The End Force Control Of 7-DOF Manipulator

As the development of industrial automation,the robots are increasingly used.In order to meet the task requirements in a complex environment,it is very important to control the force of robots effectively.The force control of the manipulator is to control the contact force between the end actuators and the environment in the constrained position.In this paper,the hybrid force/position control method is used to research the force control of the manipulator,and the working space of the manipulator is divided into the position subspace and the force subspace.The position subspace is located in the tangent direction of the contact surface,and the position control of the manipulator is performed in the space.The force subspace is in the normal direction of the contact surface,and the force control strategy is taken to the manipulator in this space.The position control of the manipulator is to control the end position of the manipulator in the workspace.The kinematics model of the manipulator should be set up first to realize the mutual conversion between the joint space and the Cartesian space coordinate.The D-H parameter method is used to establish the forward kinematics equation of the manipulator,and on this basis,the Jacobi matrix of the manipulator is derived.Since the inverse kinematics problem of the manipulator has multiple solutions,the Newton iteration method is used to obtain the optimal solution.Before the end force control of the manipulator,it needs to run smoothly from any position to the contact point,and the simulated annealing genetic algorithm is used to study the trajectory planning.The aim of force control is to achieve the end force of the manipulator effectively.The establishment of a dynamic model is the key to its research.The Lagrange method is used to deduce the dynamic equation of the manipulator,and the relation between the joint motion and the torque is obtained.In order to obtain accurate parameters,a current-based identification method is used to identify the kinetic parameters.Based on the kinematics and dynamics model of the manipulator,the hybrid force/position control method is used to research the end force control of the manipulator.The expected force and position are allocated to each joint by Jacobi matrix.The current-based force detection module is designed as force feedback,and the dynamic model is used to form the force control loop of the system.And in the position control loop,feedback is directly obtained by the builtin encoder of the manipulator.Considering the safety of the running process,the actual and theoretical values of the joint current and velocity are compared,and the difference is used as the input of the perceptron.Train the perceptron through experimentation for collision detection.If the manipulator accidentally collides with the environment during operation,gravitational compensation is applied to the current pose to achieve an emergency stop.In the experimental part of this paper,firstly we verify the correctness of the path planning method and the identification of dynamic parameters.Then design the contact-movement experiment to verify the effectiveness of the hybrid force/position control.Finally,the perceptron is trained with actual data,and the accuracy for collision detection is verified.