Ecological Adaptive Cruise Control With Optimal Lane Selection in Connected Vehicle Environments

Recent advances in transportation have enabled lane-specific measurements and lane-specific control. This paper makes use of such data to promote energy efficiency of vehicles. In particular, a Multi-Lane Adaptive Cruise Controller (MLACC) is designed which determines the optimal velocity and lane-to-drive in real-time. This cruise controller solves lane-specific optimization problems to compute an instantaneous trip cost for each lane and selects the lane that poses the lowest cost. The optimization tasks incorporate future route data and encompass multiple objectives including safety, energy efficiency and desired velocity tracking. Therefore, they can be treated as distinct Nonlinear Model Predictive Control (NMPC) problems that have to be solved altogether in each sampling time. To handle the computational load of solving multiple NMPCs in real-time, an integration of Newton and Generalized Minimal Residual numerical methods is employed. The proposed MLACC is implemented for a 2013 Toyota Prius and a wide range of simulation studies are performed to examine the controller. Specifically, hardware-in-the-loop experiments are utilized to evaluate the real-time implementability of the controller. In addition, extensive model-in-the-loop simulations are carried out and the results are compared with driver-in-the-loop experiments. Simulation results indicated that speed profiles and lane changes suggested by MLACC yield up to 27% improvement in energy consumption compared to human drivers.