| Modular robot is composed of several basic modular units with the same structure and communication mode,which has the characteristics of flexible configuration and good interchangeability.As a new wearable robot,the supernumerary robotic limbs can be fixed on the user by wearing.As an additional mechanical limb,it can assist the wearer to complete a series of complex and high-intensity tasks,which has a wide application prospect.In this paper,we combine the concepts of the two,and use a modular reconfigurable robot with chain configuration to form a supernumerary robotic limb manipulator,which makes it flexible to switch configuration and has strong adaptability.In this paper,the flexible configuration of modular supernumerary robotic limbs is analyzed,and selects six basic modular units to form a typical configuration of a sixdegree-of-freedom external limb manipulator with three parallel axes at the base end and three vertical axes at the end.The kinematics modeling of the modular external limb manipulator of this configuration was carried out by D-H parameter method,and the forward and inverse kinematics were solved and verified.Finally,the Monte Carlo method is used to solve the workspace of modular supernumerary robotic limbs in typical configuration,and the actual workspace of modular external limb robot is obtained by considering the safe working range of wearer and wearer.A position-based impedance control strategy is designed to achieve the end compliance control of the modular supernumerary robotic limbs,and a specific control process is designed for the typical compliance control application scenario of drag teaching.Through Simulink simulation,the influence of impedance parameters on the compliance control system is analyzed.The gravity compensation algorithm of the end effector is designed to eliminate the influence of the end effector’s gravity on the sensor readings.Finally,by establishing the human-robot system global coordinate system and the coordinate system relationship derivation,the compensation of the wearer’s movement is realized,and the influence of the wearer’s movement on the robotic limb movement is eliminated.Use the cylindrical bounding box to simplify the model of the modular outer limbs and obstacles.The artificial potential field method is selected as the basic algorithm for obstacle avoidance path planning.The improved artificial potential field method introduces the distance factor of the target point,establishes a new potential field function,and solves the problem of excessive gravitational force of the distant target point and the unreachable target point near the obstacle.Using the total potential energy as an index,the optimal joint angle combination is searched in the joint space,and the problem of the robot arm falling into a local minimum is solved by setting the intermediate path point as the virtual target point.Based on the above work,the algorithm flow of the obstacle avoidance path planning for supernumerary robotic limbs based on the improved artificial potential field method is given.Finally,according to the scheme and algorithm design,a modular supernumerary robotic limbs experimental platform was built.The basic performance of the modular supernumerary robotic limbs was verified through the rapid module replacement experiment and auxiliary work experiment;the compliance control algorithm was verified through the compliance drag experiment;the feasibility of the obstacle avoidance path planning algorithm was verified through the Coppelia Sim simulation. |
