The Study Of Path Planning And Collision Avoidance For Autonomous Mobile Robot

Mobile robot is an important branch of robotics. With the continuous progress of robot's technology, the robotics becomes more powerful. For the diversity and complexity of the environment, path planning has being an important aspect in the mobile robotic research field and also a crucial technology to intelligentize the robot. From the point of deal with the especial situation of environment partially unknown, this thesis researches on two mainly sides there are real-time dynamic path planning algorithm and localization and mapping with multi-sensor data fusion. Both the simulation result and the experiment data gained from mobile robot path planning system are provided in the dissertation.Firstly, this thesis analyses the researching situation is elaborated and summarized of intelligent mobile robot. And then, this thesis summarized the path planning of mobile robot and points out the development and research aspects of the path planning technology. And discusses the algorithm of path planning based on the D* algorithm (dynamic A* algorithm) and the partially-unknown environments. The D* algorithm is a kind of incompletion replanning, and makes use of the original planning information, in a way that combines the characters of the optimization and the real-time. D* algorithm is able to use the global planning and local information, also good at utilize the off-line planning and on-line planning. By researched on D* algorithm, an improved D* algorithm is studied in this thesis. In the view of process unknown obstacles, the experiment result shows that the improved D* algorithm can find a feasible path in an optimal time and which is effective and can result in higher quality path than the conventional D* algorithm at the same map environment.To consider the particularity of information partially-unknown environment, this thesis analyzes the performance and infections on sensors of robot system. The method of environment express is presented. And then, this thesis research and realize the localization and mapping of mobile robot and renew the map in time for mobile robot navigation, by using probability method which is based on Bayes estimate to fusion multi-sensor data from the environment. Finally, mobile robot path planning software in Visual Basic environment is designed and realized with the help of WiRobotX80 ActiveX controller WiRobot SDK. Such essential technology as environment map display, sensor data collection, analyze and process, map grid and renew, real-time path planning and motion control and so on can be realized by this software. And path planning experiment of research in partially-unknown environment is implemented. The results of the research show the validity of work in this paper.