Solution to redundant mechanical arm end effector posture keeping

Abstract

The invention discloses a solution to redundant mechanical arm end effector posture keeping. The solution comprises the following steps that S1, a D-H model is established for a mechanical arm to obtain a forward kinematic model, derivation is carried out on the forward kinematic model, velocity layer conversion is conducted, and target trajectory equality constraint indexes are set up; S2, numerical values of a normal vector of a mechanical arm task surface in a mechanical arm base coordinate system are measured; S3, two end effector posture keeping equality constraint indexes are built through the established D-H model and the normal vector; S4, the established target trajectory equality constraint indexes and the two end effector posture keeping equality constraint indexes are written into quadratic programming problems of a unified form; S5, the quadratic programming problems are converted into a linear variational inequality; S6, a primal-dual neural network is utilized to solve the linear variational inequality; and S7, a mechanical arm joint angle control quantity solved by the primal-dual neural network is output to a machine to achieve control over the redundant mechanical arm.

Description

technical field [0001] The invention relates to the technical field of motion planning of a redundant robot, in particular to a solution for attitude maintenance of an end effector of a redundant mechanical arm. Background technique [0002] In recent years, robotic arms have become more and more widely used in complex environments: assembly lines in industry, service robots in homes and restaurants, medical robots, rescue robots after disasters, and so on. Various applications have high requirements on the dexterity of the manipulator or mobile manipulator and the optimization ability of various indicators. Redundancy The manipulator has redundant degrees of freedom, so it can complete many subtasks while completing the main task. Therefore, compared with non-redundant manipulators, redundant manipulators have broader application prospects and research value. [0003] For redundant manipulators, the most basic problem is its inverse kinematics. The problem of inverse kin...

AI Technical Summary

Problems solved by technology

However, the algorithm based on Euler angles has many disadvantages: if it is necessary to ensure that the end effector and the brush are perpendicular to the task plane at the same time, or if there is an angle when the end effector clamps the brush, this method needs to measure and convert the angle expressions in Euler angle space, and computing the pose of the task plane

In general, traditional attitude-keeping algorithms are very complex and not convenient for practical use

In order to effectively solve the problem of posture deformation of the end effector of the manipulator, it is urgent to propose an algorithm for maintaining the posture of the end effector, which can effectively solve the problem of posture deformation of the end effector.

Images