Vol. 83, No. 1 Abstract
[Keywords: hyperspectral camera, night sensing, vehicle robot, autonomous driving, proximal sensing, RTK-GNSS, crop mapping, smart agriculture, on-the-go sensing, precision farming]
A nighttime sensing method using artificial lighting was devised to stably acquire spectral reflection data that can visualize field conditions. The final goal of this research is to develop a field scouting robot that can automatically observe and map field status. Here, we seek to prove the significance of nighttime sensing and establish a method for obtaining spectral reflection data at night. In daytime sensing, we confirmed that luminance value changed drastically due to clouds, and that the reflection spectral characteristics of sunlight shifted gradually. We succeeded in correcting the brightness deviation, which is a problem when using artificial lighting, and established a method for acquiring data at night.
[Keywords: orchard, trellis training system, Japanese pear, orchard sprayer, spray drift, close spraying]
To reduce spray drift in a trellis training system, the authors developed a low-drift sprayer with a close spraying mechanism and bi-directional flat fan nozzles. Spraying experiments of this new sprayer were conducted in a Japanese pear field. Results of a water spraying experiment showed that close spraying with the new sprayer at an air volume of 3.8 m3/s reduced drift distance with the same spraying perform<->ance as for 7.8 m3/s with a conventional orchard sprayer. A pesticide spraying experiment showed that, compared to 0.03 ppm pesticide residue for a conventional sprayer at an air volume of 7.8 m3/s, the new sprayer with an air volume of 3.8 m3/s resulted in less than 0.01 ppm pesticide residue on Japanese mustard spinach.