arXiv subjects

Yangwei You

Publications and source records attributed to Yangwei You.

10 recordsLinked to original sources

arXiv2026

One Policy, Many Embodiments: Unified Camera-Centric Action Geometry Pre-training for Heterogeneous Embodied Manipulation

Scaling generalist vision-language-action (VLA) policies is severely bottlenecked by the inherent heterogeneity of embodied data, which spans diverse robot morphologies, camera configurations, and low-level action spaces. Existing paradigms typically address this mismatch through explicit action retargeting, human-to-robot video synthesis, or dataset-specific adaptation branches, fundamentally hindering the joint learning of a unified policy. We introduce UCAG-P, a camera-centric unified action formulation that structurally aligns heterogeneous embodied datasets into a shared geometric action space. Rather than treating robot-specific commands as the shared policy target, UCAG-P represents manipulation through camera-observable anchor motion in image and camera-frame coordinates, treating robot arms, humanoids, and human hands as different embodiments of a common action schema. A geometry-conditioned action translator combines predicted motion with target-embodiment kinematics to produce executable controls. The resulting decoupled architecture allows a shared VLA policy to learn transferable manipulation geometry while retaining embodiment-specific controllability. UCAG-P is trained on 4.03K hours of robot and simulation data and 2.34K hours of human demonstrations. A single checkpoint reaches 98.3% on LIBERO, 88.7% and 89.2% on RoboTwin Easy and Hard, 82.0% zero-shot on LIBERO-Plus, and 62.0% on RoboCasa GR-1, without benchmark-specific fine-tuning.

cs.RO↗

arXiv2026

ViTacPhys: Physical Property-Aware Grasping from Human Visual-Tactile Demonstrations

Recent vision-based action models have demonstrated strong capabilities in complex manipulation, but they rarely leverage explicit object physical properties to adapt their policies. We introduce ViTacPhys, a visual-tactile framework and data acquisition system that estimates object mass and friction-coefficient classes, together with continuous stiffness, from human manipulation demonstrations. Trained on data from 60 rigid and deformable objects, ViTacPhys combines temporal visual-tactile modeling, cross-attention multimodal fusion, and a semantic prior derived from a vision-language model. On seen objects, it achieves 97.2% mass classification accuracy, 98.8% friction-coefficient classification accuracy, and a stiffness mean absolute percentage error (MAPE) of 5.51%. On held-out objects from known categories, it achieves 87.5% mass accuracy, 97.5% friction-coefficient accuracy, and a stiffness MAPE of 9.08%. We transfer ViTacPhys from the human domain to the robot domain using limited robot teleoperation data, robot-style video augmentation, and human demonstrations with matched actions, and deploy it as an online module for adaptive grasping. The resulting physical-property-conditioned policy achieves total grasping success rates of 95.0% on in-distribution objects and 83.4% on out-of-distribution objects. For out-of-distribution objects successfully grasped by both methods, its force profiles are more consistent with human teleoperation than those produced by ACT. These results demonstrate the feasibility of explicitly estimating and conditioning on object physical properties for real-world adaptive grasping.

cs.RO↗

arXiv2025

TacRefineNet: Goal-Conditioned Tactile Grasp Refinement for Edge-Prominent Objects

Accurate final grasp alignment remains challenging for edge-prominent objects such as thin plates, discs, and rods, whose sparse contacts are easily occluded and poorly resolved by depth sensing. We present TacRefineNet, a tactile-only, goal-conditioned framework for local refinement along tactilely observable pose dimensions. Given current and target multi-finger tactile images and their corresponding hand-joint configurations, a Siamese policy network directly predicts corrective wrist pose increments. The hand iteratively opens, moves, and regrasps, forming an external-dexterity tactile servoing loop. Cross-combination training pairs current and target samples, allowing targets within the sampled pose range to be specified without retraining. We collect 156,007 simulated samples from 15 plates, discs, and rods and train the policy entirely in MuJoCo before zero-shot deployment to an 11-DoF five-fingered hand with piezoresistive sensors. On seen objects, the real system achieves 80.7\% and 59.3\% success under the $10^\circ$/10\,mm criterion for fixed and random targets, respectively; after five steps, the mean errors are approximately 5.2\,mm and $3.5^\circ$. Experiments further show continuous correction under long-horizon perturbations and limited within-category transfer to unseen objects, with reduced performance for symmetric or weakly discriminative contacts. Project website is available at https://sites.google.com/view/tacrefinenet

cs.RO↗

arXiv2025

DyDexHandover: Human-like Bimanual Dynamic Dexterous Handover using RGB-only Perception

Dynamic in air handover is a fundamental challenge for dual-arm robots, requiring accurate perception, precise coordination, and natural motion. Prior methods often rely on dynamics models, strong priors, or depth sensing, limiting generalization and naturalness. We present DyDexHandover, a novel framework that employs multi-agent reinforcement learning to train an end to end RGB based policy for bimanual object throwing and catching. To achieve more human-like behavior, the throwing policy is guided by a human policy regularization scheme, encouraging fluid and natural motion, and enhancing the generalization capability of the policy. A dual arm simulation environment was built in Isaac Sim for experimental evaluation. DyDexHandover achieves nearly 99 percent success on training objects and 75 percent on unseen objects, while generating human-like throwing and catching behaviors. To our knowledge, it is the first method to realize dual-arm in-air handover using only raw RGB perception.

cs.RO↗

arXiv2025

Dual-Actor Fine-Tuning of VLA Models: A Talk-and-Tweak Human-in-the-Loop Approach

Vision-language-action (VLA) models demonstrate strong generalization in robotic manipulation but face challenges in complex, real-world tasks. While supervised fine-tuning with demonstrations is constrained by data quality, reinforcement learning (RL) offers a promising alternative. We propose a human-in-the-loop dual-actor fine-tuning framework grounded in RL. The framework integrates a primary actor for robust multi-task performance with a refinement actor for latent-space adaptation. Beyond standard physical interventions, we introduce a lightweight talk-and-tweak scheme that converts human corrections into semantically grounded language commands, thereby generating a new dataset for policy learning. In real-world multi-task experiments, our approach achieves 100% success across three tasks within 101 minutes of online fine-tuning. For long-horizon tasks, it sustains a 50% success rate over 12 consecutive operations. Furthermore, the framework scales effectively to multi-robot training, achieving up to a 2 times improvement in efficiency when using dual robots. The experiment videos are available at https://sites.google.com/view/hil-daft/.

cs.RO↗

arXiv2025

FPC-VLA: A Vision-Language-Action Framework with a Supervisor for Failure Prediction and Correction

Robotic manipulation is a fundamental component of automation. However, traditional perception-planning pipelines often fall short in open-ended tasks due to limited flexibility, while the architecture of a single end-to-end Vision-Language-Action (VLA) offers promising capabilities but lacks crucial mechanisms for anticipating and recovering from failure. To address these challenges, we propose FPC-VLA, a dual-model framework that integrates VLA with a supervisor for failure prediction and correction. The supervisor evaluates action viability through vision-language queries and generates corrective strategies when risks arise, trained efficiently without manual labeling. A dual-stream fusion module further refines actions by leveraging past predictions. Evaluation results on multiple simulation platforms (SIMPLER and LIBERO) and robot embodiments (WidowX, Google Robot, Franka) show that FPC-VLA outperforms state-of-the-art models in both zero-shot and fine-tuned settings. Successful real-world deployments on diverse, long-horizon tasks confirm FPC-VLA's strong generalization and practical utility for building more reliable autonomous systems.

cs.RO↗

arXiv2024

State Estimation Transformers for Agile Legged Locomotion

We propose a state estimation method that can accurately predict the robot's privileged states to push the limits of quadruped robots in executing advanced skills such as jumping in the wild. In particular, we present the State Estimation Transformers (SET), an architecture that casts the state estimation problem as conditional sequence modeling. SET outputs the robot states that are hard to obtain directly in the real world, such as the body height and velocities, by leveraging a causally masked Transformer. By conditioning an autoregressive model on the robot's past states, our SET model can predict these privileged observations accurately even in highly dynamic locomotions. We evaluate our methods on three tasks -- running jumping, running backflipping, and running sideslipping -- on a low-cost quadruped robot, Cyberdog2. Results show that SET can outperform other methods in estimation accuracy and transferability in the simulation as well as success rates of jumping and triggering a recovery controller in the real world, suggesting the superiority of such a Transformer-based explicit state estimator in highly dynamic locomotion tasks.

cs.RO↗

arXiv2022

Navigation with Tactile Sensor for Natural Human-Robot Interaction

Tactile sensors have been introduced to a wide range of robotic tasks such as robot manipulation to mimic the sense of human touch. However, there has only been a few works that integrate tactile sensing into robot navigation. This paper describes a navigation system which allows robots to operate in crowded human-dense environments and behave with socially acceptable reactions by utilizing semantic and force information collected by embedded tactile sensors, RGB-D camera and LiDAR. Compliance control is implemented based on artificial potential fields considering not only laser scan but also force reading from tactile sensors which promises a fast and reliable response to any possible collision. In contrast to cameras, LiDAR and other non-contact sensors, tactile sensors can directly interact with humans and can be used to accept social cues akin to natural human behavior under the same situation. Furthermore, leveraging semantic segmentation from vision module, the robot is able to identify and, therefore assign varying social cost to different groups of humans enabling for socially conscious path planning. At the end of this paper, the proposed control strategy was validated successfully by testing several scenarios on an omni-directional robot in real world.

cs.RO↗

arXiv2022

Traversability analysis with vision and terrain probing for safe legged robot navigation

Inspired by human behavior when traveling over unknown terrain, this study proposes the use of probing strategies and integrates them into a traversability analysis framework to address safe navigation on unknown rough terrain. Our framework integrates collapsibility information into our existing traversability analysis, as vision and geometric information alone could be misled by unpredictable non-rigid terrains such as soft soil, bush area, or water puddles. With the new traversability analysis framework, our robot has a more comprehensive assessment of unpredictable terrain, which is critical for its safety in outdoor environments. The pipeline first identifies the terrain's geometric and semantic properties using an RGB-D camera and desired probing locations on questionable terrains. These regions are probed using a force sensor to determine the risk of terrain collapsing when the robot steps over it. This risk is formulated as a collapsibility metric, which estimates an unpredictable region's ground collapsibility. Thereafter, the collapsibility metric, together with geometric and semantic spatial data, is combined and analyzed to produce global and local traversability grid maps. These traversability grid maps tell the robot whether it is safe to step over different regions of the map. The grid maps are then utilized to generate optimal paths for the robot to safely navigate to its goal. Our approach has been successfully verified on a quadrupedal robot in both simulation and real-world experiments.

cs.RO↗

arXiv2020

Contact-Implicit Trajectory Optimization using an Analytically Solvable Contact Model for Locomotion on Variable Ground

This paper presents a novel contact-implicit trajectory optimization method using an analytically solvable contact model to enable planning of interactions with hard, soft, and slippery environments. Specifically, we propose a novel contact model that can be computed in closed-form, satisfies friction cone constraints and can be embedded into direct trajectory optimization frameworks without complementarity constraints. The closed-form solution decouples the computation of the contact forces from other actuation forces and this property is used to formulate a minimal direct optimization problem expressed with configuration variables only. Our simulation study demonstrates the advantages over the rigid contact model and a trajectory optimization approach based on complementarity constraints. The proposed model enables physics-based optimization for a wide range of interactions with hard, slippery, and soft grounds in a unified manner expressed by two parameters only. By computing trotting and jumping motions for a quadruped robot, the proposed optimization demonstrates the versatility for multi-contact motion planning on surfaces with different physical properties.

cs.RO↗