Three-dimensional vehicle-mounted positioning navigation method based on coordinate transformation

Abstract

The invention discloses a three-dimensional vehicle-mounted positioning navigation method based on coordinate transformation. The method comprises the steps that firstly, a plurality of nonlinear motion models are established; secondly, mixing estimation is conducted on probabilities of all the motion models of a vehicle, wherein each motion model calculates a respective initial filter value separately; thirdly, a unscented Kalman filter is merged, a GPS observation value is added, the filter output of all the models is calculated in parallel, the probability of each model is updated after filtering is finished, the model most suitable for current vehicle movement is found for output interaction, and total state estimation and total covariance estimation are finally obtained; finally, whenthe vehicle is in a slope state, two different coordinate systems of a plane and a slope are established by utilizing a coordinate transformation method, a relationship of vehicle positions before and after three-dimensional coordinate transformation is obtained, and height information of the vehicle is calculated. The method actually and effectively improves the precision of positioning the vehicle in the height direction and the overall positioning precision and has high practical value.

Description

technical field [0001] The invention relates to a three-dimensional vehicle positioning and navigation method based on coordinate transformation. The vehicle positioning height information obtained by using this method can ensure the basic height positioning requirements of the vehicle. Combining with GIS, the vehicle can be positioned to a specific lane, which effectively solves the complex problem. The problem of vehicle positioning errors under overpasses and other road conditions not only improves the positioning accuracy of the vehicle height direction, but also improves the overall positioning accuracy of vehicle navigation, which can be used in complex road conditions in three-dimensional space. Background technique [0002] Under realistic conditions, most of the actual systems are nonlinear, and the system equations and observation equations of the vehicle maneuvering model are also nonlinear. The unscented Kalman filter determines the optimal gain matrix based on t...

AI Technical Summary

Problems solved by technology

[0004] The traditional interactive multi-model unscented Kalman filter has a good filtering effect on the plane motion of the vehicle, but the positioning accuracy error of the height information is large, and it cannot completely distinguish which layer the vehicle is located on the complex overpass, which is prone to navigation map matching Phenomenon of actual road conditions

Due to error factors such as GPS receiving signals and the conversion of altitude and ground height, the error of vehicle height data given by GPS is too large, ranging from a few meters to tens of meters. It is impossible to use such height data for filtering, and the obvious effect is impossible. Too accurate, the problem of too much vehicle height positioning error still exists in real life

Images