Error Analysis Modeling of Multi-legged Walking Robots

Determining precisely the motion of robot body is of great significance to the location and control of mobile robots. If the positions of each foothold and the driven joint variables are inaccurate, actual posture of the robot is likely to deviate from its required position. If the expected accuracy of robot´s locus is given, the robot can walk under expectation by controlling the accuracy of driven joints. On the basis of the inverse kinematics of a multi-legged walking robot, in this paper, the relation between accuracy of robot´s locus and errors of driven joint variables is discussed. The error expressions of a robot were obtained according to its structure and motion constraints, which include the motion errors of footholds in body and hip coordinates, the motion error of orientation matrix, and the error of driven joint variables. The equation set was obtained, which expresses relation among error of driven joint variables, structure parameters and error of robot´s locus. The formulae are derived in this paper. The analysis process is verified through a numerical example. This analysis model is the base of error compensation in controlling walking robots.