| The goal of achieving the autonomy of mobile robots has been the pursuit of people,in the field of intelligent mobile robot.Mobile robot in an unknown environment,only rely on their own sensors to perceive the surrounding environment,and then determine their own position,to complete the task of an independent operation.Autonomous positioning and map construction is assign of autonomy of mobile robots,is the premise of all other autonomous behavior.Mobile robot in the process of moving,constantly sensing the surrounding environment,and online real-time construction of the environment map at the same time autonomous positioning.Simultaneous Localization and Mapping(SLAM),which is the simultaneous localization and mapping of the environment map,is a research hotspot in the field of mobile robots in recent years.In the paper,the key technology of SLAM is studied deeply,and the algorithm is verified by the real mobile robot.Firstly,the coordinate system is established,the SLAM system model,the mobile robot movement model,sensor aware model,environmental map representation model;Secondly,the method of environmental map representation and data association method are studied deeply.The advantages and disadvantages of several map representation methods are demonstrated,and an algorithm of construction grid map is designed.The data association method based on grid map is studied.The nearest neighbor iteration(ICP)and the maximum likelihood domain model are introduced.The measurement noise of the sensor is modeled by the probabilistic method,and a scanning matching method based on the maximum likelihood domain model is designed.Thirdly,the particle filter algorithm is studied deeply.The standard particle filter algorithm is improved,which is applied to autonomous localization and map building of mobile robots.The proposed distribution function is improved by using the scanning matching method,and the sampling precision is improved.The sampling algorithm and the adaptive resampling mechanism of the kinematic model of the mobile robot are designed,which effectively reduces the risk of particle diversity degradation and improves the efficiency and precision of the algorithm.A complete SLAM algorithm based on particle filter is designed,and the algorithm is verified by the real data set collected by the mobile robot.Finally,build areal mobile robot hardware platform,for the robot equipped with laser ranging sensor,odometer sensor.Write mobile robot control program,the algorithm is implemented in C++ in the framework of robot operating system(ROS).Mobile robot in the real indoor environment movement perception of the surrounding environment,on-line realtime construction environment map at the same time autonomous positioning.The experimental data were analyzed and the experimental results were obtained. |
PDF Full Text Request