数据补偿模型的训练方法、车辆控制方法、装置及设备

By mining vehicle driving and map navigation data from road network databases and training a data compensation model, the problem of data loss in complex scenarios for autonomous vehicles is solved, thereby improving vehicle safety and decision-making accuracy.

CN117079458BActive Publication Date: 2026-07-17APOLLO INTELLIGENT CONNECTIVITY (BEIJING) TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
APOLLO INTELLIGENT CONNECTIVITY (BEIJING) TECH CO LTD
Filing Date
2023-08-07
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Autonomous vehicles lack critical data in complex scenarios, leading to visual perception errors or misjudgments that affect safety. Existing technical solutions are costly and inefficient.

Method used

By acquiring vehicle driving data and map navigation data, feature mining is performed in the road network database based on multiple dimensions to train a data compensation model, which compensates for the lack of vehicle data in complex scenarios. Target experience data provided by the cloud is then used to determine vehicle control strategies.

Benefits of technology

It improves the accuracy and safety of autonomous vehicles' decision-making in complex scenarios, reduces data compensation costs, and enhances the continuity and experience of intelligent driving behavior.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117079458B_ABST
    Figure CN117079458B_ABST
Patent Text Reader

Abstract

本公开提供了数据补偿模型的训练方法、车辆控制方法、装置及设备。本公开涉及人工智能技术领域，尤其涉及数据挖掘、数据分析、自动驾驶等技术领域。具体方案为：获取第一数据集和第二数据集；基于多个维度在路网数据库中对第一数据集和第二数据集进行交通标识特征挖掘，得到第一目标特征集；基于多个维度在地图路网数据库中对第一数据集和第二数据集进行车辆行驶行为特征挖掘，得到第二目标特征集；基于第一目标特征集和第二目标特征集训练待训练模型，得到数据补偿模型。根据本公开的方案，能够通过对现有车载软件和硬件数据的充分挖掘得到经验数据，基于经验数据来弥补车辆在复杂场景下缺失的数据，从而有助于提高自动驾驶车辆的安全性。
Need to check novelty before this filing date? Find Prior Art