Research And Design Of Wheeled Robot Path Planning System

Wheeled robots are widely used in the fields of substation inspection,medical assistance,life service and so on.Its traditional architecture has a growing hardware level that is not compatible with a unified software platform.Especially in the field of driverless vehicle,the driverless vehicle requires high real-time data and robust system,which is another challenge to the traditional robot architecture.At the same time,wheeled robot path planning technology requires a shortest and optimal path in planning path,so we need to solve many problems on the built wheeled robot model,such as low algorithm efficiency,unreasonable path planning and easy to fall into local optimum.To solve this above problems,this paper propose that use a strong open source robot operating system(Robot Operating System)to integrate the SLAM(Simultaneous Localization and Mapping)and PP(Path Planning)technology to build a wheeled robot path planning system.It also solves the problem that the traditional inspection robot lacks the ability of autonomous environment perception and the autonomous path planning.Then,the global static path planning algorithm is studied and compared.By improving the algorithm,we can improve the efficiency of the algorithm and avoid the local optimal condition,so that the path planning results are more reasonable.The main contents of this topic include the following parts:Firstly,an experimental simulation model of a wheeled robot is built.It mainly includes the wheeled robot physical model,the indoor experimental operation environment and the runway experimental operating environment.At the same time,the physical parameters of wheeled robot are designed accordingly.Secondly,the global static path planning algorithm Time-efficient A* is improved.This paper proposes an improved Bi-directional Time-effcient A* algorithm,which will run Time-efficient A* algorithm from the start point and end point to find the path.It solve the problem of unreasonable path planning by multi neighbor grid distance calculation.Thirdly,based on the physical model of wheeled robot and the experimental running environment of wheeled robot,the path planning system of wheeled robot is built on ROS.The system has the ability of autonomous environment perception,map building and autonomous planning.At the same time,this paper uses the system to carry out experimental research and result analysis.In conclusion,the wheeled robot path planning simulation system built in this paper has completed the experiment of building map and path planning,and has no obstacle collision in the process of operation.At the same time,the Bi-directional Time-effcient A* algorithm based on Time-efficient A* solves the problem that the path planning results are not ideal and the path planning length is increased.It is proved that the improved algorithm is reasonable and effective.