The field of robotics is witnessing significant advancements in manipulation and human-robot interaction. Researchers are focusing on developing more intuitive and flexible systems that can adapt to complex environments and tasks. One notable direction is the integration of affordance reasoning, which enables robots to understand the relationships between objects and actions. This has led to improvements in areas such as disassembly, grasping, and manipulation of deformable objects. Another key area of research is the development of more sophisticated control strategies, including hybrid force-position control and reinforcement learning-based approaches. These advancements have the potential to significantly improve the performance and autonomy of robots in various applications, including manufacturing, healthcare, and service robotics. Noteworthy papers in this area include Affordance-R1, which proposes a unified affordance grounding framework, and GraphCoT-VLA, which introduces a 3D spatial-aware reasoning vision-language-action model for robotic manipulation with ambiguous instructions.
Advancements in Robot Manipulation and Human-Robot Interaction
Sources
Affordance-Guided Dual-Armed Disassembly Teleoperation for Mating Parts
Modular Vacuum-Based Fixturing System for Adaptive Disassembly Workspace Integration
Incremental Language Understanding for Online Motion Planning of Robot Manipulators
Affordance-R1: Reinforcement Learning for Generalizable Affordance Reasoning in Multimodal Large Language Model
DexFruit: Dexterous Manipulation and Gaussian Splatting Inspection of Fruit
A Hybrid Force-Position Strategy for Shape Control of Deformable Linear Objects With Graph Attention Networks
In-situ Value-aligned Human-Robot Interactions with Physical Constraints
CoT-Pose: Chain-of-Thought Reasoning for 3D Pose Generation from Abstract Prompts
Grasp-HGN: Grasping the Unexpected
GraphCoT-VLA: A 3D Spatial-Aware Reasoning Vision-Language-Action Model for Robotic Manipulation with Ambiguous Instructions
Pose-RFT: Enhancing MLLMs for 3D Pose Generation via Hybrid Action Reinforcement Fine-Tuning
Selective Contrastive Learning for Weakly Supervised Affordance Grounding
MolmoAct: Action Reasoning Models that can Reason in Space
PCHands: PCA-based Hand Pose Synergy Representation on Manipulators with N-DoF
DiffPose-Animal: A Language-Conditioned Diffusion Framework for Animal Pose Estimation
Towards Affordance-Aware Robotic Dexterous Grasping with Human-like Priors
CaRoBio: 3D Cable Routing with a Bio-inspired Gripper Fingernail
Interpretable Robot Control via Structured Behavior Trees and Large Language Models
Few-shot Vision-based Human Activity Recognition with MLLM-based Visual Reinforcement Learning
A Semantic-Aware Framework for Safe and Intent-Integrative Assistance in Upper-Limb Exoskeletons