Mechanical arm on-line obstacle avoidance movement planning method

Abstract

The invention provides a mechanical arm on-line obstacle avoidance movement planning method, and relates to the technical field of mechanical arm movement planning. The mechanical arm on-line obstacleavoidance movement planning method comprises the steps that modeling is conducted on a mechanical arm and an obstacle, and a target node is determined through the inverse solution; in the pre-planning stage, sampling is conducted in the joint space to generate a new node, collision detection is conducted on the new node to determine whether the new node is added to a search tree or not, and a node which enables the cost of the new node to be lowest and is located in the search tree serves as a parent node of the new node; first k nodes are planned from the search tree and sent to a track planning device until the situation that the target node enters the re-planning stage is searched out; and in the re-planning stage, sampling is conducted nearby path nodes to optimize a path, the path isplanned again when the obstacle movement and the path conflict, and planned nodes are sent to the track planning device at the same time until a mechanical arm moves to the target node. By the adoption of the mechanical arm on-line obstacle avoidance movement planning method, the path planning and the movement of the mechanical arm are synchronously conducted, the on-line obstacle avoidance movement is achieved, and even though the position of the target node changes, sampling can also be conducted again, and the mechanical arm can be connected with the path and arrive at the target node.

Description

Technical field [0001] The invention relates to the technical field of robotic arm motion planning, in particular to an online obstacle avoidance motion planning method of a robotic arm. Background technique [0002] Motion planning is a basic problem in robotics research. Early motion planning was mainly for mobile robots. The mobile robot was regarded as a mass point on a two-dimensional plane, looking for a collision-free path from the starting point to the target point. At present, path planning methods for mobile robots are very rich. At present, such methods mainly include Dijkstra algorithm, A* algorithm, artificial potential field method, swarm intelligence algorithm, etc. For the multi-input multi-output, nonlinear, and strongly coupled high-dimensional complex system such as the robotic arm, these methods have the disadvantages of corresponding long time, large calculation amount, and easy to fall into the local optimum, and cannot realize the online obstacle avoidance ...

AI Technical Summary

Problems solved by technology

Too many sampling points will reduce the speed of finding the shortest path, and it

Images