Page 43 - 《软件学报》2025年第12期

5424 软件学报 2025 年第 36 卷第 12 期

Sciences, Beijing 100190, China)
4
(Software Security and Trustworthiness Laboratory, Nanjing Institute of Software Technology, Nanjing 211135, China)
5
(University of Chinese Academy of Sciences, Beijing 100049, China)
Abstract: With the rapid development of autonomous driving technology, the issue of vehicle control takeover has become a prominent
research topic. A car equipped with an assisted driving system cannot fully handle all driving scenarios. When the actual driving scenario
exceeds the operational design domain of the assisted system, human intervention is still required to control the vehicle and ensure the safe
completion of the driving task. Takeover performance is an extremely important metric for evaluating a driver’s performance during the
takeover process, which includes takeover reaction time and takeover quality. The takeover reaction time refers to the time from the
system’s takeover request to the driver’s control of the steering wheel. The length of the takeover response time not only reflects the
driver’s current state but also affects the subsequent handling of complex scenarios. Takeover quality refers to the quality of manual
vehicle operation by the driver after regaining control. This study, based on the CARLA driving simulator, constructs 6 typical driving
scenarios, simulates the vehicle control takeover process, and collects physiological signals and eye movement data from 31 drivers using a
multi-channel acquisition system. Based on the driver’s takeover performance, and regarding International standards, an objective takeover
performance evaluation metric is proposed, incorporating the driver’s takeover reaction time, maximum horizontal and vertical
accelerations, and minimum collision time, derived from multiple vehicle data. By combining driver data, vehicle data, and scenario data, a
deep neural network (DNN) model predicts takeover performance, while the SHAP model analyzes the impact of each feature, improving
the model’s interpretability and transparency. The experimental results show that the proposed DNN model outperforms traditional machine
learning methods in predicting takeover performance, achieving an accuracy of 92.2% and demonstrating good generalization. The SHAP
analysis reveals the impact of key features such as heart rate variability, driving experience, and minimum safe distance on the prediction
results. This research provides a theoretical and empirical foundation for the safety optimization and human-computer interaction design of
autonomous driving systems and is of great significance for improving the efficiency and safety of human-vehicle cooperation in
autonomous driving technology.
Key words: autonomous driving; takeover performance; deep neural network (DNN); interpretability; human-machine interface
自动驾驶技术的迅猛发展使得人车合作变成了更加现实、紧迫的课题 [1] . 根据 SAE 分级 [2] , 可以将自动驾驶
分为 6 个等级, 其中 L0 为手动驾驶, L5 为完全自动驾驶, 当前可应用在真实交通场景下的自动驾驶等级一般在
L3 以下. 自动驾驶汽车联合摄像头、GPS、雷达和激光雷达等传感器感知环境数据, 通过规划模块规划行驶路线,
最后执行模块执行相应的驾驶动作. L3 为有条件自动驾驶, 其辅助驾驶系统 (advanced driving assistance system,
ADS) 在驾驶过程中允许驾驶员暂时解放双手、减轻驾驶负荷, 执行一些阅读、饮食等与驾驶无关的任务 (non
driving related task, NDRT) [3,4] . 受限于当前自动驾驶技术和法律规定, 在超出自动驾驶系统的操作设计域 (operation
design domain, ODD) 时需要驾驶员对车辆进行接管并驾驶, 驾驶员仍旧是驾驶过程中重要的一环. 在这种情况下,
自动驾驶系统需要发出接管请求 (take over request, TOR), 并通过视觉、听觉或触觉的方式提醒驾驶员, 以便驾驶
员能及时接管车辆保障行车安全 [5] . 在一些安全攸关的场景下 (如即将发生碰撞、礼让行人等), 驾驶员的接管行
为的好坏对驾驶安全有着至关重要的影响 [6] . 因此, 评估并预测驾驶员的接管行为, 确保驾驶员再次从事非驾驶任
务时能够安全地接管车辆控制权具有重要意义.
评价接管行为的好坏主要通过接管绩效进行衡量, 通常来讲, 接管绩效主要包括接管反应时间和接管质量两
方面. 接管反应时间是指系统发出接管请求后到驾驶员控制方向盘的时间开销, 接管反应时间长短不仅一定程度
上反映了当前驾驶员的状态, 还对后续面对复杂场景进行操作也有一定影响. 接管质量是指驾驶员获得车辆控制
权后手动驾驶车辆的质量. 近年来, 面向有条件自动驾驶情况下的接管研究大多专注于探究不同的接管提醒方
式 [7] 、接管提醒时间 [8] 、接管场景 [9] 对接管质量的影响, 对接管质量的绩效研究相对较少. 部分学者对接管质量进
行了预测但并未对影响接管质量的因素进行分析 [10] . 同时, 接管质量的衡量目前还不具备统一的标准, 因而亟需
对接管绩效进行判定, 并探究不同接管因素对接管绩效的影响.
在自动驾驶领域, 接管数据的获取面临着多重挑战, 不仅涉及复杂的多通道数据采集设备的需求, 还需实现这
些设备与仿真平台之间的无缝对接, 以确保人-机-设备间的时间同步精度. 此外, 鉴于该领域的特殊性, 当自动驾
驶系统遇到难以自行解决的驾驶情境时, 会发出接管请求, 这对驾驶员快速且准确的接管能力提出了极高的要求.