Multi-picking of spoon-chain type seed-metering devices in potato planters causes multi-seeding, which leads to yield loss and plant diseases. This study developed a dedicated test bench for multi-picking detection and air-blowing excess seed removal, which integrated a detector based on spatial capacitive sensor, an air-blowing excess seed remover, and a controller. The feasibility of multi-picking identification via maximum net capacitance change was verified through theoretical analysis and Maxwell simulation. A two-point air-blowing strategy was adopted for air-blowing excess seed removal. Taking the success rate of air-blowing excess seed removal as the evaluation index, bench tests and parameter optimization were conducted with airflow speed for air-blowing excess seed removal, potato spoon depth, and seed-metering chain speed as variable factors. The optimal parameters were determined as an airflow speed of 31.34 m/s, a potato spoon depth of 19.67 mm, and a seed-metering chain speed of 0.29 m/s. Under this optimal parameter combination, the actual success rate of air-blowing excess seed removal reached 85.5%. This test bench provides a practical technical solution for solving multi-seeding in seed potato precision seeding, and lays a solid foundation for the optimization of seeding processes and the popularization of seed potato precision seeding technology.

The congenital yield reduction caused by miss-seeding in spoon-type seed-metering device of small and medium-sized potato planter is huge. Based on the physical mechanism of different measured capacitance values between two fixed capacitor plates with different media, a miss-seeding detection scheme based on a spatial capacitance sensor is proposed first. A simple and efficient spatial capacitance sensor that can obtain as large capacitance measurement value as possible is designed, and a dual CPU coordinated seed-monitoring and compensation control system architecture is adopted. AD7745 is selected for the capacitance measure of the spatial capacitance sensor, and the code of grating encoder is also recorded at the same time. Thereby, when each potato spoon passing through the space surrounded by the capacitor plates, the maximum net capacitance fluctuation and its corresponding position can be acquired. A suitable threshold can distinguish between normal-seeding and miss-seeding effectively. Moreover, it should be emphasized that, this monitoring system only requires one monitoring point. Then, based on obtained information, an improved miss-seeding catching-up compensation plan is put forward. By utilizing the powerful memory capability of the CPU, this system does not need to complete compensation immediately after the miss-seeding identification. Instead, the miss-seeding information and the location of the accident can be just marked in advance, and only when the opportunity arrives, can the miss-seeding catching-up compensation be truly executed. In this way, the position of the seed-monitoring points can be free from restriction, and the control strategy can therefore be significantly simplified. The soil tank test data showed that, the identification accuracy of the miss-seeding detection system was not less than 94%. When the seed-metering chain speeds are 0.2, 0.3, and 0.4 m/s, the average success rates of the miss-seeding compensation system are 94.32%, 83.65%, and 75.00%, respectively. The final miss-seeding rate can be below 3%, and the average deviation compensation rate was not higher than 30%, the miss-seeding was suppressed significantly. This system is a beneficial try in the non-photoelectric detection field and low complexity miss-seeding compensation for potato seeding.