Mobile Robot Navigation Algorithm And Platform Construction In Indoor Scene

With the continuous progress of society,mobile robot technology has been rapid development,the demand for mobile robots is also increasing.The rise of artificial intelligence,sensors,applied electronics and other fields has also injected fresh vitality into the development of mobile robots.This makes mobile robot technology permeate into all walks of life and become an indispensable partner of people.Nowadays,mobile robots tend to be intelligent and their autonomous navigation technology has become one of the hot research directions of scholars.Therefore,the research on navigation algorithm is the most important one.The mobile robot senses the external environment and its own state information through its own sensors,and then uses the navigation algorithm to plan an optimal path from the starting point to the target point in the working environment with obstacles.This is also known as mobile robot path planning technology.As one of the core components of mobile robot navigation technology,path planning technology has very important research value.Although the relevant intelligent algorithm can help mobile robot to complete the path planning,there are still some problems such as slow convergence speed,long planning path,not smooth and unable to guarantee security.In order to solve the above problems,this thesis gives a mobile robot navigation method based on hybrid artificial fish swarm algorithm.On this basis,a navigation method of mobile robot based on anti-collision A* algorithm is given.The main research work includes the following four aspects:1.Aiming at the problems of slow search speed,low convergence efficiency,easy to fall into local optimal and long path when traditional artificial fish swarm algorithm is applied to raster map,this thesis gives a mobile robot navigation method based on hybrid artificial fish swarm algorithm.By integrating the advantages of differential evolution algorithm,the searching ability and late convergence rate of fish swarm algorithm are improved.Multi-population mutation optimization strategy and adaptive parameter strategy are adopted for differential evolution algorithm to prevent the hybrid algorithm from falling into local optimal.Finally,the path planned by the hybrid algorithm is optimized twice to delete redundant inflection points and make the path smoother.2.Aiming at the problems that traditional A* algorithm can’t guarantee the safety of planning path,many redundant nodes and many turning points of movement trajectory,this thesis gives a navigation method of mobile robot based on anti-collision A* algorithm.The heuristic function of A* algorithm is improved by introducing safe distance factor to ensure that the path planned by the algorithm does not collide with obstacles.The plane structure method is used to eliminate redundant inflection points and ensure that the planned path is the optimal path.Finally,simulation experiments in MATLAB prove that the improved algorithm can plan a safe distance and smooth optimal path with obstacles.3.In order to study the dynamic obstacle avoidance problem based on anti-collision A* algorithm in unknown environment,this thesis gives a combined algorithm based on anti-collision A* algorithm and dynamic window method.When the mobile robot is in an unknown environment,only A* algorithm is not enough to avoid obstacles and reach the target point,so local obstacle avoidance function of dynamic window method is needed.A fusion sub-function is designed by using the coordinates of global optimal path nodes planned by anti-collision A* algorithm,and the evaluation function of dynamic window method is improved by using this function,so as to solve the problem that dynamic window method is easy to fall into local optimal.In the MATLAB environment,by designing the simulation environment of static unknown obstacles and dynamic unknown obstacles,it is verified that the combined algorithm in this thesis can not only plan the optimal path,but also realize the dynamic obstacle avoidance of unknown obstacles.4.This thesis designs an experimental platform based on QBot 2e intelligent car.Firstly,the host environment was configured,and the hardware structure and map construction process of the experimental platform were analyzed.Following that,the anti-collision A* algorithm in this thesis is imported into the control program of mobile robot.Finally,the visual sensor is used to construct the environment map and complete the obstacle avoidance task to verify the effectiveness and practicability of the algorithm presented in this thesis.