A dynamic hierarchical cooperative lane change strategy for off-ramp connected and autonomous vehicles in mixed traffic environment

On the expressway, vehicle lane change faces intractable challenges. Improper lane change decisions often lead to serious traffic accidents and unexpected fluctuations of traffic flow. As the connected and automated vehicles (CAVs) emerge on roads, there is an opportunity to avoid poor lane change via communications and cooperations among vehicles. However, the practical traffic environment is mixed, where CAVs and human-driven vehicles (HDVs) coexist. To make full use of cooperative lane change on the expressway, this study exhausts all possible (seven types) vehicle configurations (i.e., CAVs or HDVs) surrounding the connected and automated off-ramp vehicle (CAOV). Then, a dynamic hierarchical cooperative lane change strategy is proposed, which includes: 1) a fuzzy logic-based decision model for lane change in the upper layer; 2) an optimal trajectory planning algorithm with anti-collision safety detection in the lower layer. The proposed strategy helps the CAOV make the best lane change decision among three choices according to the environment and generate an optimal lane change trajectory ensuring driving comforts and efficiency. By numerical analysis and comprehensive simulations, the proposed strategy shows good performance in increasing the lane change success rate (LCSR) and overall traffic speed on the expressway.