robot to unknown ground environment is particularly important.In this thesis,the three-legged robot is used as the research object.Based on the energy stability margin,predictive control is used to optimize and generate the robot's gait,so that the robot can adapt to the ground with unknown obstacles.The main research work of this thesis is:1.The structure and degrees of freedom of the three-legged robot were analyzed,and DH matrix was used to model the kinematics of the robot.Based on the results of kinematic modeling,numerical simulation is used to analyze the range of the footwork workspace,which lays the foundation for gait analysis.2.Based on the structure of the robot,the gait of the three-legged robot is analyzed and designed.According to the gait characteristics of the three-legged robot and the theory of energy stability of the robot,a control strategy to improve the gait stability of the three-legged robot was proposed,and the control strategy is verified by numerical simulation.The simulation results show that this control strategy can make the robot maintain a large energy stability margin.3.Based on the control strategy,a predictive controller is designed to control the elongation and shortening of the robot's virtual legs to obtain a greater stability margin,and at the same time optimize and generate gait.In order to solve the overshoot problem in predictive control,constraint optimization is introduced,and the predictive control with variable constraints is used to optimize the state space model.The two predictive control algorithms are compared through simulation.The experimental results show that the predictive control algorithm with variable constraints can reduce the overshoot.In order to identify the unknown ground environment,the gyroscope data is introduced to identify the ground through the Euler rotation equation,and the online generation algorithm of the reference trajectory of the predictive controller is proposed.Finally,the feasibility of the reference trajectory generation algorithm is verified through simulation.4.On the V-REP physical simulation platform,build a virtual prototype of the threelegged robot.Adopt shared memory to realize data communication between V-REP and MATLAB,and build a united simulation platform.Finally,the physical simulation of the robot adapting to the obstacle is carried out through the united simulation platform.The simulation results show that the control method proposed in the thesis can make the threelegged robot adapt to walking on terrain with unknown obstacles.