Design And Implementation Of Multi-sensor Integrated Navigation System For The Indoor Environment

Global navigation satellite system(GNSS)is one of the most commonly used navigation methods of UAV and UGV.However,the signal of GNSS is relatively weak in an indoor environment.In order to make up for this defect,it is necessary to use visual SLAM technology to build integrated navigation system to achieve high-precision navigation of UAV or UGV in an indoor environment.Meanwhile,the construction of environmental map is also an indispensable function for UGV to conduct autonomous navigation in an indoor environment.Therefore,it is of great value and significance to study the problems of high-precision integrated navigation and rapid environment map construction suitable for an indoor environment.This thesis mainly studies the mapping of mobile robot,autonomous positioning and path planning in an indoor environment.The main research contents are as follows.For the design of multi-sensor indoor integrated navigation system and the demand for high-precision integrated navigation system suitable for indoor environment,this thesis designs the overall scheme of the system and puts forward the corresponding index requirements.According to the design scheme,the hardware platform of multi-sensor indoor integrated navigation system is built by using the robot development platform and lidar sensors,and the software architecture of the system is designed based on the robot operating system(ROS)on this platform.An environment map construction algorithm based on laser data filtering is proposed.In view of the problem that the calculation amount of the existing laser slam mapping algorithm is too large at present,this thesis proposes a method of laser data filtering,which eliminates the redundant part of original laser data by taking the scanning density of the laser data as the filtering criteria and can reduce the amount of data input by the mapping algorithm and improve the execution efficiency of the mapping function of the system.The validity of the laser data filtering method is verified by using the Cartographer technology of Google.In order to realize the high-precision navigation of UGV in an indoor environment on the basis of the research on the visual inertial mileage calculation method with tight data coupling,this thesis studies the visual inertial odometry technology based on visual SLAM,and proposes an algorithm of time stamp alignment processing between IMU and camera.After testing,the positioning error of the integrated navigation system is better than 10 cm.On the basis of the above,this thesis further uses the bidirectional a * algorithm and the Dynamic Window Approach as the global and local path planning methods respectively,and develops a robot system integrating the software and hardware platform,map construction,autonomous positioning and path planning based on the consideration of the obstacle avoidance requirements of mobile robot in an indoor environment.The algorithm proposed in this thesis is implemented on this platform,and related experimental verification is carried out.Experimental results show that the mobile robot system has the functions of environment map construction,indoor autonomous navigation and obstacle avoidance,and all the performance indicators meet the index requirements of the system design.