Enhancing Precision Agriculture Through PSO-Enabled Drones: Targeted Pest Control in Indonesian Rice Cultivation

This study presents an application of aerial vehicles (UAVs), commonly known as drones, integrated with computer vision technology to detect and treat pests in precision agriculture. The focus is on rice cultivation in Indonesia, where pest infestations have contributed to a decline in the area. To address this issue, the paper proposes an approach that combines the Particle Swarm Optimization (PSO) model with Visible Atmospherically Resistant Index (VARI) imaging. This combination allows for targeted spraying in areas identified as infested by pests, significantly improving efficiency. Drawing inspiration from the behavior of birds flocking or fish schooling, PSO, a stochastic technique based on population, utilizes a set of particles to evolve and find the optimum through iterative processes. Each particle, characterized by its velocity and position, navigates towards areas based on its best fitness value, thereby determining an optimal path for the drone. The effectiveness of PSO has been previously demonstrated in path planning for plantations and coconut detection for harvesting, highlighting its versatility. This research contributes to the community by showcasing the efficacy of utilizing PSO within UAVs for pest management and optimizing spraying operations. However, it should be noted that certain limitations exist, such as identifying one area and encountering challenges when translating paths into real-world conditions. Moving forward, researchers can direct their efforts toward addressing these obstacles to maximize the potential of UAVs in Precision Agriculture.