Research On Mobile Manipulator And Its Force Feedback Teleoperation Based On Shared Control Mode

Mobile manipulator has both mobile and operational capacity, and has larger operating space and better operational flexibility. It is a hot research focus in the robot research area. It is the ultimate goal of intelligent robots research to make the robots with sensors complete the specified operation task autonomously. However, being confined to the current technological level, it is difficult for robots to complete all work tasks autonomously. The combination of part autonomy and network teleoperation of the robots to realize the integration of artificial intelligence can greatly improve its ability to fulfil complex and dynamic operational tasks. The important issues of shared control based mobile robot teleoperation system including mobile manipulator design, the method of improving robot autonomy and the method to support operator improve operational efficiency. In this paper, it developed a mobile robot firstly, and then studyed the self-autonomy motion planning, and finally researched the force-feedback time-delay compensation method for mobile platform teleoperation and manipulator teleoperation.In allusion to the application background of control-sharing teleoperation, we studied an omni-directional wheeled mobile manipulator. The mobile manipulator has redundant network communication structure, which can communicate with the control network reliably. The mobile platform employs three-wheel all driving structure, which can realize omni-directional motion and is suited to narrow movement space. The wheel module employs differential drive structure which is simple and easy wiring and the turning of the wheels is free of the restrictions of angle range.restrictions. The manipulator is designed as series of 6-dof structure, and the axis of the end's three-axis intersect at one point with analytic inverse solution, so that it's easy to control. Mobile Manipulator configured infrared ranging, ultrasonic ranging, encoders, ultrasonic positioning, digital compass, camera, temperature, six-axis force / force moment and many other sensors, which provide the conditions for the robot to realize local autonomy. Omni-directional Mobile Manipulator has nine freedom, and it belongs to redundant robot. In allusion to the IKP of redundant robot, this paper proposes a kinematic inversion method based on the combination of genetic trust reign algorithm and analytical solutions. It can calculate the inverse of the robot rapiddly.Real-time motion planning of mobile manipulator is an important part of its autonomy. Motion planning includes path planning and trajectory planning. Using the traditional artificial potential field for mobile robot path planning prone to raise problems of local minimum point and destination failure and so on. In order to overcome the shortcomings of traditional methods an improved artificial potential field method is proposed. We build up an improved artificial potential field model, and based on this we get the algebraic superposition of all kinds of potential field strength, and we use trust reign algorithm for searching for the minimum point of the sum of potential field strength within a sampling period of moving range, and a number of minimum points constitute the global optimal path. Trust reign algorithm has the character of rapid convergence, and the algorithm can satisfy the real-time requirements of arithmetic. For manipulator trajectory planning, in order to ensure the real-time control, we employ the trajectory planning method based on joint space to real-time plan the trajectory of Manipulator.Manipulator represents the operational of the mobile manipulator, and its efficient teleoperaton can improve the operating performance of shared control teleoperation system of the mobile manipulator. Force-feedback technology in teleoperation system provides the operator with a force telepresence, which can greatly enhance the efficient operation of the operator. However, network transmission delay has greatly affected the operational performance of the system, and even cause system's instability. Wave variable method based on passivity can ensure system's stability with arbitrary time delays, but with poor transparency. In allusion to the force feedback teleoperation of manipulator with multiple degrees of freedom controlling problem, firstly, a universal wavelet transform formula is established for teleoperation system with multi-degree of freedom, and in order to overcome the problem of poor transparency of wave variable method, the wave prediction method is used to time-delay teleoperation system, and in allusion to the steady-state error of the master and slave end of teleoperation system caused by variable time delay and inaccuracy of the prediction model, forward correction controlling method is proposed which can effectively decreace the steady-state position error. For the changing teleoperation system's task for the robot model, the method of using adaptive wave predictor to deal with the changes of the model is proposed, and the work strategies of the adaptive wave predictor is designed.Mobile platform represents the mobility of the mobile manipulator, teleoperation of mobile platform can complement its lack of autonomous movement, and combining the teleoperaton of the mobile platform and manipulator can give full play of "mobile" and "operation" feature. The force feedback controlling strategy of mobile vehicle is studied, and for the problem of the force feedback teleoperation of mobile vehicle the combination of wave variable method and prediction technology and multi-channel teleoperation control is proposed, which can improve its transparency while ensuring the system's stability.We established a web-based teleoperation experimental system, and achieved the experiment of making the robot carry and place objects in large region, and applied dynamic programming method,mobile vehicles and manipulator's force feedback teleoperation controlling method into experiments to verify the effectiveness of the proposed method, and this lays the foundation for the realization of shared remote controlling.