Research On Path Planning And Object Tracing For Mobile Robot Based On Intelligent Optimization Algorithm

The robot navigation technology involves multiple research areas such as computer,communication,mechanical electronics,information,automation and artificial intelligence,and so on,which has a wide range of application in every domain.As the key technology of mobile robot,it has become a hot spot for domestic and foreign scholars.In a relatively closed unknown environment where mobile robots cannot accept GPS signals,such as underwater exploration,tunnels,and indoor services,mobile robots(warehouse driverless trucks,etc.)are required to estimate their position and build an abstract map of its environment and obstacles(Simultaneous Localization and Mapping).At the same time,they can effectively avoid obstacles to reach the target task point under the guidance of abstract map(Path Planning).In this paper,taking the indoor wheeled mobile robot as the simulation object,the robot path planning method and the robot target tracking and positioning method based on the intelligent bionic algorithm are studied.The main content and achievements are as follows:For the complexity(the size of the search space)and depth(the iteration step,the length of running time)of robot path planning algorithm based on ant colony algorithm,two kinds of improvement measures are put forward in rasterize environment:The global path planning method based on the potential field ant colony algorithm is proposed.The local pheromone diffusion model and the pheromone diffusion grid table are established,and the pheromone of the current path is smoothly spread along the direction of the virtual potential field force of the path point to the adjacent path and superimposed on the global pheromone,which ensures the smoothness of pheromone and enhances the pheromone concentration in the subspace of the implicit global optimal path,reduces the search space of the algorithm solution,strengthens the cooperation ability between the ants and reduces the complexity and depth of the ant colony algorithm.On the basis of the above method,geometrical method is applied to the local path planning according to the different characteristics of the local path to obtain another path to complete the pheromone updating of the two paths.Namely,one ant has the ability to search for two full paths,which improves the searching efficiency of the ant colony algorithm and reduces complexity and depth of ant colony algorithm.The ant individual's search efficiency is improved.In this paper,a novel method of object tracing based on linear model is presented for modeling of nonlinear systems.As a consequence,the measurement matrix can be factored into a state matrix and a landmark matrix using singular value decomposition,one of which contains the information of landmarks and the other contains the pose of mobile robots.The relationship matrix can be learned between the measurement matrix and landmark matrix via linear regression,and recover the original robot state matrix.In the end,this paper introduces the mobile robot simultaneous localization and mapping(SLAM)based on extended kalman filter(EKF)algorithm,and analyzes the consistency of pose estimation of robot and position estimation of landmark in different environments,and simulates the cause of inconsistency.The analysis points out that it is possible to reduce the inconsistency of robot pose estimation and position estimation by closed-loop sampling path or shortened observation period.The mobile robot is positioned in the same way as the map building,completing the grid map building,and the robot is guided by the path planed in the in rasterize environment in advanced and the robot is guided to the target point safely.