Polynomial Chaos Expansion 활용한 Anti-Lock Braking System의 통계적 모멘트 분석

초록

We propose a novel method for quantifying the braking performance distributions under vehicle parameter uncertainties using the polynomial chaos expansion (PCE). Two PCE models are constructed according to the number of stochastic inputs; a univariate model, where the vehicle mass is treated as a stochastic input, and a bivariate model, where both the vehicle mass and tire radius are modeled as stochastic inputs. The anti-lock braking system (ABS) control logic is formulated with PCE to estimate the time-dependent mean and variance of vehicle speed and position during ABS activation. The proposed method is validated against Monte Carlo simulation (MCS) in terms of accuracy and computational efficiency. Results show 96.2% reduction in computation time compared to univariate MCS and 87.6% reduction compared to bivariate MCS, while maintaining high accuracy in estimating the mean and standard deviation of speed and position.

키워드