Unmanned vehicle robust adaptive navigation positioning method based on improved graph optimization

By improving the graph optimization method and combining information from multiple sensors, a global optimization model was constructed and fault observation was isolated, which solved the problem of high-precision navigation for unmanned vehicles in satellite-denied environments, achieving high-precision navigation and positioning and improving robustness and adaptability.

CN121702371APending Publication Date: 2026-03-20THE NO 5311 FACTORY OF THE CHINESE PEOPLES LIBERATION ARMY
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-11-28
Publication Date
2026-03-20

AI Technical Summary

Technical Problem

Unmanned vehicles struggle to achieve high-precision navigation and positioning in satellite-denied environments. Existing technologies rely on satellite navigation systems and lack robustness and adaptability in complex environments.

Method used

An improved graph optimization method is adopted, which combines heterogeneous information from vehicle-mounted inertial sensors, wheel speed odometers, ultra-wideband (UWB) sensors, and lidar sensors to construct a graph optimization model. Global optimization estimation is performed using iterative least squares method, and residual fault detection algorithm is used to isolate fault observation information to achieve high-precision navigation for unmanned vehicles.

Benefits of technology

Achieving high-precision navigation and positioning in satellite-denied environments improves the robustness and adaptability of unmanned vehicles in complex environments, and can provide accurate attitude, speed and position information in real time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121702371A_ABST
    Figure CN121702371A_ABST
Patent Text Reader

Abstract

The invention discloses an unmanned vehicle robust adaptive navigation positioning method based on improved graph optimization. An inertia pre-integration factor node, a wheel speed odometer / kinematics model auxiliary factor node, an ultra wide band (UWB) distance observation factor node and a laser radar relative position constraint factor node in the graph optimization model are constructed by using observation constraint information provided by a vehicle-mounted heterogeneous navigation sensor. On the basis, an information fault of the observation sensor is discriminated in real time through an improved residual chi-square detection algorithm, and a self-adaptive fusion threshold value is set according to a fault detection result. And finally, global optimization estimation is carried out on the state of the unmanned vehicle in the graph optimization model through an iterative least square method based on a sliding window, high-precision robust fault-tolerant navigation positioning of the unmanned vehicle in a satellite denial environment is realized, and accurate navigation information, specifically including high-precision attitude, speed and position information, is provided for the unmanned vehicle in real time.
Need to check novelty before this filing date? Find Prior Art