Design Of Integrated Navigation System Based On Multi-Sensor Fusion

At present,the navigation modes of autonomous mobile robots at home and abroad can be divided into Electromagnetic Navigation,Optical Navigation,Inertial Navigation,GPS Navigation,Laser Navigation,Visual Navigation.All kinds of navigation modes have certain advantages and disadvantages.Inertial Navigation System has become a necessary navigation mode for various mobile platforms because of its superior performance.The research of Integrated Navigation System has a long history at home and abroad.At present,it mainly focuses on IMU,GPS,Vision,Magnetometer,Lidar and other sensors.The basic idea of integrated navigation is to make full use of the information provided by various sensors to complement each other,so as to improve the accuracy and robustness of navigation system.Integrated Navigation System adopts two or more different navigation means.According to the characteristics and advantages of different navigation systems.each navigation information data is fused to obtain a more accurate navigation system.The key technology is multi-sensor fusion strategy.Multi-sensor fusion refers to the optimization and processing of multi-sensor data to obtain the estimated and real values of the system.At present,the information fusion technology of integrated navigation system mainly includes track deduction,Kalman filter,Bayesian estimation,genetic algorithm,neural network,support vector machine and so on.Modern filter has good performance,but its stability and robustness are insufficient.Algorithm complexity and calculation are large,and the hardware requirements of the platform are high.It can't show its superior performance on the low-end platform,but it is difficult to realize on the general platform because of its complex logic structure and design ideas.In Integrated Navigation System,the classical method is the most widely used Kalman Filter algorithm,Extended Kalman Filter,Unscented Kalman Filter and other algorithms.This paper is devoted to the implementation of integrated navigation system based on multi-sensor fusion,and actively explores the solution of indoor mobile robot navigation system.On the basis of in-depth research and practice,the implementation method of integrated system integrating Lidar and inertial data is proposed.Lidar is a means to construct environment map,and the position and attitude information of navigation platform is analyzed by means of ICP laser scanning matching algorithm.Inertial sensors are the main means of navigation parameters calculation.The position and attitude parameters calculated by Lidar and inertial sensors are fused with Extended Kalman Filter.The integrated navigation system with excellent performance is designed and implemented.Finally,the comparative experiments on the indoor mobile robot platform designed by ourselves show that the integrated navigation system indeed has a certain advantage in terms of precision and stability.