Engine rotational dynamics can be estimated from the tooth signals of the crankshaft wheel. The accuracy of the calculation rises when the number of tooth increases. However, the engine control unit (ECU) needs to detect the tooth position and converts it into time information. It will overload the interrupt system. In order to overcome this problem, the field programmable gate array (FPGA) is proposed to implement the engine rotational dynamics calculation with stroke identification algorithm. The FPGA can accelerate the interpolation and the multiplication processes by implementing them in parallel computation. Thus, the proposed method can calculate the engine rotational dynamics in real time, and then the calculation results can be transferred into the robust Kalman filter which was established by Matlab/Simulink in a computer, i.e., a simulated ECU. Finally, a 125cc scooter is utilized to verify the proposed algorithm. The test results show that the proposed method can be used to obtain precise information of engine rotational dynamics.