A neural network assisted vehicle adaptive autonomous navigation method

By using a neural network-assisted multi-parameter parallel prediction LSTM model and variational Bayesian filtering, the error accumulation problem of vehicle inertial navigation in environments without satellite signals was solved, achieving fully autonomous and high-precision navigation for vehicles.

CN119779296BActive Publication Date: 2026-01-09BEIJING AUTOMATION CONTROL EQUIP INST
2 Cites 0 Cited by

Patent Information

Application Number
CN202411867627.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-12-18
Publication Date
2026-01-09
Estimated Expiration
2044-12-18

AI Technical Summary

Technical Problem

Existing vehicle-mounted inertial navigation systems suffer from severe error accumulation in environments without satellite signals, making it difficult to achieve high-precision autonomous navigation.

Method used

A neural network-assisted multi-parameter parallel prediction LSTM model, combined with variational Bayesian filtering, is used to achieve adaptive integrated navigation of inertial navigation data. The trained model can perform high-precision prediction of vehicle attitude and speed when GNSS is ineffective.

Benefits of technology

It enables fully autonomous, high-precision navigation of vehicles in the absence of GNSS, significantly reducing the accumulation of inertial navigation errors, and is suitable for scenarios where satellite signals fail, such as tunnels and building obstructions.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application provides a neural network assisted vehicle adaptive autonomous navigation method, which comprises the following steps: designing an improved LSTM model for multi-parameter parallel prediction; collecting inertial measurement unit data and performing inertial navigation calculation; performing inertial / satellite integrated navigation calculation when satellite navigation is effective; training the designed parallel LSTM model based on the integrated navigation result; predicting vehicle speed and attitude information based on the trained machine learning model when satellite navigation is ineffective; and using variational Bayes based on the prediction result of the machine learning model to perform adaptive integrated navigation when satellite navigation is ineffective. The application overcomes the problem of insufficient cumulative accuracy of vehicle navigation error when only relying on inertial navigation without reference information, and can realize full autonomous and high-precision navigation and positioning of the vehicle, which has important practical application value.
Need to check novelty before this filing date? Find Prior Art