Coast Down Curve Computational Modeling and Its Influence on Urban and Highway Autonomy Results

Currently, fuels development is strongly dependent on experiments. New engines and vehicles simulation methodologies contribute to speed up R & D projects deadlines, as well as reducing costs. This paper presents a modeling methodology for a vehicle deceleration load curve (coast down) prediction and simulations of coast down variations impact on urban and highway autonomies. Two coast down curve mathematical models were successfully developed and validated. The first one, based on vehicles technical specifications and empirical equations, resulted in percent differences up to 9% compared to the experimental results. This is lower than the variation established on coast down standard, which is 15%. The second, generated by regression analysis between other vehicles characteristics versus experimental results of F0 and F2 (coast down curve parameters), resulted in percent differences up to 15%, for six of the eight vehicles. A simulator of urban and highway autonomies as coast down load functions was successfully implemented. Their results presented differences of up to 1% compared to the experiments. The models and simulations presented in this paper show potential to decrease coast down and autonomies experimental tests, which require considerable human and material resources. In addition, coast down models may compensate the low coast down tracks tests availability. Their results could be used when it is not possible perform track tests.