A spatial vector projection based error sensitivity analysis method for industrial robots

Abstract

Precision design is an important approach to guarantee the accuracy of industrial robots, where error sensitivity analysis lays a vital foundation. Intuitively, if the pose error vector caused by one error component is nearly perpendicular to the sum vector that is caused by all error sources, this error component will have little influence on the accuracy of end-effector even though its norm is large. Unimportant errors may unreasonably be regarded as vital errors when their direction characteristics are ignored. In this paper, the local sensitivity index (LSI) is defined based on the projection theory of spatial vector to evaluate the contribution of an error component in a definite pose. Both the direction and norm characteristics of pose error vector are considered in the novel definition of LSI, which is quite different from the previous researches. According to the definition of expectation and variance, the average sensitivity indices (ASI) and variance sensitivity indices (VSI) of error sources are established. Finally, the precision design of a 6-DOF (degrees of freedom) industrial robot is conducted in numerical simulation based on the proposed error sensitivity analysis method. The results show that the pose accuracy of end-effector can greatly be increased by improving very few error components obtained via error sensitivity analysis. The proposed error sensitivity analysis method is quite effective, which can greatly improve precision design effects.