| IIWith the continuous progress of urbanization,large-scale storage,pipe gallery,basement and other building facilities have become the basic components of modern cities.At present,the construction of environmental map based on manual mapping and scanner technology has low efficiency,which cannot meet the needs of timeliness of data collection.Aiming at the above problems,an autonomous mobile surveying and mapping robot and a 3D mapping system are designed based on the robot SLAM theory in this thesis.The Kalman filtering method is used to integrate the data of ultra-wideband and odometer sensors to achieve accurate indoor positioning of the robot.Based on the factor graph model,the lidar and the back-end of the ultra-wideband odometer are jointly optimized for mapping,and the indoor high-precision 3D scene map is constructed.The main research contents and achievements of this thesis are as follows:(1)According to the actual needs of indoor architectural surveying and mapping operations,first of all,determine the function of the robot surveying and mapping system;Secondly,based on the theoretical basis of robot synchronous positioning and mapping technology,the hardware scheme and software scheme of the surveying and mapping robot are designed respectively.;Finally,an autonomous surveying and mapping robot and 3D mapping system for indoor buildings was builtled,which provides an experimental platform for the robot positioning and mapping algorithm research.(2)In order to solve the problem that the surveying and mapping robot lacks reliable positioning information in the indoor environment,firstly,the indoor high-precision positioning method based on ultra-wideband positioning is studied,and indoor static positioning experiments are carried out;Secondly,the odometry positioning used by the surveying and mapping robot is calibrated,and the positioning accuracy of the robot odometer is improved by55.6% compared with that before calibration.Finally,a method based on ultra-wideband fusion odometry positioning is adopted.The odometer positioning data is used as the system input,and the ultra-wideband positioning information is the observation data.The extended Kalman filter and the unscented Kalman filter are used for data fusion to realize the ultra-wideband with odometer fusion positioning.Through comparative experiments,it is verified that the proposed method can effectively improve the positioning accuracy of the robot and achieve the accurate positioning of the surveying and mapping robot in long-term mapping operations.(3)For the construction of high-precision 3D environmental map of indoor buildings,the LOAM algorithm based on solid-state lidar is improved,and the back-end optimization of factor graph based on ultra-wideband odometry factor,lidar odometry factor,and loopback constraint factor is designed.The mapping module can realize real-time positioning of 3D scenes and lowdrift mapping functions.(4)In order to verify the practicability and feasibility of the algorithm in this thesis,the surveying and mapping robot is used as the experimental platform to conduct robot positioning and mapping tests in indoor corridors and laboratory scenes.The experimental results show that the positioning accuracy of the optimized mapping algorithm proposed in this thesis is within20 cm in the average value of the two scenes,and it can realize the construction of low-drift 3D scene maps.Through the research on the surveying and mapping robot and the method of positioning and mapping,it provides a feasible solution for the autonomous positioning and high-precision map construction of the autonomous surveying and mapping robot in the indoor complex structural environment. |
PDF Full Text Request