Independence localization system for Mecanum wheel AGV

This study deals with the independent localization system of AGV which is sensor fusion strapdown inertial navigation system and laser navigation to develope more fast and accurate for the AGV. the strapdown inertial navigation system is very expensive and has large system. Hence, we propose the independent and stable localization system with sensor fusion using a low-cost strapdown inertial navigation system which consists of MEMS inertial sensor. To reduce errors of the strapdown inertial navigation system, we performed compensation using proportional method to gyro sensor and low-pass filter to accelerometer. We used transformation matrix with a result of attitude reference using Kalman filter. Finally, to compensate the result of error, fuzzy inference system is used. To verify the performance of the proposed system, experiments is performed with a Mecanum wheeled AGV and a forklift AGV. In the experiments result, we confirmed that the proposed system can estimate the position.