| With the rapid development of computer technology and artificial intelligence technology and people's living standards,robot technology is developing rapidly and indoor mobile robot has attracted more and more attention.Indoor mobile robot can complete the functions like cleaning,restaurant service and patient care,and these functions can't be separated from the basic function of autonomous navigation of robot.Therefore,SLAM(Simultaneous Localization And Mapping)and path planning play an important role in the research of mobile robots.In this paper,the SALM problem of indoor mobile robot is studied.First of all,the status of research and development of SLAM field is analyzed,and the methods of theory research,sensor and map description commonly used in this field are summarized.Secondly,we designed and manufactured a wheeled mobile robot,and completed the construction of its control system hardware platform.We mainly focus on the Advantech PCM-3363D,Arduino Mega 2560,motor driver,inclinometer,encoder,PS2 handle etc.The software part of the robot control system uses the Ubuntu 14.04 operating system,and all the programs are developed on the ROS platform.Thirdly,the Kinect sensor is introduced,its principle of obtaining depth image and coordinate transformation of camera are described.The significance and method of using RGB-D camera to return depth data as laser sensor are explained.Then,the algorithm of robot SLAM is researched.The location and construction map of the robot adopts modified Rao-Blackwellized Particle Filter(RBPF)method.The surrounding environment model is expressed by grid map.Dijkstra method and dynamic window method(DWA)to find the optimal trajectory path.Finally,We experimented with robots.In this environment,the robot can locate and construct the map in real time,and can complete the autonomous navigation based on the known map.The experimental results show that the indoor mobile robot can complete the SLAM function and achieve the desired results. |
PDF Full Text Request