Trajectory Tracking And Obstacle Avoidance Control Based On Omnidirectional Mobile Robot

With the rapid development of intelligent technology,robots are playing an increasingly important role in people's life.Nowadays,they can be seen everywhere,which were originally specially created for industrial manufacturing.As one of the important branches of robot,mobile robot has been widely studied and applied because of its characteristic of easy movement and effective expansion of working range.In order to give full play to the advantages of mobile robot in motion,people began to develop the robot structure with more powerful motion ability.In the context,the first omnidirectional mobile robot with integrity constraints and all-round motion ability was created.It can be applied to complex realistic scenarios due to its high flexibility and stability,therefore,the research on omnidirectional mobile robot has great practical significance.Based on the omnidirectional mobile robot,we studied the trajectory tracking control and obstacle avoidance method.Firstly,the mathematical modeling analysis of the four-wheel omnidirectional mobile robot was carried out,and its kinematic model was derived.On this basis,we established the dynamic error model between the center point of the mobile robot and the reference pose.Then,according to the purpose of trajectory tracking control,the cost function of model predictive control was designed,and the optimal control problem was transformed into a quadratic programming problem with constraints.Finally,the optimal control output was obtained to enable the mobile robot to track the reference trajectory stably.Aiming at the obstacle avoidance problem in the system,we developed an autonomous obstacle avoidance scheme based on the artificial potential field method and the Bug algorithm,which are used to solve the local path planning problem in the known environment and the dynamic environment respectively.As the artificial potential field method inevitably exists the local minimum problem,we integrated the wall following method into it to help the mobile robot to escape from the local minimum trap.The path was shortened by optimizing the mode switching condition and choosing the better moving direction in the following wall method.At the same time,an obstacle filling strategy was proposed,which greatly reduces the difficulty of path planning.In addition,based on the traditional Bug algorithm,a novel No target-Bug algorithm was developed,which can guide the system to avoide collision when there is no clear target point under the remote control mode.By subdividing the process of moving around obstacles in the NT-Bug algorithm into several stages,a more efficient movement is realized comparied with the traditional Bug method.Finally,a shared control strategy based on the omnidirectional mobile robot was proposed,where the operator can realize shared control of the robotic movement through a human-machine interaction manner.In the shared control system,the operator's EMG signals were transformed into motion instructions,which were used to control the moving direction and linear velocity of the mobile robot,respectively.Besides,the mobile robot system can realize autonomous obstacle avoidance in the guidance of the NT-Bug algorithm.The feasibility of the proposed shared control strategy and the effectiveness of NT-Bug algorithm were verified by two practical experiments in the indoor environment.