Research On Path Planning Of Indoor Mobile Robots

Path planning technology is an important part of the study of mobile robots and is the basis for the realization of top-level tasks such as artificial intelligence control.The purpose of path planning is to enable mobile robots to read environmental information through sensors,use path planning algorithms to plan an optimal collision-free path,and finally arrive at designated locations to complete specified tasks.In this paper,based on the comparative study of the existing path planning algorithms,based on the advantage of laser radar rapid positioning and mapping,based on ROS(Robot Operating System)system using genetic algorithm and artificial potential field method to solve the optimal path planning problem.On the one hand,it is of great significance to improve and innovate the path planning algorithm.The other party is of great significance in advancing the product application of ROS-based systems.It also fits into the integrated platform of software,hardware,and algorithms built in China Manufacturing 2025 and further development intelligent barriers,autonomous planning technology.By comparing the advantages and disadvantages of various types of path planning algorithms,it can be concluded that a single algorithm cannot solve the path planning problems in both static and dynamic environments.However,different algorithms can complement each other,and the path planning is finally determined using a combination of genetic algorithms and artificial potential field methods.The design of the environment map encoding method and genetic operators in the genetic algorithm will affect the global path planning results.According to the actual situation of this study,the map is divided by grid method,the environmental information is innovatively coded using serial number method,and some genetic operators are improved.Compared with the traditional genetic algorithm in the simulation experiment and indoor experiment,the improvement is reflected.The global path obtained by the genetic algorithm planning is better and the time is shorter.In the local path planning,the artificial potential field method has a local minimum problem.It is analyzed that the main reason for this problem is the force balance of the robot during the movement process.Aiming at this problem,the method of removing obstacles was proposed and used in combination with the method of adding obstacles.Finally,it was verified by experiments that this method successfully solved this problem.In order to combine the two algorithms for path planning and apply the path planning theory to practice,we chose to use the ROS system as the basis to design the nodes of HP+Windows+Cmder+Oracle VM VirtualBox+Liunx(Ubuntu 14.04)+ROS indigo.Raspberry+ Liunx(Ubuntu 14.04)+ nodes in the ROS indigo environment connect to the master's distributed control architecture while enabling multi-sensor information fusion and distribution.Through the development of the move_base function package to encapsulate the two algorithms,so that the improved path planning algorithm in the ROS system for intelligent control,and through the configuration of rviz successfully visualized the actual results of the operation,and ultimately achieve the expected results of the study.