A small scene high-precision map construction method

By combining LiDAR and inertial measurement units with autoware tools to build high-precision maps for small scenes, the problem of high cost and poor flexibility in high-precision map production has been solved, achieving low-cost, high-precision map construction and improving the flexibility and safety of autonomous driving systems.

CN116182833BActive Publication Date: 2025-12-16BEIJING INST OF TECH +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202211672134.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-12-26
Publication Date
2025-12-16
Estimated Expiration
2042-12-26

AI Technical Summary

Technical Problem

The production cost of existing high-precision maps is high and the process is cumbersome, making it difficult to tailor-make small-scale autonomous driving tasks, which makes it difficult to implement autonomous driving.

Method used

Scene mapping is performed using LiDAR and inertial measurement units, and point cloud maps are annotated using Autoware tools to construct a road network model, including point cloud layers, road topology layers, obstacle layers, and vehicle model layers, achieving lane-level navigation and centimeter-level positioning accuracy.

Benefits of technology

It enables the low-cost construction of high-precision small-scene maps, improves the flexibility and safety of autonomous driving systems, reduces the computational load of perception systems, and ensures the accuracy of positioning and the rationality of trajectory planning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116182833B_ABST
    Figure CN116182833B_ABST
Patent Text Reader

Abstract

The application relates to a small-scene high-precision map construction method, which comprises the following steps: using a small vehicle carrying a laser radar and an inertial measurement unit to carry out scene mapping, and acquiring a point cloud map of the scene; labeling the point cloud map based on a labeling tool, and constructing a road network model to complete the construction of the scene. The method is suitable for any scene, and after collecting environmental data, a high-precision map is manually drawn, which is highly flexible and can effectively assist the work of each sub-module of an automatic driving system.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Three-dimensional high-precision map generation method and device

    CN109993780A

  • Large-scale park mapping positioning method based on multi-sensor touch

    CN112967392A