Design of an AI ‐assisted autonomous orchard sprayer with dual spraying mechanisms

Abstract BACKGROUND This study presents the design, development, and field evaluation of Vabot, an artificial intelligence (AI)‐powered, fully electric autonomous agricultural ground vehicle intended for accurate pesticide application in orchard settings. The platform integrates a YOLOv5‐based computer vision algorithm into the control system to enable real‐time canopy detection and selective spraying. The main aim of the study was to evaluate the system's performance under real orchard conditions and compare two spraying mechanisms: a fixed‐arm system and an oscillating‐arm system. RESULTS Field experiments were performed in apple orchards at Ankara University, Türkiye, to assess droplet size, spray coverage, canopy penetration, and operational efficiency. The oscillating‐arm mechanism substantially enhanced spray uniformity and canopy penetration, achieving a coverage rate of 60.43%, compared with 14.87% for the fixed‐arm configuration. Measured droplet sizes ranged 125–242 μm for the fixed‐arm configuration and 147–182 μm for the oscillating‐arm system, indicating a more homogeneous distribution in the latter. The field efficiency of the vehicle was calculated as 41% over a 1200 m 2 area. CONCLUSION These findings demonstrate the potential of AI‐integrated autonomous systems to improve spray effectiveness, reduce pesticide usage, and minimize environmental contamination in orchards. The Vabot platform offers a practical and scalable solution for intelligent spraying in precision agriculture. Overall, the results highlight the synergy between AI and mechanical systems in delivering precise, efficient, and sustainable spraying performance in orchard environments. © 2026 Society of Chemical Industry.