Research On Omnidirectional Mobile Robot Platform And Path Planning

In the 21 st century,the application of mobile robots is more and more extensive.With the development of science and technology,many service robots have replaced human labor in warehousing and logistics industry.The research of this paper is based on the mobile robot in the warehousing and logistics industry,mainly in terms of path planning algorithm and autonomy of mobile robots.Based on the analysis of traditional ant colony algorithm and artificial potential field method,the improved algorithm is proposed,and the good performance of the improved algorithm is proved by MATLAB simulation.Finally,the path planning experiment is carried out on the omni directional mobile robot platform built in the research.The experimental results further prove the practicability of the algorithm.The main content of the research includes the following aspects:First of all,a Macanum wheeled omni directional mobile robot platform is constructed in good flexibility and high stability.The kinematic and dynamic models of the mobile robot are established to achieve the control of the robot.A combined structure of the top-level upper computer and the lower-level lower computer is designed for the controller.The control chip of the lower computer is STM32F427,which can communicate with the chassis motor module using CAN bus technology.Proportion integration differentiation control method(PID)is utilized to achieve precise control of the robot.A laser-based ROS autonomous navigation framework is designed based on the core microprocessor of the top-level control system.Secondly,as to the global path planning of mobile robots,ant colony algorithm is used.However,due to the low efficiency and blindness of the conventional ant colony algorithm,the algorithm is improved in aspect of heuristic information in the transition probability.The improved algorithm is verified by simulation tests in different environments: simple level,medium level and complex.The simulation results show that the algorithm has a good performance and can effectively complete the path search task in different environments.In addition,in a certain range,the influence of parameters on the path planning of the robot is analyzed and optimized.The result shows that the optimized parameters can definitely improve the performance of the algorithm.Then in terms of local path planning,artificial potential field method is adopted.To solve the problem that the target point cannot be reached smoothly,the repulsion field function is improved by adding a term,the distance between the mobile robot and the target point,to the repulsion field model equations.As to the local minimum problem,a multiple factor is added to the gravitational potential field function.Thus,the gravitational force of the robot is increased to prevent the gravitational force and the repulsive force from being equal.The feasibility of the improved algorithm is verified by comparison analysis of conventional artificial potential field method and the improved algorithm.The locations of the obstacles are randomly set in the experiments to test the effectiveness of the improved algorithm in an unknown environment.Finally the improved path planning algorithm is applied in the ROS framework for experiments.Experimental results show that,compared with the conventional algorithm in ROS,the improved algorithm performs better.