Minimize the Fuel Consumption of Connected Vehicles between Two Red-Signalized Intersections in Urban Traffic

Eco-driving through multiple intersections can have significant fuel benefit for road transportation. Many existing studies assume that a vehicle runs at a constant speed between intersections, which can lead to a large error in fuel prediction and trajectory optimization. By directly taking powertrain dynamics into consideration, this paper studies generic operating rules of fuel-optimal operation for connected vehicles traveling between arbitrary two red-signalized intersections. An optimal control problem is formulated to minimize engine fuel consumption, which is numerically solved by the Legendre pseudospectral algorithm. It was found that the optimal driving operation between two red-signalized intersections takes the form of either a two-stage solution, i.e., accelerating and decelerating, or a three-stage solution, i.e., accelerating, constant speed cruising, and decelerating, depending on the distance between the intersections and the speed limit. A quasi-optimal operating rule is proposed to approximate the optimal solutions, achieving much faster computational speed and less than ±1.5% fuel consumption error compared to the numerical method. The effectiveness of the quasi-optimal rule is demonstrated by applying it to selected multiintersection passing scenarios.