LAUSR

Platooning has the potential to significantly improve traffic flow, safety, and fuel consumption. In the foreseeable future, traffic will still be composed mainly of human-driven vehicles (HDVs). Thus, the importance of exploring mixed platoons consisting of automated vehicles (AVs) and HDVs. Mixed platoons control is challenging since HDVs incorporate major uncertainty and stochastic nature. In this paper, human driving stochastic behavior was analyzed. The findings of this analysis were incorporated in all aspects of controller development – modeling, estimation, and controller synthesis. An algorithm to find an optimal controller that attenuates the effect of HDVs stochasticity while keeping predefined safety constraints was developed. Using this algorithm, two $\mathcal {H}_{\infty }$ controllers are suggested: a robust controller and a gain-scheduling (GS) controller. The controllers were simulated using an information flow topology (IFT) that was suited for mixed-platoons. Results show that both controllers attenuates HDVs stochasticity. Compared to the robust controller, the GS controller improves performance mainly at high velocities.