Mechanical arm path planning method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of mechanical arm control, and provides a mechanical arm path planning method and device, electronic equipment and a storage medium. A first starting point pose and a first end point pose under a space coordinate system are obtained, and position information of an obstacle is obtained; according to the first starting point pose and the first end point pose under the space coordinate system, a corresponding second starting point pose and a corresponding second end point pose under a joint coordinate system are calculated through inverse kinematics calculation; and each joint is taken as an agent, the angles of the joints are taken as state sets of the agents, {-theta, theta, 0} is as an action set of the agents, and a multi-agent reinforcement learning algorithm is utilized to generate a path from the second starting point pose to the second end point pose according to a preset collision detection rule and a reward and punishment mechanism, wherein theta is the preset minimum movement step length of each joint. Therefore, the singular solution and multi-solution problems caused by the fact that the angles of all the joints are calculated through inverse kinematics calculation by space coordinates can be avoided, and the stability of operation of a mechanical arm is guaranteed.

Description

technical field [0001] The present invention relates to the technical field of manipulator control, in particular to a manipulator path planning method, device, electronic equipment and storage medium. Background technique [0002] The path planning technology of the manipulator is an important link in the control technology of the manipulator. The main purpose of the path planning of the manipulator is to enable the manipulator to reach the designated end point from the designated starting point without collision. The existing manual path planning methods usually only meet the requirements of a simple working environment, and the degree of automation is low. In the existing method of automatic path planning based on spatial coordinates, when planning a path, every time a candidate path point is planned, inverse kinematics calculation is required to verify whether there is a collision or whether it is at a singular point (unclear The inverse of the robot is transformed into...

AI Technical Summary

Problems solved by technology

The existing manual path planning methods usually only meet the requirements of a simple working environment, and the degree of automation is low

In the existing method of automatic path planning based on spatial coordinates, when planning a path, every time a candidate path point is planned, inverse kinematics calculation is required to verify whether there is a collision or whether it is at a singular point (unclear The inverse of the robot is transformed into the coordinate point of the rotation angle of each joint axis of the robot, that is, the point where the inverse kinematics has no solution in the robot workspace), whether there is a multi-solution state, etc., the amount of calculation is large, and the path obtained by planning is sometimes just From the perspective of coordinates, it is smooth, but the rotation angle of the joint axis of the manipulator at adjacent path points changes greatly (that is, the joint space changes greatly), resulting in unstable operation of the manipulator and potential safety hazards

Images