A path planning method for autonomous obstacle avoidance of live working manipulator based on motion primitive library

Abstract

The invention provides a live line work mechanical arm autonomous obstacle avoidance path searching method based on movement primitive libraries. A complete mechanical arm kinematic model and a mechanical arm simplified kinematic model are built, and a mechanical arm working environment is built; the movement primitive libraries are designed in a classified mode, and corresponding movement primitives are designed for the various libraries to serve as library members; distance calculation functions between a mechanical arm and a target point and between the mechanical arm and an obstacle are built in a classified mode, and the danger area threshold value and the target point arriving precision threshold value are set; a heuristic function hd(x) in the Cartesian space and a heuristic function hc(x) in the articulate space, and a heuristic rule are created, and heuristic searching is achieved; and the current pose and the expected pose of the mechanical arm and the Cartesian space distance of the obstacle are calculated, and a path is planned according to the set judgment condition. Mechanical arm operation scene information can be built rapidly and precisely, it is guaranteed that the obstacle avoidance path searching result has suboptimality and completeness, the path can be planned rapidly under the premise of meeting the kinematic constraint, and the stationarity of the mechanical arm movement process is guaranteed.

Description

Technical field [0001] The invention belongs to the field of live working robots, and specifically relates to three-dimensional reconstruction of a multi-degree-of-freedom manipulator working environment and autonomous dynamic obstacle avoidance path planning, in particular to a method for path planning of an autonomous live working manipulator. Background technique [0002] At present, my country's high-voltage live operation methods are still at the stage of manual operation and robotic arm teleoperation. During manual operation, technicians are exposed to high-voltage environments, which is extremely dangerous; teleoperation needs to ensure that the robotic arm does not interact with the tower during the operation. The device collides and is in contact with only one wire at most, which places high requirements on the ability of technicians in on-site observation, prediction, and fine operation. [0003] The C-space method is a classic and effective algorithm in solving the proble...

AI Technical Summary

Problems solved by technology

The path planning algorithm based on sampling solves the problem of huge amount of calculation in high-dimensional C-space, but its performance depends on the sampling method and the connection mode between nodes and has randomness; when the artificial potential field method is used in high-dimensional C-space, the calculation The problems of large quantities and local minima have not been well solved; the heuristic search algorithm ensures the suboptimal path and the completeness of the search by establishing a cost equation, but for the multi-degree-of-freedom manipulator problem, the traditional heuristic search algorithm The huge amount of calculation makes it difficult to realize engineering applications

For the above methods, there is still an important problem, they cannot guarantee that the search results meet the kinematic constraints of joint velocity and acceleration and the stability of the manipulator during the search process.

[0004] In the field of 3D real-time reconstruction technology of robot operating environment, on-site reconstruction technology based on point cloud and depth camera has been widely used, and 3D measurement technology based on binocular camera has also achieved some success, but how to improve the accuracy of 3D reconstruction in complex environments and real-time performance is still a problem to be solved

Images