The emergent high-speed growth of precision agriculture requires solid frameworks that will improve sustainability, efficiency, and transparency in managing vineyards. The proposed research suggests a system that uses IoT sensor networks, machine learning models, and Ethereum-based blockchain to solve two important problems: pest identification and the optimization of water resources. The IoT layer was used to implement soil moisture, climate, and imaging sensors to gather real-time data about the vineyard. At the edge, preprocessing and deep learning algorithms were used with a blockchain-enabled Convolutional Neural Network (CNN) to provide correct pest detection. At the same time, the timing of irrigation was automated by IoT-enabled soil moisture monitoring, which greatly decreased the amount of wasted water. This ensured integrity of data and trust in the farmers as the Ethereum blockchain layer offered immutable storage, smart contracts to make decisions, and secure events logging. The results of the experiments revealed that the proposed system had a 95.8% pest detection accuracy, which was higher than the traditional and baseline machine learning methods. The efficiency of water management increased too by reducing water consumption by 55.2 % and doubling the crop productivity by 26.8 %t. Moreover, the blockchain application has scored high security of 0.93, confirming that it is a reliable solution even with moderate latency overhead. In this study, the authors emphasize the opportunities of converging the IoT with blockchain and transforming viticulture through sustainable practices, data safety, and efficiency.
C.M. Nalayini, A.R. Sathyabama, S. Priyadharshini, Sumathi. S
In the modern agricultural landscape, collaborative robots have emerged as key technology for enhancing precision, efficiency and sustainability in farming practices. This research introduces a Smart Cobot Greenhouse Assistant developed to automate plant health monitoring and management with controlled environment. It identifies plant's condition and take suitable actions like watering dry plants, raising alerts for diseased ones, and leaving healthy plants unharmed. A two-dimensional workspace is modelled where the cobot is initialized at the origin$(\mathbf{0, 0})$and plants are located at respective positions$(\mathbf{x, y})$. Dynamic states such as healthy, dry and diseased are assigned and simulated through python code using a well-defined synthetic dataset. Cobot's movements and actions are animated using Matplotlib visualization to produce an efficient and sustainable greenhouse management and the information are recorded into the blockchain distributed ledger and alerts are sent to the owner for further decision. Compared with the traditional system, the proposed smart cobot achieved operational responsiveness, dynamic visual feedback, and trustworthy data handling with full transparency.
Robotic swarm intelligence is a rapidly evolving field that leverages principles of decentralized control, self-organization, and emergent behavior to enable effective coordination and collaboration in multi-robot systems. Inspired by biological swarms, such as ant colonies and bird flocks, swarm robotics focuses on the collective performance of simple agents interacting locally to achieve complex tasks. This approach enhances scalability, robustness, and adaptability in dynamic and unpredictable environments. Key applications include search and rescue, environmental monitoring, industrial automation, and military operations. Recent advancements in artificial intelligence, machine learning, and communication technologies have further improved swarm decision-making, task allocation, and formation control. This paper explores the fundamental principles, coordination strategies, and challenges in robotic swarm intelligence, highlighting future directions for optimizing collaboration in autonomous multi-robot systems.
This paper presents a selective survey of theoretical and experimental progress in the development of biologicallyinspired approaches for complex surveillance and reconnaissance problems with multiple, heterogeneous autonomous systems. The focus is on approaches that may address ISR problems that can quickly become mathematically intractable or otherwise impractical to implement using traditional optimization techniques as the size and complexity of the problem is increased. These problems require dealing with complex spatiotemporal objectives and constraints at a variety of levels from motion planning to task allocation. There is also a need to ensure solutions are reliable and robust to uncertainty and communications limitations. First, the paper will provide a short introduction to the current state of relevant biological research as relates to collective animal behavior. Second, the paper will describe research on largely decentralized, reactive, or swarm approaches that have been inspired by biological phenomena such as schools of fish, flocks of birds, ant colonies, and insect swarms. Next, the paper will discuss approaches towards more complex organizational and cooperative mechanisms in team and coalition behaviors in order to provide mission coverage of large, complex areas. Relevant team behavior may be derived from recent advances in understanding of the social and cooperative behaviors used for collaboration by tens of animals with higher-level cognitive abilities such as mammals and birds. Finally, the paper will briefly discuss challenges involved in user interaction with these types of systems.