Research On Path-Planning For Mobile Robot

Path-planning is an important topic in mobile robot field. For based on the information got by sensors the robot's actions should accord with robot's kinetic restriction and other essential restrictions, mobile robot path-planning is always a difficult problem in robot research. At present, there are a lot of methods which could be used in robot path-planning, such as the artificial potential method, the grids method and artificial intelligence methods: genetic algorithms, neural network algorithms, ant colony algorithms and so on.This thesis mainly focuses on the research of genetic algorithms and ant colony algorithms and solves the problems of path-planning in the soccer robot system with effective methods.First of all, various typical methods in mobile robot path-planning research are introduced, and these methods'performances are discussed.Next, this thesis's research presents genetic algorithms and ant colony algorithms. Genetic algorithms are discussed in two areas: discrete space and continuous space. Max-Min Ant System algorithm, which could be used in real soccer robot system by optimizing the environment model, is introduced. These methods discussed above are realized by MATLAB emulation experiments, and then merits and demerits of these two algorithms are discussed after comparing the experiments'data and figures. These two methods'feasibilities are confirmed on soccer robot competition platform.Finally, this thesis introduces the soccer robot competition platform in detail, and an algorithm of soccer robot's shooting in arc path is proposed according to people's experience of shooting. This algorithm is proven to be effective and practical by the real system test.