Efficient control of vehicles in congested traffic using model predictive control

Traffic management on road networks is emerging in the control engineering due to the strong demand for alleviation of traffic congestion in urban areas. Complex interacting behavior of vehicular traffic is one of the reasons of congestion in dense urban traffic. In congested traffic, waves of vehicle density propagate backward to the following vehicles as drivers want to keep safe clearances with frequent acceleration and braking. This paper presents a novel method aiming to alleviate congestion, save energy and reduce exhaust gas emissions by controlling an individual vehicle in a systematic way. In particular, model predictive control is used to predict the future states of the preceding traffic and regulate a safe head distance under the bounded driving torque condition of the vehicle. Numerical simulation shows that a vehicle with the proposed control scheme effectively attenuates the propagation of jamming waves and improves its energy consumption, and as a result, the following vehicles also improve their flow and energy consumption significantly.