The Developments and Challenges towards Dexterous and Embodied Robotic Manipulation: A Survey

2025/07/16 by Gaofeng Li, Ruize Wang, Li, Gaofeng +8 · 1 voice · 3 citations
Computer Science · Engineering · #Modular Robots and Swarm Intelligence #Robot Manipulation and Learning #cs.RO

paper · pdf · doi:10.48550/arxiv.2507.11840

arxiv published 2025/07/16 · arxiv updated 2025/11/18

Abstract

Achieving human-like dexterous robotic manipulation remains a central goal and a pivotal challenge in robotics. The development of Artificial Intelligence (AI) has allowed rapid progress in robotic manipulation. This survey summarizes the evolution of robotic manipulation from mechanical programming to embodied intelligence, alongside the transition from simple grippers to multi-fingered dexterous hands, outlining key characteristics and main challenges. Focusing on the current stage of embodied dexterous manipulation, we highlight recent advances in two critical areas: dexterous manipulation data collection (via simulation, human demonstrations, and teleoperation) and skill-learning frameworks (imitation and reinforcement learning). Then, based on the overview of the existing data collection paradigm and learning framework, three key challenges restricting the development of dexterous robotic manipulation are summarized and discussed.

Citations

• GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
• DexH2R: Task-oriented Dexterous Manipulation from Human to Robots
• π₀: A Vision-Language-Action Flow Model for General Robot Control
• Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning
• GRUtopia: Dream General Robots in a City at Scale
• Open-TeleVision: Teleoperation with Immersive Active Visual Feedback
• Sim2Real Manipulation on Unknown Objects with Tactile-based Reinforcement Learning
• DexCap: Scalable and Portable Mocap Data Collection System for Dexterous Manipulation
• Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation
• Tactile Active Inference Reinforcement Learning for Efficient Robotic Manipulation Skill Acquisition
• RoboGen: Towards Unleashing Infinite Data for Automated Robot Learning via Generative Simulation
• VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models
• Affordances from Human Videos as a Versatile Representation for Robotics
• UniDexGrasp++: Improving Dexterous Grasping Policy Learning via Geometry-aware Curriculum and Iterative Generalist-Specialist Learning
• Identifying Expert Behavior in Offline Training Datasets Improves Behavioral Cloning of Robotic Manipulation Policies
• VideoDex: Learning Dexterity from Internet Videos
• DexGraspNet: A Large-Scale Robotic Dexterous Grasp Dataset for General Objects Based on Simulation
• A4T: Hierarchical Affordance Detection for Transparent Objects Depth Reconstruction and Manipulation
• Anthropomorphic Twisted String-Actuated Soft Robotic Gripper with Tendon-Based Stiffening
• Deep Learning Approaches to Grasp Synthesis: A Review
• Dexterous Imitation Made Easy: A Learning-Based Framework for Efficient Dexterous Manipulation
• Challenges and Outlook in Robotic Manipulation of Deformable Objects
• A Survey of Embodied AI: From Simulators to Research Tasks
• How to Train Your Robot with Deep Reinforcement Learning; Lessons We've Learned
• Contact Mode Guided Sampling-Based Planning for Quasistatic Dexterous Manipulation in 2D
• Solving Challenging Dexterous Manipulation Tasks With Trajectory Optimisation and Reinforcement Learning
• Vision-based Teleoperation of Shadow Dexterous Hand using End-to-End Deep Neural Network
• Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations
• Kernelized Movement Primitives
• Generative Adversarial Imitation Learning
• Learning Dexterous Manipulation for a Soft Robotic Hand from Human Demonstration
• Dynamical Movement Primitives: Learning Attractor Models for Motor Behaviors

Cited by

• UniBYD: A Unified Framework for Learning Robotic Manipulation Across Embodiments Beyond Imitation of Human Demonstrations
• In-Hand Manipulation of Articulated Tools with Dexterous Robot Hands with Sim-to-Real Transfer
• CookBench: A Long-Horizon Embodied Planning Benchmark for Complex Cooking Scenarios
• Interactive Imitation Learning for Dexterous Robotic Manipulation: Challenges and Perspectives -- A Survey
• Dexterous Manipulation through Imitation Learning: A Survey

Discussions

• I’m prepared to change my view as the facts on the ground change, but at the present moment this strikes me as very “assume a can opener” [bsky, 1 points, 1 comments]

Related

• Dexterous Manipulation through Imitation Learning: A Survey
• Towards Human-level Dexterity via Robot Learning
• Survey of Learning-based Approaches for Robotic In-Hand Manipulation
• Dexterous Imitation Made Easy: A Learning-Based Framework for Efficient Dexterous Manipulation
• DexterityGen: Foundation Controller for Unprecedented Dexterity