Design and Implementation of A Strawberry-Picking Robot Using Artificial Vision in Greenhouses

Pacosoncco-Caira R.A.; Quispe-Barra M.A.; Arredondo-Mamani J.R.; Apaza-Cruz J.L.; Vidangos-Ponce G.R. · Smart Innovation, Systems and Technologies (2026) · DOI: 10.1007/978-3-032-24262-4_21

Resumen

In the Puno region (Àcora district), manual strawberry harvesting in greenhouses faces multiple limitations, including high labor demand, fruit damage due to inadequate handling, and low process efficiency, all of which directly affect farmers’ productivity and income. In response, this study designed and implemented a strawberry-harvesting robot with computer vision, using a quantitative approach with a quasi-experimental design to evaluate its performance under field conditions. Quantitative metrics such as detection accuracy, harvesting time, and operational efficiency were analyzed to validate its functionality. The robot features a compact design and corrosion-resistant materials, which facilitate its maneuverability and durability in confined spaces and humid greenhouse environments. Its computer-vision system estimates the distance from the detected strawberry’s visual area, achieving an average detection efficiency of 98.5 % and 47 s per strawberry harvested under optimal lighting conditions (12:00 p.m.). The results demonstrate the robot’s potential as a viable technological solution to reduce labor dependency, minimize fruit damage, and improve harvest efficiency, in addition to contributing to cost reduction and increased productivity. Future research will seek to optimize its robustness to light variations and its scalability for larger crop areas. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Instituciones: Universidad Nacional del Altiplano