Research On The Technology Of Collision Detection And Physical Collaboration For Human-robot Interaction

As expected,it was predicted that robots will eventually interact and work closely with humans in diverse everyday environments as well as support them in industrial scenarios.How to ensure the safety of human beings is the premise of the above coexistence and collaboration.It is necessary to prevent accidental collisions in time,deal with unavoidable or intentional physical contact on the basis of safety and reliability,and make perception-based motion response.Thes are also core issues in the development of physical human-robot interaction(pHRI).Therefore,in view of the existing robot structure,this paper aims to explore the possibility of humancomputer safe physical interaction without external sensors.It mainly studies and discusses from the aspects of collision detection,time-varying threshold,intention detection and external force information acquisition.Firstly,the establishment of accurate robot dynamics model is the basis of the research work in this paper.The Newton-Eulerian method combined with the screw theory is used to derive the dynamic equation,which reduces the computational complexity of the traditional modeling method and improves the fastness and realtime.At the same time,in order to describe the robot motion more accurately,a parameter identification method based on least squares idea is proposed to accurately identify the joint nonlinear friction model.Then,the disturbance observer based on generalized momentum is introduced and optimized,and the tracking of external disturbance can be realized.Finally,a time-varying threshold establishment method for error compensation is proposed,which improves the detection accuracy of fixed threshold in the collision process.The purpose of the above research is to ensure the safety of human beings in the coexistence of human and robot.Based on this,the content of physical human-robot collaboration is discussed in this paper.Firstly,the robot's characteristics are used to determine the contact intention between human and robot by frequency.High frequency represents accidental collision,while low frequency soft contact represents intentional behavior.Robots will react differently according to human expectations,such as stopping directly or entering collaborative mode.This frequency resolution is achieved by filtering.In order for the robot to respond in the most appropriate way,collecting physical information from contact or collision events is an essential part.In this paper,using the Jacobian matrix property of the robotic force domain,combined with the disturbance observer output to solve the contact force magnitude and action position,and a reasonable derivation process is given.Finally,all the algorithms and research conclusions in the paper are verified on the Adams-Simulink simulation platform and the six-DOF robot experimental platform,and the experimental results are analyzed and discussed.