Abstract: In order to adapt to the development trend of precision agriculture and improve the accuracy of the yield measurement system, a set of impulse yield measurement systems was developed. The impulse sensor module, Beidou module, DTU transmission module, and upper computer were included in the system. The information of the impulse sensor was collected and sent to the upper computer through CAN communication. On the upper computer, the yield data was filtered by the IMM-UKF algorithm, and then the grain yield was obtained according to the calibration curve. Through the vehicle no-load test, IMM-UKF algorithm, KF, and UKF algorithm were used for comparison and verification. The results showed that the IMM-UKF algorithm has a better filtering effect, and the mean square deviation is 0.000 42 N. Through the calibration test of the sensor, the relationship between the detection value of the sensor and the actual grain weight was determined. After the field yield test, the yield map of each field was made. The test results show that the average error of the yield measurement system was 3.911%.