论文部分内容阅读
驾驶人潜在危险预知能力(APRP)对道路交通安全有重要影响,国内相关的理论研究及其训练系统相对滞后。首先,研究驾驶人APRP的内在机理与表现特征,建立评估驾驶人APRP的算法模型,其中评估指标为潜在危险的响应时间ti与识别准确度Yi这2个参量。最终,建立了一个基于真实交通场景视频数据库的APRP评估系统。APRP测试结果表明:新手驾驶人(ID)与熟练驾驶人(ED)在与交通事故的发生概率及其严重程度密切相关的ti与Yi这2个指标上存在显著差异。根据测试结果,讨论新手驾驶人与熟练驾驶人APRP存在差异的原因,同时验证APRP评估指标与算法模型的合理性与有效性。APRP评估体系可应用于驾驶人的主动安全意识教育与训练,用以提高驾驶人的安全意识与驾驶能力。
The Predictability of Driver Potential Hazard (APRP) has a significant impact on road traffic safety, and the relevant domestic theoretical research and its training system lag behind. First of all, the intrinsic mechanism and performance characteristics of driver APRP are studied, and an algorithm model for evaluating driver APRP is established. The evaluation index is two parameters: potential response time ti and identification accuracy Yi. Finally, an APRP evaluation system based on the real traffic scene video database is established. The APRP test results show that there are significant differences between the novice driver (ID) and skilled driver (ED) on ti and Yi, which are closely related to the probability and seriousness of traffic accidents. According to the test results, the reasons for the difference between novice drivers and skilled drivers APRP are discussed, and the rationality and validity of APRP evaluation indicators and algorithm models are verified. The APRP assessment system can be applied to driver’s initiative safety awareness education and training to improve drivers’ safety awareness and driving ability.