A jam-absorption driving system for reducing multiple moving jams by estimating moving jam propagation

Abstract

Jam-absorption driving (JAD) is a novel connected and automated vehicle-based (CAV-based) control strategy that uses either a single or several absorbing cars to clear moving jams. However, a practical control system has not yet been developed. This study presents a real-time control system for operating JAD against multiple moving jams on freeway sections under a partial CAVs and heterogeneous (mixture of cars and trucks) traffic environment. The system dynamically predicts absorbing end points by estimating the propagation of multiple moving jams without using a traffic model. Subsequently, it selects suitable CAVs as absorbing cars in traffic via vehicle-to-infrastructure communication to absorb moving jams. A control model embedded with complete/synergetic JAD strategy decision-making and moving jams–CAVs matching process was developed to guarantee moving jams can be properly absorbed as early as possible. Simulation results demonstrated that the proposed JAD system successfully reduced successive multiple moving jams in the toy model (with two artificial successive moving jams) and hypothetical bottleneck (with series moving jams generated by a speed limit zone) scenarios. Moreover, sensitivity analysis indicated that the JAD system effectively reduced moving jams and enhanced traffic performance (e.g., about 10% fuel consumption reduction) under approximately 1% to 10% CAV market penetration rates (MPRs). This implies that the proposed JAD system can be implemented at low CAV MPRs in the future.