汽车工程 ›› 2023, Vol. 45 ›› Issue (8): 1383-1391.doi: 10.19562/j.chinasae.qcgc.2023.08.009

所属专题: 智能网联汽车技术专题-感知&HMI&测评2023年

• • 上一篇    下一篇

智能网联汽车协同感知信任度动态计算与评价方法

朱冰,姜泓屹,赵健(),韩嘉懿,刘彦辰   

  1. 吉林大学,汽车仿真与控制国家重点实验室,长春 130022
  • 收稿日期:2023-02-22 修回日期:2023-04-06 出版日期:2023-08-25 发布日期:2023-08-17
  • 通讯作者: 赵健 E-mail:zhaojian@jlu.edu.cn
  • 基金资助:
    国家自然科学基金(U22A20247);吉林省科技发展计划项目(20220201023GX);吉林省自然科学基金(20210101057JC)

A Method for Dynamically Calculating and Evaluating the Trustworthiness of Collaborative Perception of Intelligent Connected Vehicles

Bing Zhu,Hongyi Jiang,Jian Zhao(),Jiayi Han,Yanchen Liu   

  1. Jilin University,State Key Laboratory of Automotive Simulation and Control,Changchun 130022
  • Received:2023-02-22 Revised:2023-04-06 Online:2023-08-25 Published:2023-08-17
  • Contact: Jian Zhao E-mail:zhaojian@jlu.edu.cn

摘要:

协同感知技术增强了智能网联汽车的感知性能,但协同网络中出现的异常车辆和恶意攻击信息会影响协同感知结果的真实性与有效性。针对该问题,本文结合车辆协同网络中的检测结果提出一种基于遮挡状态判别与检测有效性识别策略的智能网联汽车信任度动态聚合评价方法。仿真结果表明,本文提出的协同感知信任评价方法提升了协同感知车辆对检测结果的可信度,增强了智能网联汽车协同感知过程的鲁棒性,并实现了信任管理模型面对高信任车辆突发恶意攻击的动态识别。

关键词: 智能网联汽车, 协同感知, 动态聚合, 信任评价

Abstract:

Collaborative perception technology enhances the perceptual performance of intelligent connected vehicles, but abnormal vehicles and malicious attack information in the collaborative network can affect the authenticity and effectiveness of collaborative perception results. To address this problem, an intelligent connected vehicle trustworthiness dynamic aggregation and evaluation method is proposed in this paper, based on occlusion state discrimination and detection effectiveness identification strategy, combined with detection results in the vehicle collaborative network. Simulation results show that the proposed collaborative perception trust evaluation method improves the reliability of detection results for collaborative perception vehicles, enhances the robustness of the intelligent connected vehicle collaborative perception process, and realizes dynamic identification of sudden malicious attacks of high-trust vehicles by the trust management model.

Key words: intelligent connected vehicle, cooperative perception, dynamic aggregation, trust evaluation