Speed Optimization Strategy for Connected Vehicles on Ecological Roads

This paper proposes an eco-friendly trajectory-based traffic control strategy for connected vehicles on ecological roads with wildlife crossing corridors to reduce energy consumption while improving the protection of wildlife. Surveillance environment is constructed above the corridor to detect wildlife crossing events and broadcast the information to the roadside unit, which will further transmit information to the upcoming vehicles by Internet of Vehicles for preoptimizing speed trajectory. To minimize the risk of wildlife-vehicle collision, the protection strategy is described as adjusting vehicle speeds in advance to pass alternatively in turn for vehicles and wildlife. A dual-objective control model is thus proposed to minimize the total energy consumption and crossing corridor time under safe driving. The Deep Deterministic Policy Gradient algorithm is used to train the dual-objective model. Comprehensive experiments are investigated based on the practical data in Chongming Island in Shanghai. The result shows that the speed control strategy is reliable and feasible for connected vehicles, which can save energy consumption by an average of 60.5% while ensuring animal safety.