Tilling depth was usually detected manually with a ruler in many cultivation areas in China, which was inefficient and the result had a high uncertainty, which made subsoiling monitoring difficult. A depth algorithmic model was studied on basis of the attitude of tractor’s three-point hitch for measuring tilling depth and based on which a subsoiling management system was developed and implemented. For establishing the measurement model, the movement path of the tractor and subsoiler was analyzed, and by studying the dynamical equation, the tilling depth measurement model was proposed. Tilling depth was calculated in the model by detecting the attitude of the tractor and subsoiler. In order to verify the accuracy of the measurement method, an ARM kernel based tilling depth measurement sensor and a monitoring system were developed. The system integrated GPS, GPRS and SD card together, which had ability of tilling depth measuring, working speed measuring, working position measuring and heading measuring. The working data was stored in SD card of the system, and was transmitted to the data center by GPRS at the same time for further process. The system calibration test showed that the maximum error was 0.88cm，the average error was less than 0.21cm, the root mean square error (RMSE) was less than 0.66cm. Using the calibrated system to take several tests in field, the result showed that the maximum error was 1.18cm, the average error was less than 0.45cm and the RMSE was less than 0.64cm, which indicated that the system had a high accuracy and stability. The system supplied an information-based method for monitoring the operation quality of subsoiling, and solved the problems caused by manual measuring. The system was useful in farming industry.