To monitor young leaves, which are an important diagnostic factor of growth in strawberries, a method that exposes the growth point was investigated by applying airflow from above. The results of wind tunnel experiments showed that 18 out of 19 growth points could be exposed at an air velocity of 6 m/s, and the frequency of exposure per unit time was calculated to be 89％. Additionally, the position of the airflow was studied assuming that the movement of the quadcopter follows the staggered rows of crops. By flying directly above the crop row, it was possible to optimally expose the growing point 76％ of the time while moving 400 mm in front of and behind the plant.

A boom height measurement device with three position-sensitive detectors （PSD） was developed for an automatic spray height control system. We confirmed that the PSD could stably measure spray height outdoors using an infrared transmission filter. Then, the PSDs were evaluated for their output characteristics for leaves of major crops. Additionally, the interference areas of the PSDs were identified for optimal placement. Finally, the prototype was compared to an ultrasonic sensor in a potato field to evaluate its accuracy. The RMS of the device based on the ultrasonic sensor with a target spray height of about 60 cm was 5.5 cm. The developed device was useful as a sensor to measure the spray height.

In this study, we developed a garlic sorter and experimentally evaluated the performance of a working system consisting of a garlic trimmer and a sorter. The garlic sorter detects trimmed roots and sorts them into three containers corresponding to three classes, namely, good, bad, and scratched using a deep learning model. If a trimmed garlic is sorted into the bad class, it will be trimmed and detected again until it is sorted into the good class. The results of testing the working system showed that the detection and sorting by the sorter were faster than those by an expert. The final pass rate of the good class was high irrespective of the experience of the operator.

Positioning accuracy of Quasi-Zenith Satellite System （QZSS） was experimented by using a robot tractor in mountainous fields. The robot tractor can plow the fields while automatically driven by using a real-time kinematic global navigation satellite system （RTK-GNSS）. Two multi-frequency GNSS receivers were additionally mounted on the tractor to measure the positioning accuracy and the number of used satellites by different augmentation signals. The positioning data of tractor were collected in seven fields under three methods, which are RTK by GPS and GLONASS, CLAS （centimeter level argumentation service） by QZSS, and RTK by GPS, GLONASS, Galileo, QZSS and Beidou. The result shows that the CLAS method of the QZSS can measure the stable position of tractor.

We have developed a new technology to prevent accidents involving hand-fed threshing, which is one of the most typical accidents associated with a head-feeding combine harvester. Based on a survey of existing technologies, a glove with a magnet attached to it was worn by the operator, and the glove was detected by a metal sensor or a magnetic sensor installed in the hand-threshing section of the combine harvester. The magnetic core coil was installed on a two-row combine harvester, and a method for noise reduction in the detection signal due to the movement of metal parts around the coil, as well another to determine the detection threshold, were proposed. The specifications of the magnet were also examined.