Accurate estimation of vehicle status parameters is important to vehicle stability control. Combination of 3-DOF vehicle dynamics model and adaptive control algorithm, proposed a method that modifies adaptive law through compensation was proposed to estimate vehicle parameters correctly. Bases on the method, the vehicle dynamics model was simplified to reduce the calculation burden and improve the real time performance. The starting and acceleration driving conditon and double lane change conditon were designed for the simulation and the hardware-in-loop test. Finally, both the simulation results and hardware-in-loop test results indicate that the proposed method could estimate the vehicle mass and moment of inertia more precisely and faster compared to the adaptive algorithm without compensation and satisfy the requirement of vehicle online estimation.