| In the simulation of the mobile robot navigation,it is important to reconstructing the real ground environment fast and accurately in unfamiliar environment.The mobile robot simulates the movement route and action in the reconstructed environment.After the simulation is successful,the mobile robot can be operated in practice and carry out various exploratory tasks.Through the comparison and selection of three-dimensional reconstruction method and camera calibration method,based on Metashape three-dimensional reconstruction software,the influence of different pixels on reconstruction model is compared.A series of two-dimensional images in real scene are collected by Canon 70D SLR camera.By comparing the influence of parameters on visual effect,modeling time and distance error of the model,the reconstruction parameters are selected and optimized.The causes of errors in the model are analyzed,and the reconstruction of real and high-precision small and large three-dimensional terrain environment is realized.The methods of collecting images and placing marker points are given.The methods of dealing with holes in the model are put forward,and the methods of reducing model errors are put forward.The method ensures that the model can be reconstructed with high accuracy and truthfulness in the shortest time.The error analysis and comparison of the distance between the marker point and the real marker point in the final small and large-scale scene reconstruction model are carried out.The results show that the method can not only construct a three-dimensional model with no holes,smooth surface,texture and high authenticity,but also the average relative error of the model is less than 0.1%.With the increase of the scale of the model,the average relative error of the model decreases.Through the validation and analysis of the model,it is proved that it can provide important ground data support for the three-dimensional simulation operation of mobile robots. |
PDF Full Text Request