Integrating Distracted Driving Detection with Vehicle Platooning for Semi-Autonomous Driving in Smart Transportation

Engineering notes

Key topics: autonomous driving, carla, deployment, control. See the paper for implementation details and experimental results.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。

Original abstract

This paper presents a novel integrated control framework that combines Distracted Driving Detection (DDD) with adaptive Vehicle Platooning (VP) to enhance both road safety and platooning performance. The proposed system uses real-time computer vision techniques to detect distraction indicators viz. non-frontal head pose, prolonged eye closure, and hands off the steering wheel-and dynamically adapts platoon behavior using a feedback control loop. Vehicle-to-vehicle (V2V) communication enables cooperative decision-making, allowing the system to initiate role reassignment, adjust inter-vehicle spacing, or trigger safe mode transitions in response to driver inattention. Unlike conventional platooning methods that assume fully attentive drivers, our framework actively monitors cognitive states and embeds safety-aware responses into the control strategy. Experimental validation using the CARLA and SUMO simulators was conducted over a $\mathbf{3 0}$-minute simulation period with platoon sizes of 5 vehicles. Results show a 27 % reduction in collision rates, from 0.14 to 0.10 collisions per vehicle kilometer, improved platoon stability, and increased fuel efficiency under distraction-induced scenarios. These findings demonstrate the system's robustness and practical viability for deployment in semi-autonomous and mixed-autonomy traffic environments.

Links and sources

• Official / arXiv page

Comments