In order to meet the demand of high accuracy and strong robustness of veh icle tracking in the intelligent transportation system,a new adaptive multi-fea ture fusion tracking algorithm is proposed in this paper.The proposed algorithm overcomes the partic le degeneration phenomenon well by using finite-difference extended Kalman filter to produce op timization proposal distribution function. An adaptive multi-feature fusion method is pro posed to overcome the defects of the additive fusion and the multiplicative fusion.The n ew method uses static and dynamic characteristics as complementary observables i n the framework of the improved particle filter.Experimental results show that this method is effective in enha ncing the accuracy and robustness of vehicle tracking system in different environments.