In the underwater environment, in order to preserve rare and endangered objects or to eliminate the exotic invasive species that can destroy the ecosystems, it is essential to classify objects and estimate their number. It is very difficult to classify objects and estimate their number. While YOLO shows excellent performance in object recognition, it recognizes objects by processing the images of each frame independently of each other. By accumulating the object classification results from the past frames to the current frame, we propose a method to accurately classify objects, and count their number in sequential video images. This has a high classification probability of 93.94% and 97.06% in the test videos of Bluegill and Largemouth bass, respectively. The proposed method shows very good classification performance in video images taken of the underwater environment.

Detecting objects in digital images is challenging in computer vision, traditionally requiring manual threshold selection. However, object detection has improved significantly with convolutional neural networks (CNNs), and other advanced algorithms, like region-based convolutional neural networks (R-CNNs) and you only look once (YOLO). Deep learning methods have various applications in agriculture, including detecting pests, diseases, and fruit quality. We propose a lightweight YOLOv4-Tiny-based object detection system with a circular bounding box to accurately determine chrysanthemum flower harvest time. The proposed network in this study uses a circular bounding box to accurately classify the degree of chrysanthemums blooming and detect circular objects effectively, showing better results than the network with the traditional rectangular bounding box. The proposed network has excellent scalability and can be applied to recognize general objects in a circular form.

The manufacturing industry utilizes computing technology, robot technology, artificial intelligence, and IoT to improve production processes and quality. In particular, object detection technology is used in various industrial fields, and object detection methods based on deep learning are attracting attention. The tobacco processing industry requires automated production facilities, and quality control for defects in product appearance is essential. Mainly because tobacco products are sold at high prices, poor appearance is a significant issue in terms of consumer complaints and processing costs. Therefore, accurate cigarette detection is essential. We propose a modified network structure based on the YOLOv4-Tiny network, and use it to build a network optimized for cigarette detection. The modified network uses a single circular bounding box for learning and fast detection. It utilizes visual techniques, such as gradient-weighted Class Activation Mapping (Grad-CAM) to analyze the degree of activation of the network to construct an optimal network. This reduces the size of the network and increases processing speed, while maintaining detection accuracy. This paper is expected to play an important role in quality control and efficient production in the manufacturing industry.

This paper addresses the challenge of enabling a mobile robot to reach target coordinates and posture in minimal time by tackling the complexities of kinematic optimal control problems. Determining optimal control gains is analytically challenging, especially when incorporating dynamic models. The mobile robot’s control system comprises two controllers: a kinematic controller, optimized using genetic algorithms and cubic polynomial interpolation, to generate the optimal trajectory, and a dynamic controller using an adaptive PID approach to ensure stable and robust performance despite variations in system parameters like the robot’s weight. The proposed method demonstrates superior performance over existing methods, particularly under dynamic changes in robot characteristics. Simulation results show that the kinematic controller generates optimal trajectories more effectively than traditional techniques, while the adaptive PID controller maintains optimal performance despite disturbances and mass variations. In experiments with various disturbances, the proposed method consistently reached target coordinates with minimal timing discrepancies, outperforming comparative methods. Additionally, it significantly reduces spatial constraints and learning time. These findings contribute to the practical design of mobile robot controllers, ensuring efficient and stable operation, especially in scenarios requiring rapid derivation of optimal gains. Future work should focus on integrating kinematic and dynamic controllers into intelligent systems for mobile robots with arbitrary positions and postures, further enhancing autonomy and adaptability.

Traditional kinematic control of mobile robots primarily regulates position and orientation, often neglecting dynamic factors such as acceleration and torque, making it suitable for low-speed operations. However, to enhance safety and efficiency, it is crucial to consider dynamic constraints in the robot’s control system, including maximum velocity and angular velocity. Existing kinematic control methods typically fail to incorporate these dynamic limitations. This paper proposes a novel method for optimizing control gains for mobile robots, factoring in dynamic constraints. We introduce optimal control concepts to kinematic and dynamic controllers, employing a genetic algorithm to identify optimal control gains. Furthermore, we leverage a neural network with robust interpolation capabilities to select control gains for arbitrary initial poses effectively. The trained neural network accurately predicts control gains across various initial conditions, as simulation results confirm. The performance of the proposed neural network controller for diverse mobile robot postures is nearly equivalent to that of a controller using optimization gains derived from a genetic algorithm. In experiments with various robot postures, the maximum performance error time recorded was 0.44 seconds, reflecting a delay of 3.2% in arrival time. This approach enables mobile robots to reach target destinations with improved stability and performance, addressing the limitations inherent in traditional kinematic control methods.