Recent Developments in Robotic Grasping Detection

Robotic grasping detection, a specialized domain in robotics, focuses on enabling robots to autonomously identify optimal object - grasping positions through sensory data analysis. This innovation has transformative implications for robotic interactions in diverse environments, including agriculture. The integration of deep learning techniques has significantly advanced this field, allowing robots to learn from extensive datasets and adapt to varied contexts, enhancing their versatility. This article provides a comprehensive survey of methodologies for estimating grasping configurations, explores diverse sensor modalities for data acquisition, and categorizes robotic grasping into two paradigms: known and unknown objects, based on familiarity. The study culminates in outlining critical considerations for developing robust grasping detection systems.