Research On Workspace,Singularity And Path Planning Of Parallel Robots

Robot is the key factor in realizing automated production as well as promoting quality of human life.Ministry of industry and information technology of People's Republic of China try to promote industrial robots and service robots,thus making the development plan of robot industry(2016-2020).Compared to serial robots,parallel robots have the advantages of big stiffness,high precision,concise structure and heavy capacity,thus being widely used in parallel machine tools and motion simulators.However,because parallel robots have the drawbacks of relatively small workspace and complex singularities inside the workspace,their promotion and application are limited.Therefore,related theoretical research is important for both improving the performance and promoting its application.Supported by the national natural science funds of China,"Hybrid Energy Flow Method for Prediction of Broadband Vibration and Acoustics in Extra Large Structures Such As High-Speed Train Carriages"(51675306),"Active Control of Precisoin Equipments' Multi-Dimensional Vibration Using Flexible Metamorphic Mechanisms"(51275275),and Foundation of State Key Laboratory of Automotive Simulation and Control,"Ambulance Vibration Stretcher and Its Active Control Based on Reconfigurable Parallel Mechanisms"(20161112),this paper completed the research on workspace,singularity,singularity-free workspace and path planning of parallel robots.Firstly,the kinematic and dynamic model of a general 6-DOFs parallel robot was built.The functions of jacobian matrix,inverse position solutions,inverse velocity solutions,inverse acceleration solutions and input forces were got.The kinematics of the 3RPR parallel robot and kinematics and dynamics of the 6/6 Stewart parallel robot were deduced.Secondly,based on the kinematic model of the general 6-DOFs parallel robot,a novel universal numerical approach for all types of workspace determination of parallel robots was proposed by means of discretizing orientation and position at the same time.The efficiency of the algorithm was analyzed.Then,the workspaces of the 3RPR parallel robot and 6/6 Stewart robot were calculated by the proposed approach.Thirdly,the definitions of constant orientation singularity locus,maximum singularity locus,absolute singularity locus and singularity probability were proposed,and their physical meanings were analyzed.The methods to graphically illustrate the singularity locus and singularity probability were proposed.Then,the singularity locus and singularity probability of the 3RPR parallel robot and 6/6 Stewart robot were illustrated by the proposed approach.The singularity locus and singularity probability can recognize singularity and guide the planning of workspace.Fourthly,to solve the problem that completely singularity-free workspace is relatively too small but functioning in the maximal workspace cannot guarantee singularity-free path,the definition of connected singularity-free workspace was proposed.And the approach to determine the connected singularity-free workspace was proposed by adapting undirected graph theories in data structure.Then the connected singularity-free workspaces of the 3RPR parallel robot and 6/6 Stewart robot were calculated by the proposed approach.Fifthly,optimum performance path planning was studied by adapting theories of weighed graph in data structure.Then a genetic algorithm to calculate the minimum time of the optimum performance path was proposed by means of random discretization.Finally,the optimum performance path and the minimum time of the 6/6 Stewart robot were calculated by the proposed approaches.