Vehicle-mounted navigation method and device based on machine learning and computer equipment

Abstract

The invention relates to a vehicle-mounted navigation method and device based on machine learning and computer equipment. The method comprises the steps of: building a VDM-assisted IMU error correction model, building a BDS/IMU/VDM integrated navigation model, building a machine learning training model when the BDS is in a normal navigation state, taking the position and speed obtained by mechanical arrangement and VDM correction of an IMU as a training input sample, taking the BDS/IMU/VDM combined filtering output as an output sample, and when the BDS is in an unlocked state, starting a machine learning model which is matured by training to predict a filtering output value, and correcting the position and speed information of IMU/VDM calculation. Through test in a tunnel, an underground garage and a satellite signal shielding area, a simplified machine learning network can be used for establishing auxiliary navigation information and realizing real-time auxiliary inertial navigation/wheel speed updating, so that long-time positioning error divergence is avoided, and the navigation and the positioning precision of the system is greatly improved.

Description

technical field [0001] The present application relates to the technical field of surveying, mapping and navigation, and in particular, to a vehicle navigation method, device and computer equipment based on machine learning. Background technique [0002] GNSS / INS integrated navigation technology is a widely used navigation and positioning technology, especially in the field of vehicle navigation and automatic driving. In an open environment, GNSS can provide continuous high-precision navigation and positioning information. Once it enters the sheltered area, the navigation performance drops sharply; inertial navigation can provide short-term high-precision positioning information, effectively making up for the vehicle navigation and positioning performance in the GNSS loss-of-lock state. However, when the satellite loses lock for a long time, such as entering a long tunnel, a narrow tree hole, or an urban canyon, the inertial navigation positioning error accumulates over time....

AI Technical Summary

Problems solved by technology

In recent years, with the rapid development of artificial intelligence, intelligent algorithms have gradually moved from theory to application. Using the powerful self-learning and prediction capabilities of intelligent algorithms, they can make up for the missing information of GNSS and obtain virtual observations of GNSS lost locks. However, the current intelligent The algorithm also has the disadvantages of low operating efficiency and difficulty in ensuring real-time performance

Images