Research On Seamless Switching Indoor-outdoor Positioning And Navigation Of Firefighting Robot

The main purpose of using firefighting robot is to replace the firefighter in firefighting,reduce casualties and improve the efficiency of rescue.In this paper,according to the positioning and navigation needs of the firefighting robot,a method of indoor and outdoor seamless switching positioning and navigation was designed.The main contents of this paper are listed as follows:According to the particularity of the firefighting robot using scene,a mechanical structure of the firefighting robot that can work in extreme environment was designed.The sensor selection of the firefighting robot was carried out,the overall architecture of the positioning and navigation system was designed,and the movement model of the firefighting robot was studied.Through the SLAM(Simultaneous Localization and Mapping)algorithm under the ROS system,the firefighting robot can recognize the environment.The advantages and disadvantages of Hector-SLAM algorithm and Rao-Blackwellized-SLAM algorithm were studied,the thesis used RBPF-SLAM algorithm,considering the characteristics of firefighting robot using scenarios and hardware modules.Aiming at the problem of particles dilution and particles diversity shortage of RBPF-SLAM algorithm,an improvement was made.The effectiveness of the improved RBPF-SLAM algorithm was verified by simulation,and the reliability of the improved algorithm was verified by performing experiment in the experimental scene.A UWB/GPS(Ultrawide Band / Global Positioning System)indoor and outdoor seamless switching positioning assistant system was designed to improve the positioning accuracy of the firefighting robot.The positioning system used GPS/IMU(Global Positioning System / Inertial Navigation System)combined positioning in the outdoor environment,and switches to UWB/IMU combined positioning in the indoor environment where GPS signals cannot be sensed.At the same time,the indoor and outdoor positioning seamless switching algorithm based on the threshold mechanism and the positioning accuracy of multiple sensors was designed,the system can achieve the most accurate positioning in the current region in the process of movement,and at the same time,the problems of uneven switching and ping-pong effect are improved.The effectiveness of the algorithm is verified by experiments on a firefighting robot.On the basis of environment identification and initial localization,two path planning algorithms were adopted to realize the navigation of the firefighting robot.Two global path planning algorithms,A* and Dijkstra,were studied.A* algorithm with low computational complexity and high computational efficiency was selected through comparison of simulation results,and DWA algorithm was used for local path planning of firefighting robot.The fusion path planning navigation algorithm was implemented on the simulation software and the experimental platform,and the obstacle can be dynamically avoided during the navigation process.The firefighting robot finally reaches the target point,which verifies the reliability of the navigation algorithm.