Optimisation of energy efficiency based on average driving behaviour and driver's preferences for automated driving

The implementation of anticipating driving styles in adaptive cruise control systems promises to considerably reduce fuel consumption of vehicles. As drivers have to accept the optimised driving styles of such systems, which implement longitudinally automated driving, the optimisation results should not deviate strongly from the average driving behaviour. This work presents an approach to the optimisation of the vehicle's longitudinal dynamics, which is based on a predicted average driving profile. The proposed approach ensures that the optimisation results meet the expectations of drivers by directly accounting for driver's preferences on weighting up travel time against fuel consumption relative to the average driving profile. Based on human decision finding, rational and intuitive planning decisions are modelled in a cost function and represent optimisation constraints. The approach generally includes information from vehicle‐to‐vehicle and vehicle‐to‐infrastructure communication (V2X), which is an extension to the state‐of‐the‐art. This study describes the optimisation approach and presents a method to determine suitable optimisation parameters in order to consider driver's preferences. The optimisation approach is applied in a simulated test drive and improvements in fuel economy are analysed. Finally, the authors sketch a reference system architecture to prove the feasibility of the presented approach.