Mobile mechanical arm repetitive motion planning of non-convex anti-noise return-to-zero neural network

Abstract

The invention discloses a mobile manipulator repetitive motion planning method based on a non-convex anti-noise return-to-zero neural network. The method comprises the following steps: a, obtaining an integral mobile manipulator kinematics equation of a mobile platform and a manipulator based on integrity constraint and space coordinate transformation; b, time-varying quadratic programming of multiple sub-tasks of joint and wheel speed limitation, trajectory tracking and repeated motion is constructed for repeated motion of the movable mechanical arm; and c, proposing a non-convex anti-noise return-to-zero neural network to solve a quadratic programming model of the mobile mechanical arm under noise disturbance. The non-convex anti-noise return-to-zero neural network controller is designed on the basis that the difference value between a given expected trajectory function and an actual motion trajectory serves as an error function, noise interference in the trajectory tracking process of the mobile mechanical arm is restrained, and robustness is enhanced. And the non-convex set restrains the error change rate from a relatively large non-convex set to a very small non-convex set through twice restraint, so that the stability is enhanced, and a trajectory tracking task is finally completed.

Description

technical field [0001] The invention relates to the field of mobile robots, in particular to a trajectory tracking planning method for a four-wheel mobile mechanical arm based on a repetitive motion anti-noise type zeroing neural network. Background technique [0002] With the continuous progress of artificial intelligence technology, the intelligent robot industry has ushered in vigorous development. An intelligent robot is a machine system that fully simulates human beings in terms of integrated perception, thinking and effects. It can replace humans in various fields. When a mobile manipulator works in a dynamic, unknown and complex environment, it should have complete autonomy, that is, the system should have the ability to perceive, plan, maneuver and coordinate. planning, motion control, and cooperative control. According to the different control objectives, the motion control of mobile manipulators can be divided into three types: point stabilization, path following...

AI Technical Summary

Problems solved by technology

For example, the motor of the i-th joint of the mobile robot manipulator, its output joint velocity or joint acceleration may be discontinuous, that is, the pulse per second used to control the motor in the robot is expressed in a discrete form, which is obviously a non- In the case of convex mapping, the convex boundary constraint does not meet the actual situation

Images