Cooperative Robotic Arm Collision Detection Research Based On Second-order Feedforward External Force Observer

Security is a key attribute of collaborative robots and has become a hot topic of research today.In order to ensure the safety of human-machine cooperation,the robot must be able to accurately detect external unconscious collisions and can make corresponding safety reactions while identifying the location of the external force action points.Most of the current lightweight cooperative robot arm research is based on force sensors and high-performance torque sensors,which not only increases the cost of the arm but also increases the complexity.Based on generalized momentum observer combined with dynamic threshold,the collision detection of without external force sensors is studied in this paper.The six-degree-of-freedom robot arm is used as the research object,and the dynamic model of the flexible joint robot arm with external disturbance is established in this paper.The second-order feedforward generalized momentum observer model is designed.Then the friction model is added to the model and optimize the joint parameters by least squares method.The multiplication method optimizes the joint parameters and performs simulation and verification on this basis.The research results show that the collision detection algorithm and dynamic threshold setting proposed in this paper can realize the external collision force detection function,thus promoting the development of human-machine cooperation security.The main contents are as follows:Firstly,based on the dynamics modeling of a six-degree-of-freedom cooperative robot arm.The kinematics of the six-degree-of-freedom robot arm is analyzed.The Cartesian space coordinate system and the Jacobian matrix of the joint space are established for the robot arm.Based on the Lagrangian theorem,the simplified flexible joint robot arm dynamic model under external disturbance is established.Perform force analysis to provide theoretical support for accurately identifying the position of the robot arm and achieving external force determination.Secondly,based on the optimization of generalized momentum,the robot arm collision detection algorithm.The first-order generalized momentum and second-order generalized momentum observer design are carried out,and the comparison experiments of various generalized momentum observer collision algorithms are carried out.The effectiveness of the improved second-order observer is verified by simulation software and based on the data of the experimental platform accuracy.Again,the optimization of the robot arm parameters based on the least squares method.The current and torque output values are optimized by the least squares method,the friction model is optimized,and the observer output values under the static threshold and the dynamic threshold state are compared to verify the accuracy of the dynamic threshold for collision detection,and the external force observer output is realized.The value exceeds the dynamic threshold.The arm immediately switches to the stop-and-go mode.Finally,an experimental study on the collision detection of a cooperative robot arm.Combined with the system to build a six-degree-of-freedom cooperative robot arm experimental platform,the collision algorithm is used to realize the real-time online data acquisition during the operation of the robot arm.By conducting contact experiments on different arms of the robot arm,the force position and the joint between the joints are obtained.Influencing the relationship,the Jacobian matrix mapping method is used to analyze the observer output values of each joint,and the effectiveness of the collision detection algorithm and the external force determination for the sensitivity of the collision detection is realized.In this paper,based on the safety of human-machine cooperation robot,the collision detection method without external torque sensor is developed.By compiling the collision detection algorithm and setting the dynamic threshold,it provides theoretical support for human-machine cooperation safety,and accelerates the research of next-generation cooperative robot.Carry out effective accumulation.