Online adaptive approach for a game-theoretic strategy for Complete Vehicle Energy Management

This paper introduces an adaptive approach for a game-theoretic strategy on Complete Vehicle Energy Management. The proposed method enhances the game-theoretic approach such that the strategy is able to adapt to real driving behavior. The classical game-theoretic approach relies on one probability distribution function whereas the proposed approach is made adaptive by using dedicated probability distribution functions for different drive patterns. Owing to the adaptability of the proposed approach, the strategy is further refined by proposing dedicated objective functions for the driver player and for the auxiliary player. Next, an algorithm is developed to classify the measured driving history into one of the pre-defined drive pattern and employ the corresponding game-theoretic strategy. Multiple strategies are simulated with a model of a parallel hybrid heavy-duty truck with a battery and electric auxiliaries. The fuel reduction results are compared and the adaptive game-theoretic approach shows an improved and a more robust performance over different drive-cycles compared to the non-adaptive one.