Stochastic Fuel Efficient Optimal Control of Vehicle Speed

Abstract

Stochastic dynamic programming (SDP) is applied to generate control policies that adjust vehicle speed to improve average fuel economy without degrading, significantly, the average travel speed. The SDP policies take into account statistical patterns in traffic speed and road topography. Specific problems of fuel efficient in-traffic driving and fuel efficient lead vehicle following are considered, and it is shown how these problems can be treated within an SDP framework. Simulation results are summarized to quantify fuel economy improvements, and experimental results are reported for the fuel efficient lead vehicle following case. The properties of vehicle speed trajectories induced by SDP policies are examined. © Springer International Publishing Switzerland 2014.