Research On Path Planning Of Redundant Robot

With the development of economy and the progress of science and technology,medical rehabilitation robot is getting more and more attention,the mechanical arm as the main mechanism of its load capacity,quality,flexibility,obstacle avoidance ability,the ability to perform complex tasks are put forward higher requirements.Research on path planning of redundant robot is the theoretical basis of robot's task.In this thesis,inverse kinematics for redundant manipulator and path planning for redundant manipulator were researched,which was armed at a 7-DOF redundant manipulator for Independent research and development.The main research contents were as follows:First of all,kinematics of manipulator was analyzed,a 7-DOF manipulator model was framed by D-H parameters method,and its Kinematics equation was derived.For inverse kinematics,inverse solution based on singularity robust inverse(SRI)was studied,using damping least-square method to avoid singularity,using joint limits as the optimization index to solve the kinematics inverse solution.The problem of the joint velocity transfinite near the singular structure by the pseudo inverse method was solved.The inverse kinematics equation of redundant manipulator was established,which can avoid the problem that the joint angular velocity is too large,and has a high accuracy of inverse solution,the pseudo inverse method and the singular robust inverse method were simulated and compared.Secondly,the path planning of the manipulator in the free space was analyzed.The joint space path planning was carried out by using the three spline function with the constraint condition of initial and ending angular velocities.The interpolation method was used to realize the straight line and circular arc motion in Cartesian space,and the path planning process was verified by simulation.The obstacle avoidance for manipulator was improved,on the basis of the traditional artificial potential field method.The collision detection technique based on ellipsoidal bounding box was introduced,and then the collision detection model was established.In order to solve the problem of conversion from low dimensional space to a high dimensional space,and the manipulator could not be looked as a particle,the search space from the Cartesian space was transformed to the joint space,the process of obstacle avoidance and path planning was the process of the total potential energy of mechanical arm gradually reduced.A method of establishing virtual target point was proposed,which could solve the problem of minimum point.Thirdly,combined with MFC and OpenGL technology,the 7-DOF manipulator interaction control system was established,which was composed of debugging system and control system.According to the function module of the program,four modules were established,which were simulation preview module,algorithm decision module,information transmission module and information storage module.Finally,the algorithm was validated by the 7-DOF manipulator,and the experimental results were analyzed.Firstly,the effectiveness and accuracy of the proposed algorithm were verified,and the effectiveness of the obstacle free path planning algorithm based on interpolation and the obstacle avoidance path planning algorithm based on improved artificial potential field method was verified.