Summary
Predicting how groups of cars follow each other, or platoons, is important for safe and smooth driving. Previous prediction methods focused mainly on accuracy but sometimes failed to model how disturbances, like sudden braking, realistically spread through the platoon. The authors developed a new AI model called SSP-DMGTimeNet that uses physical rules about how disturbances travel between cars to make better predictions. This model not only predicts individual car movements accurately but also keeps the whole platoon stable by preventing unrealistic amplifications of disturbances. Their tests show that including physical constraints helps balance accurate predictions with realistic, stable traffic behavior.
vehicle platoontrajectory predictionstring stabilityspatiotemporal modelingcausal attentiondisturbance propagationmulti-scale temporal representationphysics-constrained learninghead-to-tail amplification
Authors
Yuhang Wang, Kailang Ma, Zirui Li, Mingfeng Fan, Kitae Jang, Changju Lee, Heye Huang
Abstract
Existing car-following prediction methods mainly optimize trajectory accuracy, while rarely considering whether predicted disturbances propagate realistically along a vehicle platoon. This limitation may lead to accurate but string-unstable predictions. We propose SSP-DMGTimeNet, a physics-constrained learning framework for spatiotemporal trajectory prediction of vehicle platoons. The model combines multi-scale temporal representations with cross-vehicle interaction features to capture complex and time-varying platoon dynamics. A propagation-delay-aware causal attention mechanism explicitly models upstream-to-downstream disturbance propagation by learning response delays between adjacent vehicles and accumulating them along the platoon. In addition, time- and frequency-domain string-stability losses relieve disturbance amplification across both adjacent vehicles and arbitrary sub-platoons during training. Experiments on HighD show that SSP-DMGTimeNet achieves an unstable-window rate of 0.65\% for five-vehicle platoons and a maximum head-to-tail amplification of 0.898 on the ground-truth excitation subset, while maintaining competitive trajectory prediction performance. In zero-shot evaluation on NGSIM US-101 and I-80, the model achieves velocity MAEs of 1.316~m/s and 1.252~m/s, with unstable-window rates of 3.90\% and 4.10\%, respectively. These results demonstrate that incorporating platoon-level physical constraints can effectively balance trajectory prediction accuracy and disturbance propagation stability.