Outdoor Mobile Robot Localization Based On Road Informatioin

Global Positioning System(GPS) is now widely used for outdoor mobile robot localization. However, since the satellite signal is easily affected by the external environment, such as weather, building block, it sometimes can not provide ideal positioning effect. The use of laser scanners, vision sensors for outdoor mobile robot localization is the current hotspot. However, these methods typically have shortcomings such as large computation, easily affected by the environment and expensive cost.In response to these problems, an outdoor mobile robot localization method based on the road information is proposed. This subject extracts road information from geographic information systems and makes use of coordinate transformation algorithms and robots terrain inclined model(RTI) for the extracted road information to get the road map. The attitude of the robot, speed and GPS data are measured by sensors in real time when the robot is traveling on the road. Then, the particle filter is used to fuse the measurement data and the above-described road map to get the position of the robot.This paper describes the experimental platform. The IMU module was designed to measure the attitude of the robot. Meanwhile, due to the widespread use and critical acclaim of Android smartphones, the Android phone app was developed for the subject to receive sensor data, live run localization algorithm and show the location of the robot on the map.Three experiments were designed to verify the location algorithm for this subject. Roads of different type and length were selected for experiments. One experiment was mainly used to verify the localization algorithm performance and to analyze the impact of the algorithm and initial positioning error of positioning errors; the second experiment was used to verify the localization algorithm can still provide accurate position of the robot on roads of long distance and different type; the third experiment was mainly used to validate the algorithm performance in a complex urban road environment including minor arterial, service way, tunnels and urban canyons. A detail analysis of the experimental results was made in this paper to show that the algorithm can get the ideal positioning effect on long-distance localization and complex urban road conditions.