The boomerang effect in the traffic oscillation growth of commercial automated vehicles: theoretical analysis and experimental evidence

As the deployment of commercial automated vehicles increases, significant attention is being paid to their traffic flow characteristics, and evaluating their impact necessitates an appropriate modeling approach. Previous studies on the traffic oscillation growth pattern of automated vehicles often perform stability analysis assuming stationarity. However, in real traffic, achieving stationarity can take considerable time. Hence, studying the temporal dimension is important. In this paper, we consider the temporal dimension in analyzing the growth pattern of traffic oscillations. First, we theoretically demonstrate that the temporal dimension induces transient metastability, whereby the vehicle platoon exhibits unstable behavior over finite time intervals and becomes stable only after a sufficiently long time. Under certain perturbation frequencies, transient metastability causes oscillations to damp near the front of the platoon, while lower-frequency transient oscillations can emerge from a certain vehicle index toward the rear of the platoon. We term this phenomenon the “boomerang effect”. Second, we conducted a vehicle platoon experiment subjecting commercial vehicles equipped with longitudinal driving automation to perturbations of varying frequency and amplitude. The experiment was designed to study the platoon’s response to different perturbation characteristics, and to assess the theoretical prediction of the boomerang effect. The experimental results confirm that: (i) oscillation growth is strongly dependent on perturbation characteristics; (ii) the boomerang effect occurs within a specific frequency range as predicted theoretically. This work provides new insights into the dynamics of automated vehicle platoons, and offers direct implications for the design of strategies to improve traffic stability.