Physical AI

Direct VLM-to-VLA Adaptation via Language-Action Grounding

CLAP (Causal Language-Action Prediction) turns pretrained VLMs into VLAs by prepending a natural-language action plan to each numeric action chunk, so action prediction is causally conditioned on it with no architectural modification, action expert, or extra annotation. 2B CLAP hits 90.8% on LIBERO (+14.9 pt over VLA-0) with 4-5 hours of fine-tuning.

Online Robot Manipulation Error Detection and Description

We formulate robot failure detection as online video temporal grounding, predicting the temporal boundaries and a cause-and-effect description of each manipulation failure as the video streams.

Robust Object Insertion via Masked Encoding and Soft Wrist

A soft wrist collects safe tactile demos, and masked trajectory encoding retrieves them to insert unseen pegs at 85% success.

Compliance-Enabled Tactile Phase Retrieval for Few-Shot Insertion Learning

A compliant wrist reveals insertion phases from touch, so phase-consistent retrieval lifts few-shot peg-in-hole success by 13 points.

A Slack-Compensating, Helical-Buckling-Attenuating Flexible-Shaft Transmission for Lightweight Multi-DoF Manipulation

This study proposes SHAFT, a flexible-shaft transmission mechanism that compensates for slack with a proximal tensioner to suppress helical buckling in lightweight, remotely actuated manipulators. Compared with a flexible shaft without a tensioner, SHAFT improves transmission efficiency by about 30% and reduces angular transmission error by about 65% on a path with four 90-degree bends. Using SHAFT, we built a 6-DoF arm with a 1-DoF gripper whose arm module weighs only 280 g.

A Layout-Aware Benchmark Reveals Memory Layout Matters for Graph-based ANNS on GPU

Decoupling graph topology from memory layout on GPU. Vertex reordering alone yields up to 80% QPS gain at equal recall.

Can Compact Language Models Search Like Agents? Distillation-Guided Policy Optimization for Preserving Agentic RAG Capabilities

We investigate whether compact language models (0.5–1B) can acquire sophisticated agentic retrieval-augmented generation (RAG) behavior. While existing agentic RAG systems rely on multi-billion-parameter models, compact models suffer from poor initial outputs, sparse rewards, and unstable reinforcement learning dynamics. To address these challenges, we propose Distillation-Guided Policy Optimization (DGPO), a two-phase training framework combining cold-start knowledge distillation and selective teacher-guided reinforcement learning.

Grounded Vision-Language Interpreter for Integrated Task and Motion Planning

While recent advances in vision-language models (VLMs) have accelerated the development of language-guided robot planners, their black-box nature often lacks safety guarantees and interpretability crucial for real-world deployment. This paper proposes ViLaIn-TAMP, a hybrid planning framework for enabling verifiable, interpretable, and autonomous robot behaviors.

A Flexible Wrist with Anisotropic and Switchable Stiffness for Contact-Rich Manipulation

CLAW is a lightweight, low-cost flexible wrist that improves learned contact-rich manipulation through passive 6-DoF compliance and switchable anisotropic stiffness.

Robust and Resilient Soft Robotic Object Insertion with Compliance-Enabled Contact Formation and Failure Recovery

By integrating VLM-guided failure recovery into compliance-enabled contact formation, we achieve robust and resilient soft robotic object insertion without force sensing or controller retraining.

Learning Diffusion Policies from Demonstrations For Compliant Contact-rich Manipulation

DIPCOM: Learning Diffusion Policies from Demonstrations For Compliant Contact-rich Manipulation. (Code to be released soon)

Soft Conical Universal Robotic Hand for Scooping Granular Media from Containers of Various Sizes

This study proposes SCU-Hand, a soft and reconfigurable end-effector designed to scoop powdered materials from containers of various sizes using a general-purpose robotic arm. By deforming a single thin sheet into a conical shape, the SCU-Hand achieves adaptive surface contact without complex sensors or control algorithms, while maintaining the stiffness needed for efficient scooping. Experimental results demonstrate that SCU-Hand achieves over 95% scooping performance across different container sizes and granular materials, outperforming conventional rigid tools by up to 20%.

Lightweight Four-Motor Prosthetic Hand Enabling Precision–Lateral Dexterous Manipulation

We developed a lightweight (311g) prosthetic hand that enables both basic hand postures and precision–lateral manipulation using only four motors. This design offers a solution for daily life activities that require rotational manipulation capabilities.

Worm Gear-based Adaptive Variable Elasticity for Decoupling Actuators from External Forces

present WAVE, a worm gear-based variable stiffness actuator that decouples external forces from actuators to achieve safe, compliant, and robust manipulation.

a miniature-sized wheeled robot for swarm robotics research

maru (= miniature assemblage adaptive robot unit) is a custom-made, miniature-sized, two-wheeled robot designed specifically for tabletop swarm robotics research. This platform enables researchers and hobbyists to explore the dynamics of swarm robotics in a controlled, tabletop environment.

Visuo-Tactile Zero-Shot Object Recognition with Vision-Language Model

null

Learning Variable Compliance Control From a Few Demonstrations for Bimanual Robot with Haptic Feedback Teleoperation System

CompACT: Learning Compliance Control via Action Chunking with Transformers. (Code to be released soon)

Pose Estimation of a Cable-Driven Serpentine Manipulator Utilizing Intrinsic Dynamics via Physical Reservoir Computing

explore a new approach to pose estimation, where intrinsic nonlinear dynamics of a lightweight cable-driven serpentine manipulator are exploited via physical reservoir computing.

Low-Cost Air Hockey Robot Using a Five-Bar Linkage Mechanism Driven by Position-Control Servomotors

explore a new challenge in the development of a low-cost air hockey robot for human-robot-interaction research.

Audio-Visual Self-Supervision for Frame-Level Player-wise Offensive Shot Detection in Table Tennis Matches

explore a new challenge in audio-visual self-supervision for frame-level player-wise offensive shot detection in table tennis matches.

Retrieval-Augmented Policy Training for Cooperative Manipulation by Mobile Robots

introduces the Multi-Agent Coordination Skill Database, allowing multiple mobile robots to efficiently use past memories to adapt to new tasks.

Language-Guided Pattern Formation for Swarm Robotics with Multi-Agent Reinforcement Learning

This paper explores how to leverage the vast knowledge encoded in large language models to tackle pattern formation challenges for swarm robotics systems.

SliceIt! - A Dual Simulator Framework for Learning Robot Food Slicing

A real2sim2real approach to learning robotic food slicing tasks with rigid robot manipulators

When to Replan? An Adaptive Replanning Strategy for Autonomous Navigation using Deep Reinforcement Learning

propose an adaptive replanning strategy using deep reinforcement learning

Robotic Object Insertion with a Soft Wrist through Sim-to-Real Privileged Training

zero-shot sim-to-real transfer for soft robotic object insertion.

Symmetry-aware Reinforcement Learning for Robotic Assembly under Partial Observability with a Soft Wrist

we propose to leverage the symmetry for sample efficiency by augmenting the training data and constructing auxiliary losses to force the agent to adhere to the symmetry.

An Electromagnetism-Inspired Method for Estimating In-Grasp Torque from Visuotactile Sensors

Tilt torque estimation for gel-based visuotactile sensors

Benchmarking Actor-Critic Deep Reinforcement Learning Algorithms for Robotics Control With Action Constraints

a benchmark for evaluating action-constrained reinforcement learning algorithms

Learning Dual-Arm Manipulation from Demonstration Translated from a Human and Robotic Arm

present a multi-modal robotic grinding system that utilizes both audio and visual feedback

Learning Robotic Assembly by Leveraging Physical Softness and Tactile Sensing

develop an anomaly detection model using a Variational Autoencoder to identify the timing of these transitions and employ learning and heuristic-based controllers to navigate the peg tip to the hole and perform the insertion.

Learning Robotic Powder Weighing from Simulation for Laboratory Automation

employs a sim-to-real transfer learning approach using a domain randomization (DR) technique for robotic powder weighing.

Robotic Powder Grinding with Audio-Visual Feedback for Laboratory Automation in Materials Science

present a multi-modal robotic grinding system that utilizes both audio and visual feedback

Periodic Multi-Agent Path Planning

Study of multi-agent path planning where agents appear periodically.

Counterfactual Fairness Filter for Fair-Delay Multi-Robot Navigation

explore a new problem setting and solution algorithms for fairness-aware multi-robot navigtaion tasks

Risk-aware Path Planning via Probabilistic Fusion of Traversability Prediction for Planetary Rovers on Heterogeneous Terrains

The official code respository for "Risk-aware Path Planning via Probabilistic Fusion of Traversability Prediction for Planetary Rovers on Heterogeneous Terrains" (ICRA 2023)

Fracture Anticipation by Breaking Reusable Fragile Objects

Plastic table-top cable-driven robotic arm with all motors located at the base link

explore a new challenge in transfer RL, where only a set of source policies collected under unknown diverse dynamics is available for learning a target task efficiently.

Learning to Construct Cooperative Timed Roadmaps for Multi-agent Path Planning in Continuous Spaces

explore data-driven multi-agent path planning in continuous space. Our approach learns a generative model from planning demonstrations to build effective roadmaps, which improves planning efficiency while keeping high success rate & solution quality.

Quasistatic contact-rich manipulation via linear complementarity quadratic programming

Novel online planner for contact-rich manipulation using linear complementarity quadratic programming

Prioritized Safe Interval Path Planning for Multi-Agent Pathfinding with Continuous Time on 2D Roadmaps

explore a challenging multi-agent pathfinding problem for hundreds of agents moving on a 2D roadmap with continuous time.

Path Planning using Neural A* Search

Novel data-driven search-based planner based on differentiable A*

Learning to Balance the Safety-Efficiency Trade-off in Interactive Crowd-aware Robot Navigation

A robot that learns when to ask a crowd to make room and when to detour, raising navigation success in dense crowds.

Multi-Source Policy Aggregation for Transfer Reinforcement Learning between Diverse Environmental Dynamics

explore a new challenge in transfer RL, where only a set of source policies collected under unknown diverse dynamics is available for learning a target task efficiently.

Training-Free Peg Insertion Using a General-Purpose Vision-Language Model with Tactile Feedback

A general-purpose vision-language model interprets visual and tactile observations and selects robot actions without additional tactile or manipulation training.