Branch Interference Sensing and Handling by Tactile Enabled Robotic Apple Harvesting

In the dynamic and unstructured environment where horticultural crops grow, obstacles interference frequently occur but are rarely addressed, which poses significant challenges for robotic harvesting. This work proposed a tactile-enabled grasping method that combines deep learning, tactile sensing, soft robots. By integrating fin-ray fingers with embedded sensing arrays customized perception algorithms, robot gains ability to sense handle branch during harvesting process thus reduce potential mechanical fruit damage. Through experimental validations, an overall 83.3–87.0% status detection success rate, promising handling have been demonstrated. The can also be extended broader applications wherever undesirable foreign object intrusion needs addressed.