Picking robot arm based on adaptive neural network and control method thereof

Abstract

The invention discloses a picking robot arm based on an adaptive neural network and a control method thereof. With the picking robot conducting quick picking as a target, key techniques of a picking robot arm controlling method and denoising of a robot vision image are achieved, and quick picking of the picking robot is achieved. By means of the control method, an RBF neural network control algorithm is adopted for calculating interference, and a kinetic model of the picking arm under an ideal condition is firstly built; then according to uncertain items existing in the system, an adaptive neural network controller is designed to correct the dynamic model; then the stability of the controller is proved through a Lyapunov function, and a system convergence condition is given, and thus the robot arm is subjected to negative feed control and offset interference. The picking robot is short in fruit identifying and locating time, and the picking efficiency is high; the control method is high in interference resistance and self-learning capability, and the uncertain items of the model can be quickly and accurately estimated and compensated.

Description

technical field [0001] The invention relates to a visual servo control-based, automated picking robot arm and a control method thereof, in particular to a picking robot arm based on an adaptive neural network and a control method thereof. Background technique [0002] With the development of science and technology, agricultural robot technology is becoming more and more mature driven by technology, which makes the development of picking robots reach an unprecedented height. As the "eyes" of the picking robot, the vision system is used to perceive the environment and realize the identification and positioning of the fruit target. Whether it can accurately and quickly identify the picking target directly affects the reliability and real-time performance of the picking robot. It affects the picking efficiency of picking robots. Therefore, the recognition of the target is one of the key links of the picking robot, and it has also attracted the attention of many scholars, and ha...

AI Technical Summary

Problems solved by technology

In this way, negative feedback control is carried out on the robot arm to offset the interference, which solves the problems of longer identification and positioning time of the fruit, continuous picking interruption, and low picking efficiency

Images