Dexterous Manipulation and Adaptive Robotic Grasping: From Mechanisms to Intelligent Control

Manipulation is central to robotics applications ranging from industrial automation to service, logistics, healthcare, and human–robot collaboration. Conventional parallel-jaw grippers and rigid end-effectors are effective for structured tasks, but often lack adaptability when handling diverse objects, deformable items, or cluttered scenes. In response, the community has developed dexterous multi-finger hands, underactuated and compliant grippers, and soft robotic end-effectors that exploit morphology and compliance to simplify control while improving robustness.At the same time, advances in tactile sensors, high-speed perception, and data-driven learning have enabled contact-aware manipulation, enabling robots to detect slip, estimate contact states, and adjust grasps online. However, challenges remain in modeling contact, achieving reliable generalization, and deploying systems that are both performant and safe in real environments. These trends motivate a broad, interdisciplinary forum for new ideas and validated systems.Robotic manipulation has progressed rapidly, yet achieving reliable, human-like dexterity remains a key bottleneck for robots operating in open-world environments. Practical manipulation systems must not only grasp objects, but also regrasp, reposition, and manipulate them under uncertainty in shape, pose, friction, and contact dynamics. This calls for tight integration across hand/gripper mechanisms, compliance design, multimodal sensing (vision–tactile–proprioception), and robust control and planning.The goal of this Research Topic is to gather recent advances that enable dexterous manipulation and adaptive grasping across robotic hands, grippers, and soft/compliant end-effectors. We aim to highlight methods spanning mechanical innovation, perception-driven grasp synthesis, contact-rich motion planning, and learning-based manipulation policies that generalize beyond curated benchmarks. Contributions that bridge simulation and real-world deployment, address safety and reliability, and demonstrate performance in unstructured scenarios are particularly encouraged. By connecting robotics, control, AI, and mechanical design communities, this topic seeks to accelerate progress toward versatile, deployable manipulation systems.This Research Topic welcomes original research articles, reviews, and perspectives on theories, methods, and systems that advance robotic grasping and manipulation. We encourage contributions with rigorous evaluation in simulation and/or physical experiments, and value reproducibility, open benchmarks, and strong ablation/analysis when applicable. Relevant themes include (but are not limited to):1.Design and modeling of dexterous robotic hands and grippers2.Soft robotic manipulation and compliant grasping mechanisms3.Adaptive and intelligent grasp planning algorithms4.Learning-based manipulation and reinforcement learning for grasping5.Tactile sensing and perception for robotic hands6.Motion planning and control for dexterous manipulation7.Human-inspired manipulation strategies8.Applications of dexterous manipulation in industry, service robotics, and healthcareSubmissions may include theoretical studies, experimental validation, system design, and real-world robotic applications that contribute to advancing the field of robotic grasping and manipulation. Topic editors: Han YUAN, Juntian Qu, Qiukai Qi Source: https://www.frontiersin.org/research-topics/79085/dexterous-manipulation-and-adaptive-robotic-grasping-from-mechanisms-to-intelligent-control