Microscopic Calibration and Validation of Car-Following Models -- A Systematic Approach

摘要

Calibration and validation techniques are crucial in assessing thedescriptive and predictive power of car-following models and their suitabilityfor analyzing traffic flow. Using real and generated floating-car andtrajectory data, we systematically investigate following aspects: Datarequirements and preparation, conceptional approach including localmaximum-likelihood and global LSE calibration with several objective functions,influence of the data sampling rate and measuring errors, the effect of datasmoothing on the calibration result, and model performance in terms of fittingquality, robustness, parameter orthogonality, completeness and plausibleparameter values.