Tracking and real-time obstacle avoidance reaction planning algorithm applied to industrial robot

Abstract

The invention provides a tracking and real-time obstacle avoidance reaction planning algorithm applied to an industrial robot. The tracking and real-time obstacle avoidance reaction planning algorithmcomprises the steps: 1, the to-be-tracked curve pose, and the real-time distance between a tail end rigid body and an obstacle are input; 2, mapping from a robot joint space to a working space is given through forward kinematics; 3, the obstacle is detected through a visual sensor, and the distance to the obstacle is calculated; 4, the path is planned, and the rigid body of the working space is subjected to tracking and obstacle avoidance; 5, a joint angle is calculated according to a Jacobian pseudo-inverse method in the joint space; 6, the angular constraint problem is handled through a quick search spanning tree; 7, constraint is detected, if the constraint is not met, the step 6 is repeated, and if the constraint is met, the step 8 is executed; 8, a target is reached, the joint angleis obtained, the joint space control moment is calculated accordingly, if the target is reached, the algorithm process is completed, otherwise the step 9 is executed; and 9, an initial point is updated, and executing is conducted again from the step 3.

Description

technical field [0001] The invention relates to an industrial robot path planning and motion control technology, in particular to a tracking and real-time obstacle avoidance response planning algorithm applied to an industrial robot. Background technique [0002] In engineering, people require robots to complete specific tasks in the fastest or least energy-consuming way, so optimization methods are often used to deal with robot motion planning problems. However, the complexity of the robot's motion equations will make it very difficult or even impossible to solve the corresponding optimization problems. Using the theory of Lie groups and Lie algebras, a simple equivalent description of complex robot motion equations can be given. More importantly, the first and second derivatives of torque with respect to any joint variable of the robot can be accurately calculated. In this way, when solving the minimum energy consumption problem, the exact expression of the gradient and ...

AI Technical Summary

Problems solved by technology

Substantial difficulty in obtaining gradients in optimization problems when {en

Images