The field of robotic manipulation and sensing is moving towards more advanced and innovative solutions. Researchers are exploring new materials, designs, and control systems to improve the performance and adaptability of robots in various environments. Soft robotics, in particular, is gaining attention for its potential to create more flexible and human-like robots. Additionally, the development of new sensing technologies, such as visuotactile sensing, is enabling robots to better understand and interact with their surroundings. Noteworthy papers in this area include those that propose novel frameworks for robotic manipulation, such as A0, which leverages hierarchical affordance-aware diffusion models to decompose manipulation tasks into high-level spatial affordance understanding and low-level action execution. Another notable paper is ViTa-Zero, which introduces a zero-shot visuotactile pose estimation framework that combines visual and tactile information to estimate the 6D pose of objects.
Advances in Robotic Manipulation and Sensing
Sources
Leveraging Passive Compliance of Soft Robotics for Physical Human-Robot Collaborative Manipulation
CATCH-FORM-ACTer: Compliance-Aware Tactile Control and Hybrid Deformation Regulation-Based Action Transformer for Viscoelastic Object Manipulation
CATCH-FORM-3D: Compliance-Aware Tactile Control and Hybrid Deformation Regulation for 3D Viscoelastic Object Manipulation
Spectral Normalization for Lipschitz-Constrained Policies on Learning Humanoid Locomotion
TensorNEAT: A GPU-accelerated Library for NeuroEvolution of Augmenting Topologies
Road Grip Uncertainty Estimation Through Surface State Segmentation
Neural Fidelity Calibration for Informative Sim-to-Real Adaptation
BiFlex: A Passive Bimodal Stiffness Flexible Wrist for Manipulation in Unstructured Environments
ZeroGrasp: Zero-Shot Shape Reconstruction Enabled Robotic Grasping
3DAffordSplat: Efficient Affordance Reasoning with 3D Gaussians
CAP-Net: A Unified Network for 6D Pose and Size Estimation of Categorical Articulated Parts from a Single RGB-D Image
Improving Swimming Performance in Soft Robotic Fish with Distributed Muscles and Embedded Kinematic Sensing
Learning Transferable Friction Models and LuGre Identification via Physics Informed Neural Networks
Robust Visual Servoing under Human Supervision for Assembly Tasks
Practical Insights on Grasp Strategies for Mobile Manipulation in the Wild
A0: An Affordance-Aware Hierarchical Model for General Robotic Manipulation
Taccel: Scaling Up Vision-based Tactile Robotics via High-performance GPU Simulation
Force and Speed in a Soft Stewart Platform
ViTa-Zero: Zero-shot Visuotactile Object 6D Pose Estimation