As a result of climate change and global warming, plant diseases and pests are drawing attention because they are dispersing more quickly than ever before. The tomato leaf miner destroys the growth structure of the tomato, resulting in 80 to 100 percent tomato loss. Despite extensive efforts to prevent its spread, the tomato leaf miner can be found on most continents. To protect tomatoes from the tomato leaf miner, inspections must be performed on a regular basis throughout the tomato life cycle. To find a better deep neural network (DNN) approach for detecting tomato leaf miner, we investigated two DNN models for classification and segmentation. The same RGB images of tomato leaves captured from real-world agricultural sites were used to train the two DNN models. Precision, recall, and F1-score were used to compare the performance of two DNN models. In terms of diagnosing the tomato leaf miner, the DNN model for segmentation outperformed the DNN model for classification, with higher precision, recall, and F1-score values. Furthermore, there were no false negative cases in the prediction of the DNN model for segmentation, indicating that it is adequate for detecting plant diseases and pests.

Bipedal robots have gained increasing attention for their human-like mobility which allows them to work in various human-scale environments. However, their inherent instability makes it difficult to control their balance while they are physically interacting with the environment. This study proposes a novel balance controller for bipedal robots based on a behavior cloning model as one of the machine learning techniques. The behavior cloning model employs two deep neural networks (DNNs) trained on human-operated balancing data, so that the trained model can predict the desired wrench required to maintain the balance of the bipedal robot. Based on the prediction of the desired wrench, the joint torques for both legs are calculated using robot dynamics. The performance of the developed balance controller was validated with a bipedal lower-body robotic system through simulation and experimental tests by providing random perturbations in the frontal plane. The developed balance controller demonstrated superior performance with respect to resistance to balance loss compared to the conventional balance control method, while generating a smoother balancing movement for the robot.

In this study, we proposed a novel machine-learning-based functional electrical stimulation (FES) control algorithm to enhance gait rehabilitation in post-stroke hemiplegic patients. The electrical stimulation of the muscles on the paretic side was controlled via deep neural networks, which were trained using muscle activity data from healthy people during gait. The performance of the developed system in comparison with that of a conventional FES control method was tested with healthy human subjects.

This paper presents a novel approach to enhance pose estimation accuracy and precision in mobile robotics using multiple graphical markers. While traditional single-marker methods using graphical markers such as ArUco offer simple implementation, their performance is susceptible to environmental variations and measurement errors. To address these limitations, we propose a robust pose estimation algorithm that leverages multiple markers simultaneously. Our approach incorporates two key mechanisms: spatial consistency verification to detect invalid markers within the marker array, and temporal stability analysis to identify and exclude outlier measurements. The algorithm enhances pose estimation accuracy by averaging measurements from valid markers while preventing any single marker from dominating the estimation process. The effectiveness of our approach was validated through experiments using both fixed and drone-mounted camera configurations. The results demonstrated that the pose estimation using multiple markers significantly improved both accuracy and precision compared with single-marker approaches. In fixed-camera experiments, the proposed method showed reduced mean errors and standard deviations in both position and orientation measurements across various camera poses. Similarly, in drone-mounted camera experiments, our approach exhibited superior stability with significantly lower measurement variations during hovering maneuvers. These improvements were particularly pronounced in challenging scenarios, such as when the camera was tilted at large angles relative to the marker plane. This research contributes to the advancement of reliable pose estimation methodologies in mobile robotics and autonomous systems, with potential applications across diverse environments where precise position and orientation measurements are crucial.

This study proposes the Weight-Incorporating A* (WIA*) algorithm for mobile robot path planning. The WIA* algorithm integrates three weight factors into the Conventional A* cost function: an Obstacle Collision (OC) weight factor for collision avoidance, a Path Distance (PD) weight factor for path length optimization, and a Driving Suitability (DS) weight factor for environmental considerations. Experimental validation was conducted using nine 2D grid maps and a 3D virtual environment. The results show that WIA* achieved zero obstacle collisions compared to an average of 9.11 collisions with Conventional A*. Although WIA* increased path length by 12.69%, it reduced driving suitability cost by 93.88%, achieving zero cost in six out of nine test environments. The algorithm demonstrates effective collision-free path generation while incorporating environmental factors for practical mobile robot navigation.

Drones are increasingly used in various fields such as agriculture, logistics, and disaster response due to their agility and versatility. In indoor plant factories, small drones are used to monitor crop conditions and collect environmental data. However, small drones require frequent recharging due to their limited battery capacity, making autonomous charging systems essential for uninterrupted operation of drones. This study proposes an autonomous charging station designed for small drones in indoor plant factories. The system employs a wired charging mechanism to enhance charging efficiency, and a 3-degree-of-freedom (DOF) pose alignment system, utilizing an XY plotter and turntable, to correct drone landing errors. The alignment system ensures that drones, landing with random positions and orientations, are automatically adjusted to the correct position for charging. Experiments demonstrated that the charging station successfully aligned and charged drones with a 93% success rate on the first attempt. Even in cases of failure, the system automatically retried until a 100% success rate was achieved. This autonomous drone charging system has the potential to significantly enhance operational efficiency in indoor plant factories and can be adapted for various drone models in future applications.