ABSTRACT To accurately measure sowing depth, we designed a multisensor data fusion algorithm based on the Kalman filter and applied it to a dual-channel ultrasonic sowing depth measurement system. We installed two ultrasonic sensors on the frame above the depth limit wheel and the soil covering-compacting wheel. The terrain-like plate was installed on the central axis of the depth limit wheel to simulate the surface relief and calculate the seeding depth value. The Kalman filter algorithm filters out the clutter generated when the two ultrasonic sensors collect the depth data and then fuses the two ultrasounds of the filtered data. The test shows that when the vehicle speed is 4 km/h , the fluctuation ranges of the collected data of ultrasonic 1 and 2 are 45-60 mm and 35-70 mm , respectively, and the fluctuation ranges of the filtered data are 52-58 mm and 48-60 mm , respectively. After data fusion processing, the data fluctuate in a small range of approximately 55 mm . When the vehicle speed is 6 km/h and 8 km/h, the obtained results have the same trend as the experimental results at 4 km/h . The results show that the sowing depth measurement system is effective.