The turning radius and ground width of the chassis are important factors affecting the efficiency of four-wheel chassis on small lands. A four-wheel self-propelled chassis suitable for small plots is designed. Through controlling the rotation of the wheel, the chassis actively rotates the front drive axle around the steering device, which can drive the chassis to rotate at any angle. ADAMS simulation is established to simulate the turning track model of the swing chassis. When the chassis adopts a wheel diameter of 800 mm and a wheel width of 100 mm, the most compact wheel track-to-wheelbase ratio of the chassis meeting the steering requirements is 1:1.6. The physical model verifies that the chassis meets the requirements of small block steering.

View full abstract

This study aims to enhance the efficiency of a duck farm by developing a deep learning-based weight estimation system. We utilized Convolution Neural Network (CNN) and Vision Transformer (ViT) models, applying weighted random sampling (WRS) to address dataset imbalance. RGB images of ducks were used for training. The ViT model with WRS achieved the highest accuracy, with a mean absolute error (MAE) of 0.11, mean squared error (MSE) of 0.02, and a coefficient of determination (R²) of 0.7554. These findings demonstrated that WRS significantly improved the accuracy of regression AI models, enabling precise weight estimation from imbalanced data, thereby contributing to the productivity and efficiency of duck farming.

View full abstract

The objective of this study is to develop an AI-based fruit monitoring system to address the issue of image background noise when automatically counting tomatoes grown in greenhouse horticulture. The system consists of a scanning device and a rail-guided vehicle. This system creates a panoramic image of the crop canopy and automatically counts the number of fruits using deep learning-based fruit detection model (Mask R-CNN). A focused illumination unit on our scanning device creates a light-dark contrast between the near and far sides of the camera, removing unnecessary objects from the background of the images. We applied our system to detect tomato fruits grown in an experimental greenhouse and successfully achieved good performance (AP₅₀ = 0.95).

View full abstract

Strawberries have high respiration rates, and their quality deteriorates quickly during storage. Therefore, we investigated quinone, ascorbic acid, and anthocyanin content changes during storage of fruits. Additionally, we further investigated whether these substances could be measured nondestructively in strawberry fruit before storage (day 0) and on days 3 and 5 of storage using hyperspectral imaging. Partial least squares (PLS) regression analysis was performed using the obtained image data, and a calibration curve was created and evaluated using complete cross-validation. The anthocyanin and quinone contents increased during the storage with weight loss. The accuracy of the quinone content calibration curve using hyperspectral imaging was high (R² = 0.78), suggesting it may be used to measure quinone content during storage nondestructively.

View full abstract

In this study, semi-continuous cultivation of microalga (Botryococcus braunii) was conducted by aeration with a mixture of ammonia (NH₃) and carbon dioxide (CO₂), which simulates the waste gas generated during the composting of livestock waste, to investigate the possibility of using NH₃, an odorant substance, as a nitrogen source for microalgal cultivation. By using NH₃ as a nitrogen source, more than 1.3 times more nitrogen could be fixed as organic nitrogen while maintaining a higher algal density than when potassium nitrate (KNO₃) was used as the nitrogen source. In addition, a comparison of three different nitrogen sources, namely, KNO₃, ammonium sulfate ((NH₄)2SO₄), and NH₃, revealed that the use of NH₃ as a nitrogen source resulted in higher nitrogen absorption by the alga.

View full abstract