Energy-Optimal Speed Planning in Urban Multiple-Intersection Scenarios Based on the Pruning Algorithm, Ant Colony Algorithm, and Dynamic Programming

This study focuses on the speed planning problem for connected electric vehicles in a multiple signalized intersection scenario. A dynamic programming (DP) algorithm can generate a more energy-efficient speed trajectory for a vehicle. However, it is computationally heavy and time-consuming. Therefore, this paper proposes a double-layer speed planning scheme to achieve the optimization of the energy used by electric vehicles (EVs) along routes with multiple intersections. In the first layer, to reduce the computational dimension of speed planning problems, we design a convex space planning method that considers multiple signalized intersections by combining the pruning algorithm and the ant colony algorithm (P-ACA). In the second layer, the energy optimization problem for electric vehicles crossing multiple intersections within the constraints of the convex space was constructed, and the energy-optimal speed was obtained based on DP. Next, a model predictive control (MPC) controller was designed to achieve accurate and rapid longitudinal speed control of electric vehicles. The simulation results show that the method proposed in this paper has considerable energy-saving potential in multiple-intersection scenarios. Compared with the modified smart driver model (M-IDM), energy consumption is reduced by 23.44%. The proposed double-layer speed-planning scheme significantly reduces the computation time required for speed planning. Finally, hardware-in-the-loop (HIL) tests were conducted to verify the effectiveness of the controller under real-time conditions.