For the abnormal halt of industrial robots in the working process resulting in their postures in an unknown state and the failure of automatic return to the initial position from the halt position,a homing method for industrial robot is proposed,which includes designing a robot return path,classifying the robot's pose information by combining the k-nearest neighbour algorithm and constructing return space and spatial classification threshold,reconstructing the robot feature data for decision tree model training andexcuting the robot regression program after using the trained model decision return path.Thus theautomatic return of industrial robots to the initial position being realized.The experimental results show that the success rate of robot return is 93.3%in random posture and with various tools.
industrial robothoming methodspatial structurepath designdecision tree algorithmk-nearest neighbour algorithm
国家科技期刊平台
NSTL
万方数据
