Optimization of vehicle speed in urban road using dynamic programming

Recently, air pollution and global warming are considered as major social issues. Transportation is one of the contributors to air pollutant emissions by consuming fossil fuel. Energy-efficient driving, which reduces energy consumption by improving driving control, is needed to reduce emissions. In this paper, optimal speed profile for energy-efficient driving is generated by using dynamic programming that can achieve a global optimal solution. A conventional vehicle is selected as a target vehicle and modeled based on longitudinal dynamics. The vehicle travels in the urban road, taking into account different road conditions and constraints(e.g., traffic signal and speed limit). It is assumed that the vehicle has access to the traffic signal system data and geographical information by vehicle-to-information(V2I). The selected roads are discretized for optimization, and the road characteristics and traffic signals are determined according to each location. Meanwhile, the cost function includes the fuel consumption of the vehicle and trip time adjusted by weight factor. Optimization results showed that the vehicle consumes minimal fuel with fewer vehicle stops due to the traffic light. We analyzed the optimal speed profile generated in a specific case and the iterative optimization results set a suitable cost boundary by converging the trip time.