Answers are grounded in this paper and cite their sources.
This paper presents a novel two-claw robotic gripper with haptic feedback that supports diverse grasping modes inspired by human taxonomy, including precision grasps, power grasps, and flip-and-pinch actions, requiring different postures, force distributions, and impedance characteristics. The gripper is driven by a servo motor with multiple springs, enabling an autonomous transition from a four-bar linkage for position control to a five-bar mechanism for impedance control upon object contact. In addition, a biomimetic tactile sensor, TacSlim, is developed to enhance haptic feedback during grasping of objects of varying sizes. The sensor features a wedge-shaped profile, fingerprints, and a nail-like structure. Its geometry is jointly optimized with the gripper’s structural parameters to ensure that the surface normal of TacSlim aligns with the contact regions across different object postures. Furthermore, we develop an integrated visual–tactile perception, action planning, and control framework, comprising modules for vision-based grasp mode and posture prediction, tactile sensing, arm–gripper coordination, and iterative learning-based variable impedance control to handle disturbances and uncertainties. These modules collectively enable fully autonomous execution of multifunctional grasping tasks. Experimental results demonstrate that the proposed gripper successfully grasps objects ranging from 2 mm to 64 mm in size and from 10 g to 4 kg in weight, achieving high detection accuracy and consistently high success rates across a wide variety of object shapes and conditions.
No takes yet. Share an insight, caveat, or question.
Lu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: