Sensing Technologies in Robotic Manipulators for Low-Damage Fruit and Vegetable Grasping: Principles, Integration, and Applications

Fruits and vegetables vary greatly in shape and are easily damaged, and are typically handled in highly unstructured environments, posing significant challenges to stable, accurate, and low-damage robotic manipulation based solely on predefined trajectories and fixed gripping parameters. Sensing technologies are therefore essential for target recognition, spatial localization, quality assessment, and grasp-state feedback. This review first analyzes the sensing requirements associated with the geometric, mechanical, and quality-related characteristics of fruits and vegetables, and then systematically examines the operating principles, information characteristics, and functional suitability of optical, electrical, and acoustic sensing technologies. Based on the spatial relationship between sensing modules and manipulators, existing sensor integration strategies are further classified into externally mounted, palm-centered, finger-contact, embedded conformal, and coordinated multi-position configurations, followed by a review of their applications in field harvesting and postharvest sorting. Key challenges are subsequently discussed, including biological variability, environmental interference, structural coupling, sensor durability, multimodal information fusion, and the lack of standardized evaluation criteria. Future research directions are highlighted in terms of sensing–structure co-design, flexible functional materials, task-oriented multimodal sensing, transferable state estimation, and standardized evaluation frameworks. This review provides a systematic reference for sensing-technology selection, sensor–manipulator integration, and the engineering development of robotic systems for fruit and vegetable grasping.