Research On Impedance And Vision Control Based Handling Robot For Chemistry Laboratory

In the chemical industry,handling hazardous materials requires special protective measures and operating methods to guarantee the safety of staff.In purchase to avoid the risk of personal injury and environmental pollution,handling robots can be used to replace the manual personnel in the handling and transportation of hazardous materials.This paper focuses on the motion control of the handling robot arm,and proposes a hybrid position/force impedance control method based on adaptive Jacobi matrix and RBF neural network,and also combines machine vision target detection technology to assist the robot arm in handling operations.The main contents of this paper are as followed.The kinematics of the robotic arm is established with a detailed analysis and modeling,including the establishment of the base coordinate system of the robotic arm and the forward and inverse kinematic equations.A hybrid position/force impedance control method is proposed,which utilizes an adaptive Jacobi matrix and radial basis function neural network(RBFNN)for control.The control method consists of an outer loop and an inner loop,where the outer loop uses a combination of PID control and impedance equation to quickly eliminate force tracking errors and reduce force overshoot.While the inner loop uses an adaptive Jacobi method to estimate the uncertain Jacobi matrix and uses an adaptive RBFNN to approximate the uncertainty term in the system and the estimation error of the Jacobi matrix,and a robust term is used to compensate for the approximation error of the RBFNN and external disturbances.This method can effectively control the motion and force of the robotic arm and improve its control accuracy and robustness.The effectiveness and feasibility of the control method are illustrated by Matlab simulation results.Relevant techniques and methods for machine-vision target detection are discussed,and the effects of different lighting conditions on target recognition are explored.A method based on bilinear interpolation downsampling and Canny edge detection is proposed to improve the accuracy of the recognition under different illumination conditions,and it is experimentally verified with April Tag tag block as the experimental object.The results show that the proposed method can effectively recognize target objects under the different illumination conditions.The feasibility of the proposed method is verified by experiments.The effectiveness of the proposed method is tested by using Jet Max robotic arm for target tracking,grasping and other experiments.