Automatic Calibration Method For External Parameters Of Mobile Mapping System Based On Combination Of Multi-Feature Control And Constraint

The mobile mapping system(MMS)can quickly achieve the acquisition of high-precision and high-density three-dimensional spatial geographic information data.Today,global informatization has become a contemporary irreversible trend.With the development of technologies such as computers,sensors,network communications,"3S",MMS s are also making rapid progress and development.Mobile mapping technology continues to expand and penetrate into many application fields,and its accuracy is also facing increasingly demanding requirements.It is worth mentioning that the error of the external parameters of the MMS is one of its main sources of error,which directly affects the overall accuracy and quality of the laser point cloud.How to quickly and accurately calibrate the external parameters of the system is the premise and guarantee for it to obtain high-precision 3D spatial information data.Based on the positioning equation of MMSs and an analysis of the factors affecting their positioning accuracy,an error model for MMSs has been deduced.With reference to the traditional calibration method,this paper proposes a calibration method for external parameters of MMS based on the combination of multi-feature control and constraint.Main tasks as follows:1.For the external parameters between the laser scanner and the POS,an external parameters calibration method for MMS based on the combination of multi-feature control and constraint is proposed.This method simultaneously combines any one or more features that can be described by a certain mathematical model with an indefinite number of features.And the features with known real space position as the control conditions,and the unknown features as the constraints,Then,based on the Gauss-Helmert adjustment model,the conditions of all features are combined to construct the final normal equation to estimate the external parameters.2.Based on the e Random Sampling Consensus(RANSAC)algorithm,automatic feature extraction and matching is realized.That is,only need to give the observation data of the laser points cloud,the type of features it contains and the parameters of control features,the algorithm will automatically find and match homologous features that meet certain conditions,and automatically construct the normal equations to carry out the calibration of the external parameters of the MMS.It avoids the feature extraction through manual operation and realize the automatic calibration of external parameters.3.Using the Vsurs-E MMS as a platform,three calibration experiments based on multi-feature control,constraint,combination of control and constraint were designed,and an in-depth comparison was made from the aspects of method reliability,model parameter correlation analysis.The results show that the method based on the combination of multi-feature control and constraint not only achieves the complementary advantages of traditional methods,good practicability,and wide application range,but also further limits the correlation between the parameters in the calibration process through the combination of multiple features and the combination of control conditions and constraint.