Eco-Driving Optimization Method for Heterogeneous CAVs at an Isolated Unsignalized Intersection

This paper presents an eco-driving optimization method for heterogeneous connected autonomous vehicles (CAVs) of mixed electric vehicles (EVs) and traditional fuel vehicles (FVs) at an isolated unsignalized intersection. An optimization method is designed to compute the arrival time of each vehicle in the heterogeneous vehicle fleet. Furthermore, we use quadratic programming to optimize the speed trajectory for EVs and the Gauss Pseudospectral method for FVs. Then, an eco-driving optimization method based on a rolling optimization framework that combines the time schedule algorithm and trajectory optimization method is proposed. We make a comparative study with the dynamic programming algorithm and simulation results show that our method outperforms the baselines, where the average energy consumption of EVs and FVs has been reduced by 0.09% and 0.14%, respectively.