Path Planning Of Mobile Robot

With the development of information and artificial intelligence technology, the technology of autonomous mobile robot with perception, decision and movement functions is also developed quickly. Mobile robot navigation is an important direction of the robot study, and the path planning is the basic point of robot navigation, is also an important aspect of artificial intelligence in robotics area. It is that a robot figures out a path from start position to goal position and avoids collision in an environment which contains some obstacles. At present, many scholars have put forward their methods and many achievements have been created. Path planning is not an independent aspect in the research of mobile robot. It is relative with many other aspects,such as sense system, communication system and the rules of collaboration. This thesis mainly does the research on the path planning problem for mobile robot. The results of simulation demonstrate the validity of proposed theory and methods. The mainly comment and achievements of the thesis are as follows:1. Summarize current research progress and the tendency of mobile robot, introduce the different development level of foreign and domestic and sums up the basic problem of the path-planning for the mobile robot. The research meaning of the study is expatiated in the end.2. Briefly introduce the biological foundation, composing system, process and character of genetic algorithm, explain the definition of common terms in genetic algorithm, and discuss several realization methods of different genetic operators.3. At first introduce the main idea of D*Lite algorithm, and point out two shortages of D*Lite algorithm that it have redundant computation and its data structure is not good by searching the process of D*Lite algorithm, then design a path-planning algorithm based on optimizing the D*Lite algorithm, at last prove the optimized algorithm conquer shortages of the D*Lite algorithm and has a better performance by comparing to the D*Lite algorithm in experiment.4. At first introduce the main idea of ARA* algorithm, and point out two shortages of ARA* algorithm that it storages too many nodes with no use in memory and its computation is redundant sometime by analyzing the process of ARA* algorithm, then design a path-planning algorithm based on optimizing the ARA* algorithm, at last prove the optimized algorithm conquer shortages of the ARA* algorithm and has a better performance by comparing to the ARA* algorithm in experiment.5. At the base of classical genetic algorithm and ARA* algorithm, use grids method to do environment modeling, produce initial population by mixing two-way ARA* algorithm and random generation algorithm, and design three genetic operators which can be used to grid method, then design a path-planning method based on improved genetic algorithm, at last prove the efficiency of the improved algorithm with experiment.Finally,the work of this dissertation is summarized and prospective of further research on the path-planning of mobile robot is discussed..