A Novel Tree Search-Based Ramp Merging Strategy in a Multi-Lane Mixed Traffic Flow

To address the problems of ramp merging, this paper proposes a multi-lane centralized merging strategy in mixed traffic flow. First, vehicles group into platoon blocks, with CAVs serving as the lead vehicle to guide the following HDVs to merge safely. Second, a hierarchical optimization framework is established. An OMS-DFST algorithm is proposed in the upper layer to dynamically search for the optimal merging sequence of platoon blocks. Based on the merging sequence derived from the upper layer, an optimal cooperative merging control method is presented in the lower layer to determine the optimal trajectory. By taking efficiency, fuel consumption, pollution emissions, and comfort as cost functions, the optimal control input is solved by applying the Pontryagin principle. Results indicate the proposed strategy can effectively improve efficiency and comfort, and compared to the traditional FIFO algorithm, the vehicle’s fuel consumption is reduced by 5.5%, and pollution emissions are decreased by 4.5%.