Design and development of a vision-guided adaptive and adjustable robotic gripper for harvesting apples
Franklin Open · 2026-09 · 16 · 100716
DOI: 10.1016/j.fraope.2026.100716 ↗Kurz gefasst
Fachartikel in der Zeitschrift Franklin Open, 2026-09. Verfasst von Samriddha Das und 2 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Ministry of Electronics and Information Technology.
Abstract (Originalsprache)
This study presents the design, fabrication, and laboratory validation of a vision-assisted adaptive robotic gripper for apple harvesting. The proposed end-effector introduces a dual actuation mechanism that integrates a slider–crank and wedge system to achieve a single-degree-of-freedom adaptive grasp, combined with Fin Ray Effect-based flexible TPU fingers for compliant fruit handling. Unlike conventional rigid grippers, the developed mechanism provides a variable trapezoidal workspace (0–158 mm at the fingertip), enabling accommodation of apples with varying diameters while maintaining structural simplicity. The prototype, fabricated using additive manufacturing (TPU fingers and PLA housing), weighs 950 g and demonstrates a load-carrying capacity of 1 kg, exceeding typical apple weights and thereby confirming mechanical robustness. Static structural simulation predicted a maximum contact force of 4.18 N, while experimental validation using a calibrated Force Sensitive Resistor recorded a peak force of 3.91 N, resulting in a relative deviation of 6.56%. These values remain below damage thresholds reported in literature, indicating low risk of fruit bruising under controlled conditions. For vision-based triggering, a YOLOv8s-seg instance segmentation model trained on a laboratory-scale dataset achieved an mAP@0.5 of 0.924, enabling automated actuation through bounding-box detection and ultrasonic distance verification. The gripper demonstrated a complete operational cycle time of 12 s (including a 5 s holding delay) and a peak power consumption of 7.32 W during loading. The results establish the feasibility of integrating soft adaptive mechanics with vision-triggered actuation for gentle and reliable apple handling. While validated in a controlled laboratory environment, the proposed system provides a modular foundation for future integration with robotic manipulators and field-scale orchard deployment.
Förderung
- Ministry of Electronics and Information Technology
Zitieren
APA
Das, S., Paul, A., & Machavaram, R. (2026). Design and development of a vision-guided adaptive and adjustable robotic gripper for harvesting apples. Franklin Open, 16, 100716. https://doi.org/10.1016/j.fraope.2026.100716
Vancouver
Das S, Paul A, Machavaram R. Design and development of a vision-guided adaptive and adjustable robotic gripper for harvesting apples. Franklin Open. 2026;16:100716. doi:10.1016/j.fraope.2026.100716
BibTeX
@article{Das2026,
author = {Das, Samriddha and Paul, Ayan and Machavaram, Rajendra},
title = {Design and development of a vision-guided adaptive and adjustable robotic gripper for harvesting apples},
journal = {Franklin Open},
year = {2026},
volume = {16},
pages = {100716},
doi = {10.1016/j.fraope.2026.100716}
}
EUVIMED