Multilane Eco-Cruise for Battery Electric Vehicles With Terrain and Traffic Previews

The eco-cruise (EC) technique is recognized as an effective method to cut energy consumption in battery electric vehicles (BEVs). Prior studies have examined the road slope and traffic disturbance during speed optimization. However, the obligation to remain in a single lane limits the potential for energy savings and travel efficiency, thereby affecting the efficacy of existing EC controllers. This article investigates multilane EC (MLEC) for BEVs, taking into account both terrain and traffic previews. To address the complexity of mixed-integer nonlinear programming, we introduce a novel MLEC controller that combines a predictive EC strategy and a lane-changing strategy. We also develop a speed prediction method using self-learning Markov model and an efficient alternating direction method of multipliers (ADMMs) for the nonlinear program, aiming to implement the controller in practice. The performance of the MLEC controller is validated through comprehensive microtraffic simulations and hardware-in-the-loop (HIL) experiment. Results indicate that the proposed lane-changing strategy leads to a 2.99% reduction in energy consumption. When combined with speed planning, the MLEC achieves 6.68% and 24.56% reduction in energy consumption and driving time, aligning closely with the global benchmark. Furthermore, an average update computation time of 23.5 ms during HIL testing confirms its potential suitability for on-board control.