Wenzhen Yuan

Ruohan Gao, Zilin Si, Yen-Yu Chang et al. · stanford

Objects play a crucial role in our everyday activities. Though multisensory object-centric learning has shown great potential lately, the modeling of objects in prior work is rather unrealistic. ObjectFolder 1.0 is a recent dataset that introduces 100 virtualized objects with visual, acoustic, and tactile sensory data. However, the dataset is small in scale and the multisensory data is of limited quality, hampering generalization to real-world scenarios. We present ObjectFolder 2.0, a large-scale, multisensory dataset of common household objects in the form of implicit neural representations that significantly enhances ObjectFolder 1.0 in three aspects. First, our dataset is 10 times larger in the amount of objects and orders of magnitude faster in rendering time. Second, we significantly improve the multisensory rendering quality for all three modalities. Third, we show that models learned from virtual objects in our dataset successfully transfer to their real-world counterparts in three challenging tasks: object scale estimation, contact localization, and shape reconstruction. ObjectFolder 2.0 offers a new path and testbed for multisensory learning in computer vision and robotics. The dataset is available at https://github.com/rhgao/ObjectFolder.

Jiacheng Ruan, Wenzhen Yuan, Zehao Lin et al.

Large visual-language models (LVLMs) have achieved great success in multiple applications. However, they still encounter challenges in complex scenes, especially those involving camouflaged objects. This is primarily due to the lack of samples related to camouflaged scenes in the training dataset. To mitigate this issue, we construct the MM-CamObj dataset for the first time, comprising two subsets: CamObj-Align and CamObj-Instruct. Specifically, CamObj-Align contains 11,363 image-text pairs, and it is designed for VL alignment and injecting rich knowledge of camouflaged scenes into LVLMs. CamObj-Instruct is collected for fine-tuning the LVLMs with improved instruction-following capabilities, and it includes 11,363 images and 68,849 conversations with diverse instructions. Based on the MM-CamObj dataset, we propose the CamObj-Llava, an LVLM specifically designed for addressing tasks in camouflaged scenes. To facilitate our model's effective acquisition of knowledge about camouflaged objects and scenes, we introduce a curriculum learning strategy with six distinct modes. Additionally, we construct the CamObj-Bench to evaluate the existing LVLMs' capabilities of understanding, recognition, localization and count in camouflage scenes. This benchmark includes 600 images and 7 tasks, with a total of 9,449 questions. Extensive experiments are conducted on the CamObj-Bench with CamObj-Llava, 8 existing open-source and 3 closed-source LVLMs. Surprisingly, the results indicate that our model achieves a 25.84% improvement in 4 out of 7 tasks compared to GPT-4o. Code and datasets will be available at https://github.com/JCruan519/MM-CamObj.

Fengyu Yang, Chenyang Ma, Jiacheng Zhang et al.

The ability to associate touch with sight is essential for tasks that require physically interacting with objects in the world. We propose a dataset with paired visual and tactile data called Touch and Go, in which human data collectors probe objects in natural environments using tactile sensors, while simultaneously recording egocentric video. In contrast to previous efforts, which have largely been confined to lab settings or simulated environments, our dataset spans a large number of "in the wild" objects and scenes. To demonstrate our dataset's effectiveness, we successfully apply it to a variety of tasks: 1) self-supervised visuo-tactile feature learning, 2) tactile-driven image stylization, i.e., making the visual appearance of an object more consistent with a given tactile signal, and 3) predicting future frames of a tactile signal from visuo-tactile inputs.

Shubham Kanitkar, Helen Jiang, Wenzhen Yuan

When humans grasp objects in the real world, we often move our arms to hold the object in a different pose where we can use it. In contrast, typical lab settings only study the stability of the grasp immediately after lifting, without any subsequent re-positioning of the arm. However, the grasp stability could vary widely based on the object's holding pose, as the gravitational torque and gripper contact forces could change completely. To facilitate the study of how holding poses affect grasp stability, we present PoseIt, a novel multi-modal dataset that contains visual and tactile data collected from a full cycle of grasping an object, re-positioning the arm to one of the sampled poses, and shaking the object. Using data from PoseIt, we can formulate and tackle the task of predicting whether a grasped object is stable in a particular held pose. We train an LSTM classifier that achieves 85% accuracy on the proposed task. Our experimental results show that multi-modal models trained on PoseIt achieve higher accuracy than using solely vision or tactile data and that our classifiers can also generalize to unseen objects and poses.

Yuanzhe Liu, Jingyuan Zhu, Yuchen Mo et al.

Recent advancements in vision-language-action (VLA) models have shown promise in robotic manipulation, yet they continue to struggle with long-horizon, multi-step tasks. Existing methods lack internal reasoning mechanisms that can identify task-relevant interaction cues or track progress within a subtask, leading to critical execution errors such as repeated actions, missed steps, and premature termination. To address these challenges, we introduce PALM, a VLA framework that structures policy learning around interaction-centric affordance reasoning and subtask progress cues. PALM distills complementary affordance representations that capture object relevance, contact geometry, spatial placements, and motion dynamics, and serve as task-relevant anchors for visuomotor control. To further stabilize long-horizon execution, PALM predicts continuous within-subtask progress, enabling seamless subtask transitions. Across extensive simulation and real-world experiments, PALM consistently outperforms baselines, achieving a 91.8% success rate on LIBERO-LONG, a 12.5% improvement in average length on CALVIN ABC->D, and a 2x improvement over real-world baselines across three long-horizon generalization settings.

Wenzhen Yuan, Wutao Xiong, Fanchen Yu et al.

Multi-agent systems (MAS) demonstrate clear advantages in tackling complex problems by coordinating diverse agents and external tools. However, most existing orchestration methods rely on static workflows or serial agent scheduling, and are further constrained by heterogeneous interface protocols between tools and agents. This leads to high system complexity and poor extensibility. To mitigate these issues, we propose Agent-as-Tool, a unified parallel orchestration paradigm that abstracts both agents and tools into a standardized, learnable action space with protocol normalization and explicit state feedback. Building on this paradigm, we train a lightweight orchestrator, ParaManager, which decouples planning decisions from subtask solving, enabling state-aware parallel subtask decomposition, delegation, and asynchronous execution. For training, we adopt a two-stage ParaManager training pipeline. It improves robustness by incorporating supervised fine-tuning (SFT) trajectories equipped with recovery mechanisms, and further applies reinforcement learning (RL) to achieve an optimal balance among task success, protocol compliance, diversity, and reasoning efficiency. Experiments show that ParaManager achieves strong performance across multiple benchmarks and exhibits robust generalization under unseen model pools.

Mohammad Amin Mirzaee, Harsh Gupta, Wenzhen Yuan

Despite advances in dexterous hand manipulation, robotic hand design is still largely decoupled from task-driven evaluation and control, limiting systematic optimization. Existing robotic hand co-design approaches are often limited in scope, optimizing a small subset of design parameters. We introduce a comprehensive parametric framework for robotic hand generation that unifies palm structure, finger kinematics, fingertip geometry, and fine-scale surface curvatures within a single design space. Fine geometric features are introduced through parametric surface deformation kernels that directly influence contact interactions. We validate the framework on design optimization in grasp stability tasks in simulation and real-world dynamic scenarios. Our framework produces simulation- and fabrication-ready hand models and will be released as open-source to enable rapid design iteration for dexterous hand co-design optimization frameworks and cross-embodiment policy training and control research.

Jiacheng Ruan, Wenzhen Yuan, Xian Gao et al.

Although large visual-language models (LVLMs) have demonstrated strong performance in multimodal tasks, errors may occasionally arise due to biases during the reasoning process. Recently, reward models (RMs) have become increasingly pivotal in the reasoning process. Specifically, process RMs evaluate each reasoning step, outcome RMs focus on the assessment of reasoning results, and critique RMs perform error analysis on the entire reasoning process, followed by corrections. However, existing benchmarks for vision-language RMs (VLRMs) typically assess only a single aspect of their capabilities (e.g., distinguishing between two answers), thus limiting the all-round evaluation and restricting the development of RMs in the visual-language domain. To address this gap, we propose a comprehensive and challenging benchmark, dubbed as VLRMBench, encompassing 12,634 questions. VLRMBench is constructed based on three distinct types of datasets, covering mathematical reasoning, hallucination understanding, and multi-image understanding. We design 12 tasks across three major categories, focusing on evaluating VLRMs in the aspects of process understanding, outcome judgment, and critique generation. Extensive experiments are conducted on 21 open-source models and 5 advanced closed-source models, highlighting the challenges posed by VLRMBench. For instance, in the `Forecasting Future', a binary classification task, the advanced GPT-4o achieves only a 76.0% accuracy. Additionally, we perform comprehensive analytical studies, offering valuable insights for the future development of VLRMs. We anticipate that VLRMBench will serve as a pivotal benchmark in advancing VLRMs. Code and datasets will be available at https://github.com/JCruan519/VLRMBench.

Jacqueline Yau, Katherine J. Mimnaugh, Evan G. Center et al.

Cybersickness poses a serious challenge for users of virtual reality (VR) technology. Consequently, there has been significant effort to track its occurrence during VR use with passive measures like brain activity recorded through electroencephalogram (EEG). To classify cybersickness accurately, including in real time, machine learning algorithms which can extract meaningful signals from the rest of the brain data will be required. However, EEG datasets are typically very small and very high in variability between participants, which makes building effective models extremely challenging. To address these concerns, we first introduce a framework for neural networks which has subject-adaptive training with calibration and interpretation for classification given limited and imbalanced EEG data. Which features the models determine are most useful can be visualized by plotting interpretability maps from integrated gradients and class activation. The framework is demonstrated here with convolutional neural networks and transformer models. Using a set of brain data recorded with EEG while participants viewed a stimulus in VR designed to elicit cybersickness, we show which spatio-temporal EEG features (from electrodes and time steps) were most important for discomfort classification. Across 12 runs of our framework with three different neural networks over multiple random seeds, the models consistently pointed to the same scalp locations as having patterns of brain data that were the most helpful in determining whether or not a sample of EEG data belonged to someone who was experiencing cybersickness. These results help clarify a hidden pattern in other related research and can be used as tagged features for better real-time cybersickness classification with EEG in the future. We provide our code at [anonymized] to enable feature interpretation across different neural network architectures.

Langzhe Gu, Hung-Jui Huang, Mohamad Qadri et al.

Accurate object geometry estimation is essential for many downstream tasks, including robotic manipulation and physical interaction. Although vision is the dominant modality for shape perception, it becomes unreliable under occlusions or challenging lighting conditions. In such scenarios, tactile sensing provides direct geometric information through physical contact. However, reconstructing global 3D geometry from sparse local touches alone is fundamentally underconstrained. We present TouchAnything, a framework that leverages a pretrained large-scale 2D vision diffusion model as a semantic and geometric prior for 3D reconstruction from sparse tactile measurements. Unlike prior work that trains category-specific reconstruction networks or learns diffusion models directly from tactile data, we transfer the geometric knowledge encoded in pretrained visual diffusion models to the tactile domain. Given sparse contact constraints and a coarse class-level description of the object, we formulate reconstruction as an optimization problem that enforces tactile consistency while guiding solutions toward shapes consistent with the diffusion prior. Our method reconstructs accurate geometries from only a few touches, outperforms existing baselines, and enables open-world 3D reconstruction of previously unseen object instances. Our project page is https://grange007.github.io/touchanything .

Wenzhen Yuan, Shengji Tang, Weihao Lin et al.

Reinforcement learning (RL) has significantly enhanced the reasoning capabilities of large language models (LLMs), but its reliance on expensive human-labeled data or complex reward models severely limits scalability. While existing self-feedback methods aim to address this problem, they are constrained by the capabilities of a single model, which can lead to overconfidence in incorrect answers, reward hacking, and even training collapse. To this end, we propose Reinforcement Learning from Coevolutionary Collective Feedback (RLCCF), a novel RL framework that enables multi-model collaborative evolution without external supervision. Specifically, RLCCF optimizes the ability of a model collective by maximizing its Collective Consistency (CC), which jointly trains a diverse ensemble of LLMs and provides reward signals by voting on collective outputs. Moreover, each model's vote is weighted by its Self-Consistency (SC) score, ensuring that more confident models contribute more to the collective decision. Benefiting from the diverse output distributions and complementary abilities of multiple LLMs, RLCCF enables the model collective to continuously enhance its reasoning ability through coevolution. Experiments on four mainstream open-source LLMs across four mathematical reasoning benchmarks demonstrate that our framework yields significant performance gains, achieving an average relative improvement of 16.72\% in accuracy. Notably, RLCCF not only improves the performance of individual models but also enhances the group's majority-voting accuracy by 4.51\%, demonstrating its ability to extend the collective capability boundary of the model collective.

Zilin Si, Wenzhen Yuan

Simulation is widely used in robotics for system verification and large-scale data collection. However, simulating sensors, including tactile sensors, has been a long-standing challenge. In this paper, we propose Taxim, a realistic and high-speed simulation model for a vision-based tactile sensor, GelSight. A GelSight sensor uses a piece of soft elastomer as the medium of contact and embeds optical structures to capture the deformation of the elastomer, which infers the geometry and forces applied at the contact surface. We propose an example-based method for simulating GelSight: we simulate the optical response to the deformation with a polynomial look-up table. This table maps the deformed geometries to pixel intensity sampled by the embedded camera. In order to simulate the surface markers' motion that is caused by the surface stretch of the elastomer, we apply the linear elastic deformation theory and the superposition principle. The simulation model is calibrated with less than 100 data points from a real sensor. The example-based approach enables the model to easily migrate to other GelSight sensors or its variations. To the best of our knowledge, our simulation framework is the first to incorporate marker motion field simulation that derives from elastomer deformation together with the optical simulation, creating a comprehensive and computationally efficient tactile simulation framework. Experiments reveal that our optical simulation has the lowest pixel-wise intensity errors compared to prior work and can run online with CPU computing. Our code and supplementary materials are open-sourced at https://github.com/CMURoboTouch/Taxim.

Arpit Agarwal, Tim Man, Wenzhen Yuan

Tactile sensing has seen a rapid adoption with the advent of vision-based tactile sensors. Vision-based tactile sensors provide high resolution, compact and inexpensive data to perform precise in-hand manipulation and human-robot interaction. However, the simulation of tactile sensors is still a challenge. In this paper, we built the first fully general optical tactile simulation system for a GelSight sensor using physics-based rendering techniques. We propose physically accurate light models and show in-depth analysis of individual components of our simulation pipeline. Our system outperforms previous simulation techniques qualitatively and quantitative on image similarity metrics. Our code and experimental data is open-sourced at https://labs.ri.cmu.edu/robotouch/tactile-optical-simulation/

Ruoyu Wang, Shiheng Wang, Songyu Du et al.

Soft bodies made from flexible and deformable materials are popular in many robotics applications, but their proprioceptive sensing has been a long-standing challenge. In other words, there has hardly been a method to measure and model the high-dimensional 3D shapes of soft bodies with internal sensors. We propose a framework to measure the high-resolution 3D shapes of soft bodies in real-time with embedded cameras. The cameras capture visual patterns inside a soft body, and a convolutional neural network (CNN) produces a latent code representing the deformation state, which can then be used to reconstruct the body's 3D shape using another neural network. We test the framework on various soft bodies, such as a Baymax-shaped toy, a latex balloon, and some soft robot fingers, and achieve real-time computation ($\leq$2.5ms/frame) for robust shape estimation with high precision ($\leq$1% relative error) and high resolution. We believe the method could be applied to soft robotics and human-robot interaction for proprioceptive shape sensing. Our code is available at https://ai4ce.github.io/Deep-Soft-Prorioception/

Seoyeon Lee, Mohammad Amin Mirzaee, Wenzhen Yuan

We present GelSphere, a spherical vision-based tactile sensor designed for real-time continuous surface scanning. Unlike traditional vision-based tactile sensors that can only sense locally and are damaged when slid across surfaces, and cylindrical tactile sensors that can only roll along a fixed direction, our design enables omnidirectional rolling on surfaces. We accomplish this through our novel sensing system design, which has steel balls inside the sensor, forming a bearing layer between the gel and the rigid housing that allows rolling motion in all axes. The sensor streams tactile images through Wi-Fi, with online large-surface reconstruction capabilities. We present quantitative results for both reconstruction accuracy and image fusion performance. The results show that our sensor maintains geometric fidelity and high reconstruction accuracy even under multi-directional rolling, enabling uninterrupted surface scanning.

Mohammad Amin Mirzaee, Hung-Jui Huang, Wenzhen Yuan

Scanning large-scale surfaces is widely demanded in surface reconstruction applications and detecting defects in industries' quality control and maintenance stages. Traditional vision-based tactile sensors have shown promising performance in high-resolution shape reconstruction while suffering limitations such as small sensing areas or susceptibility to damage when slid across surfaces, making them unsuitable for continuous sensing on large surfaces. To address these shortcomings, we introduce a novel vision-based tactile sensor designed for continuous surface sensing applications. Our design uses an elastomeric belt and two wheels to continuously scan the target surface. The proposed sensor showed promising results in both shape reconstruction and surface fusion, indicating its applicability. The dot product of the estimated and reference surface normal map is reported over the sensing area and for different scanning speeds. Results indicate that the proposed sensor can rapidly scan large-scale surfaces with high accuracy at speeds up to 45 mm/s.

Harsh Gupta, Yuchen Mo, Shengmiao Jin et al.

High-resolution tactile sensors have become critical for embodied perception and robotic manipulation. However, a key challenge in the field is the lack of transferability between sensors due to design and manufacturing variations, which result in significant differences in tactile signals. This limitation hinders the ability to transfer models or knowledge learned from one sensor to another. To address this, we introduce a novel method for extracting Sensor-Invariant Tactile Representations (SITR), enabling zero-shot transfer across optical tactile sensors. Our approach utilizes a transformer-based architecture trained on a diverse dataset of simulated sensor designs, allowing it to generalize to new sensors in the real world with minimal calibration. Experimental results demonstrate the method's effectiveness across various tactile sensing applications, facilitating data and model transferability for future advancements in the field.

Ruihan Gao, Kangle Deng, Gengshan Yang et al.

3D generation methods have shown visually compelling results powered by diffusion image priors. However, they often fail to produce realistic geometric details, resulting in overly smooth surfaces or geometric details inaccurately baked in albedo maps. To address this, we introduce a new method that incorporates touch as an additional modality to improve the geometric details of generated 3D assets. We design a lightweight 3D texture field to synthesize visual and tactile textures, guided by 2D diffusion model priors on both visual and tactile domains. We condition the visual texture generation on high-resolution tactile normals and guide the patch-based tactile texture refinement with a customized TextureDreambooth. We further present a multi-part generation pipeline that enables us to synthesize different textures across various regions. To our knowledge, we are the first to leverage high-resolution tactile sensing to enhance geometric details for 3D generation tasks. We evaluate our method in both text-to-3D and image-to-3D settings. Our experiments demonstrate that our method provides customized and realistic fine geometric textures while maintaining accurate alignment between two modalities of vision and touch.

Arpit Agarwal, Mohammad Amin Mirzaee, Xiping Sun et al.

GelSight family of vision-based tactile sensors has proven to be effective for multiple robot perception and manipulation tasks. These sensors are based on an internal optical system and an embedded camera to capture the deformation of the soft sensor surface, inferring the high-resolution geometry of the objects in contact. However, customizing the sensors for different robot hands requires a tedious trial-and-error process to re-design the optical system. In this paper, we formulate the GelSight sensor design process as a systematic and objective-driven design problem and perform the design optimization with a physically accurate optical simulation. The method is based on modularizing and parameterizing the sensor's optical components and designing four generalizable objective functions to evaluate the sensor. We implement the method with an interactive and easy-to-use toolbox called OptiSense Studio. With the toolbox, non-sensor experts can quickly optimize their sensor design in both forward and inverse ways following our predefined modules and steps. We demonstrate our system with four different GelSight sensors by quickly optimizing their initial design in simulation and transferring it to the real sensors.

Shengmiao Jin, Yuchen Mo, Wenzhen Yuan

Manipulating arbitrary objects in unstructured environments is a significant challenge in robotics, primarily due to difficulties in determining an object's center of mass. This paper introduces U-GRAPH: Uncertainty-Guided Rotational Active Perception with Haptics, a novel framework to enhance the center of mass estimation using active perception. Traditional methods often rely on single interaction and are limited by the inherent inaccuracies of Force-Torque (F/T) sensors. Our approach circumvents these limitations by integrating a Bayesian Neural Network (BNN) to quantify uncertainty and guide the robotic system through multiple, information-rich interactions via grid search and a neural network that scores each action. We demonstrate the remarkable generalizability and transferability of our method with training on a small dataset with limited variation yet still perform well on unseen complex real-world objects.

Akhil Padmanabha, Arpit Agarwal, Catherine Li et al.

Three-dimensional (3-D) skin surface reconstruction offers promise for objective and quantitative dermatological assessment, but no portable, high-resolution device exists that has been validated and used for depth reconstruction across various body locations. We present a compact 3-D skin reconstruction probe based on GelSight tactile imaging with a custom elastic gel and a learning-based reconstruction algorithm for micron-level wrinkle height estimation. Our probe, integrated into a handheld probe with force sensing for consistent contact, achieves a mean absolute error of 12.55 micron on wrinkle-like test objects. In a study with 15 participants without skin disorders, we provide the first validated wrinkle depth metrics across multiple body regions. We further demonstrate statistically significant reductions in wrinkle height at three locations following over-the-counter moisturizer application. Our work offers a validated tool for clinical and cosmetic skin analysis, with potential applications in diagnosis, treatment monitoring, and skincare efficacy evaluation.

Hung-Jui Huang, Mohammad Amin Mirzaee, Michael Kaess et al.

Accurately perceiving an object's pose and shape is essential for precise grasping and manipulation. Compared to common vision-based methods, tactile sensing offers advantages in precision and immunity to occlusion when tracking and reconstructing objects in contact. This makes it particularly valuable for in-hand and other high-precision manipulation tasks. In this work, we present GelSLAM, a real-time 3D SLAM system that relies solely on tactile sensing to estimate object pose over long periods and reconstruct object shapes with high fidelity. Unlike traditional point cloud-based approaches, GelSLAM uses tactile-derived surface normals and curvatures for robust tracking and loop closure. It can track object motion in real time with low error and minimal drift, and reconstruct shapes with submillimeter accuracy, even for low-texture objects such as wooden tools. GelSLAM extends tactile sensing beyond local contact to enable global, long-horizon spatial perception, and we believe it will serve as a foundation for many precise manipulation tasks involving interaction with objects in hand. The video demo is available on our website: https://joehjhuang.github.io/gelslam.

Ruihan Gao, Wenzhen Yuan, Jun-Yan Zhu

Deep generative models have various content creation applications such as graphic design, e-commerce, and virtual Try-on. However, current works mainly focus on synthesizing realistic visual outputs, often ignoring other sensory modalities, such as touch, which limits physical interaction with users. In this work, we leverage deep generative models to create a multi-sensory experience where users can touch and see the synthesized object when sliding their fingers on a haptic surface. The main challenges lie in the significant scale discrepancy between vision and touch sensing and the lack of explicit mapping from touch sensing data to a haptic rendering device. To bridge this gap, we collect high-resolution tactile data with a GelSight sensor and create a new visuotactile clothing dataset. We then develop a conditional generative model that synthesizes both visual and tactile outputs from a single sketch. We evaluate our method regarding image quality and tactile rendering accuracy. Finally, we introduce a pipeline to render high-quality visual and tactile outputs on an electroadhesion-based haptic device for an immersive experience, allowing for challenging materials and editable sketch inputs.

Sudharshan Suresh, Zilin Si, Joshua G. Mangelson et al.

Knowledge of 3-D object shape is of great importance to robot manipulation tasks, but may not be readily available in unstructured environments. While vision is often occluded during robot-object interaction, high-resolution tactile sensors can give a dense local perspective of the object. However, tactile sensors have limited sensing area and the shape representation must faithfully approximate non-contact areas. In addition, a key challenge is efficiently incorporating these dense tactile measurements into a 3-D mapping framework. In this work, we propose an incremental shape mapping method using a GelSight tactile sensor and a depth camera. Local shape is recovered from tactile images via a learned model trained in simulation. Through efficient inference on a spatial factor graph informed by a Gaussian process, we build an implicit surface representation of the object. We demonstrate visuo-tactile mapping in both simulated and real-world experiments, to incrementally build 3-D reconstructions of household objects.

Raj Kolamuri, Zilin Si, Yufan Zhang et al.

Rotational displacement about the grasping point is a common grasp failure when an object is grasped at a location away from its center of gravity. Tactile sensors with soft surfaces, such as GelSight sensors, can detect the rotation patterns on the contacting surfaces when the object rotates. In this work, we propose a model-based algorithm that detects those rotational patterns and measures rotational displacement using the GelSight sensor. We also integrate the rotation detection feedback into a closed-loop regrasping framework, which detects the rotational failure of grasp in an early stage and drives the robot to a stable grasp pose. We validate our proposed rotation detection algorithm and grasp-regrasp system on self-collected dataset and online experiments to show how our approach accurately detects the rotation and increases grasp stability.

Jihong Zhu, Andrea Cherubini, Claire Dune et al.

Deformable object manipulation (DOM) is an emerging research problem in robotics. The ability to manipulate deformable objects endows robots with higher autonomy and promises new applications in the industrial, services, and healthcare sectors. However, compared to rigid object manipulation, the manipulation of deformable objects is considerably more complex, and is still an open research problem. Addressing DOM challenges demand breakthroughs in almost all aspects of robotics, namely hardware design, sensing, (deformation) modeling, planning, and control. In this article, we review recent advances and highlight the main challenges when considering deformation in each sub-field. A particular focus of our paper lies in the discussions of these challenges and proposing future directions of research.

Tingguang Li, Krishnan Srinivasan, Max Qing-Hu Meng et al.

Robotic in-hand manipulation has been a long-standing challenge due to the complexity of modelling hand and object in contact and of coordinating finger motion for complex manipulation sequences. To address these challenges, the majority of prior work has either focused on model-based, low-level controllers or on model-free deep reinforcement learning that each have their own limitations. We propose a hierarchical method that relies on traditional, model-based controllers on the low-level and learned policies on the mid-level. The low-level controllers can robustly execute different manipulation primitives (reposing, sliding, flipping). The mid-level policy orchestrates these primitives. We extensively evaluate our approach in simulation with a 3-fingered hand that controls three degrees of freedom of elongated objects. We show that our approach can move objects between almost all the possible poses in the workspace while keeping them firmly grasped. We also show that our approach is robust to inaccuracies in the object models and to observation noise. Finally, we show how our approach generalizes to objects of other shapes.

Negin Heravi, Wenzhen Yuan, Allison M. Okamura et al.

Rich haptic sensory feedback in response to user interactions is desirable for an effective, immersive virtual reality or teleoperation system. However, this feedback depends on material properties and user interactions in a complex, non-linear manner. Therefore, it is challenging to model the mapping from material and user interactions to haptic feedback in a way that generalizes over many variations of the user's input. Current methodologies are typically conditioned on user interactions, but require a separate model for each material. In this paper, we present a learned action-conditional model that uses data from a vision-based tactile sensor (GelSight) and user's action as input. This model predicts an induced acceleration that could be used to provide haptic vibration feedback to a user. We trained our proposed model on a publicly available dataset (Penn Haptic Texture Toolkit) that we augmented with GelSight measurements of the different materials. We show that a unified model over all materials outperforms previous methods and generalizes to new actions and new instances of the material categories in the dataset.

Shaoxiong Wang, Jiajun Wu, Xingyuan Sun et al.

Perceiving accurate 3D object shape is important for robots to interact with the physical world. Current research along this direction has been primarily relying on visual observations. Vision, however useful, has inherent limitations due to occlusions and the 2D-3D ambiguities, especially for perception with a monocular camera. In contrast, touch gets precise local shape information, though its efficiency for reconstructing the entire shape could be low. In this paper, we propose a novel paradigm that efficiently perceives accurate 3D object shape by incorporating visual and tactile observations, as well as prior knowledge of common object shapes learned from large-scale shape repositories. We use vision first, applying neural networks with learned shape priors to predict an object's 3D shape from a single-view color image. We then use tactile sensing to refine the shape; the robot actively touches the object regions where the visual prediction has high uncertainty. Our method efficiently builds the 3D shape of common objects from a color image and a small number of tactile explorations (around 10). Our setup is easy to apply and has potentials to help robots better perform grasping or manipulation tasks on real-world objects.

Roberto Calandra, Andrew Owens, Dinesh Jayaraman et al.

For humans, the process of grasping an object relies heavily on rich tactile feedback. Most recent robotic grasping work, however, has been based only on visual input, and thus cannot easily benefit from feedback after initiating contact. In this paper, we investigate how a robot can learn to use tactile information to iteratively and efficiently adjust its grasp. To this end, we propose an end-to-end action-conditional model that learns regrasping policies from raw visuo-tactile data. This model -- a deep, multimodal convolutional network -- predicts the outcome of a candidate grasp adjustment, and then executes a grasp by iteratively selecting the most promising actions. Our approach requires neither calibration of the tactile sensors, nor any analytical modeling of contact forces, thus reducing the engineering effort required to obtain efficient grasping policies. We train our model with data from about 6,450 grasping trials on a two-finger gripper equipped with GelSight high-resolution tactile sensors on each finger. Across extensive experiments, our approach outperforms a variety of baselines at (i) estimating grasp adjustment outcomes, (ii) selecting efficient grasp adjustments for quick grasping, and (iii) reducing the amount of force applied at the fingers, while maintaining competitive performance. Finally, we study the choices made by our model and show that it has successfully acquired useful and interpretable grasping behaviors.

Shan Luo, Wenzhen Yuan, Edward Adelson et al.

Vision and touch are two of the important sensing modalities for humans and they offer complementary information for sensing the environment. Robots could also benefit from such multi-modal sensing ability. In this paper, addressing for the first time (to the best of our knowledge) texture recognition from tactile images and vision, we propose a new fusion method named Deep Maximum Covariance Analysis (DMCA) to learn a joint latent space for sharing features through vision and tactile sensing. The features of camera images and tactile data acquired from a GelSight sensor are learned by deep neural networks. But the learned features are of a high dimensionality and are redundant due to the differences between the two sensing modalities, which deteriorates the perception performance. To address this, the learned features are paired using maximum covariance analysis. Results of the algorithm on a newly collected dataset of paired visual and tactile data relating to cloth textures show that a good recognition performance of greater than 90\% can be achieved by using the proposed DMCA framework. In addition, we find that the perception performance of either vision or tactile sensing can be improved by employing the shared representation space, compared to learning from unimodal data.

Wenzhen Yuan, Yuchen Mo, Shaoxiong Wang et al.

Humans represent and discriminate the objects in the same category using their properties, and an intelligent robot should be able to do the same. In this paper, we build a robot system that can autonomously perceive the object properties through touch. We work on the common object category of clothing. The robot moves under the guidance of an external Kinect sensor, and squeezes the clothes with a GelSight tactile sensor, then it recognizes the 11 properties of the clothing according to the tactile data. Those properties include the physical properties, like thickness, fuzziness, softness and durability, and semantic properties, like wearing season and preferred washing methods. We collect a dataset of 153 varied pieces of clothes, and conduct 6616 robot exploring iterations on them. To extract the useful information from the high-dimensional sensory output, we applied Convolutional Neural Networks (CNN) on the tactile data for recognizing the clothing properties, and on the Kinect depth images for selecting exploration locations. Experiments show that using the trained neural networks, the robot can autonomously explore the unknown clothes and learn their properties. This work proposes a new framework for active tactile perception system with vision-touch system, and has potential to enable robots to help humans with varied clothing related housework.

Roberto Calandra, Andrew Owens, Manu Upadhyaya et al.

A successful grasp requires careful balancing of the contact forces. Deducing whether a particular grasp will be successful from indirect measurements, such as vision, is therefore quite challenging, and direct sensing of contacts through touch sensing provides an appealing avenue toward more successful and consistent robotic grasping. However, in order to fully evaluate the value of touch sensing for grasp outcome prediction, we must understand how touch sensing can influence outcome prediction accuracy when combined with other modalities. Doing so using conventional model-based techniques is exceptionally difficult. In this work, we investigate the question of whether touch sensing aids in predicting grasp outcomes within a multimodal sensing framework that combines vision and touch. To that end, we collected more than 9,000 grasping trials using a two-finger gripper equipped with GelSight high-resolution tactile sensors on each finger, and evaluated visuo-tactile deep neural network models to directly predict grasp outcomes from either modality individually, and from both modalities together. Our experimental results indicate that incorporating tactile readings substantially improve grasping performance.

Siyuan Dong, Wenzhen Yuan, Edward Adelson

A GelSight sensor uses an elastomeric slab covered with a reflective membrane to measure tactile signals. It measures the 3D geometry and contact force information with high spacial resolution, and successfully helped many challenging robot tasks. A previous sensor, based on a semi-specular membrane, produces high resolution but with limited geometry accuracy. In this paper, we describe a new design of GelSight for robot gripper, using a Lambertian membrane and new illumination system, which gives greatly improved geometric accuracy while retaining the compact size. We demonstrate its use in measuring surface normals and reconstructing height maps using photometric stereo. We also use it for the task of slip detection, using a combination of information about relative motions on the membrane surface and the shear distortions. Using a robotic arm and a set of 37 everyday objects with varied properties, we find that the sensor can detect translational and rotational slip in general cases, and can be used to improve the stability of the grasp.

Wenzhen Yuan, Chenzhuo Zhu, Andrew Owens et al.

Hardness is among the most important attributes of an object that humans learn about through touch. However, approaches for robots to estimate hardness are limited, due to the lack of information provided by current tactile sensors. In this work, we address these limitations by introducing a novel method for hardness estimation, based on the GelSight tactile sensor, and the method does not require accurate control of contact conditions or the shape of objects. A GelSight has a soft contact interface, and provides high resolution tactile images of contact geometry, as well as contact force and slip conditions. In this paper, we try to use the sensor to measure hardness of objects with multiple shapes, under a loosely controlled contact condition. The contact is made manually or by a robot hand, while the force and trajectory are unknown and uneven. We analyze the data using a deep constitutional (and recurrent) neural network. Experiments show that the neural net model can estimate the hardness of objects with different shapes and hardness ranging from 8 to 87 in Shore 00 scale.

Wenzhen Yuan, Shaoxiong Wang, Siyuan Dong et al.

For machines to interact with the physical world, they must understand the physical properties of objects and materials they encounter. We use fabrics as an example of a deformable material with a rich set of mechanical properties. A thin flexible fabric, when draped, tends to look different from a heavy stiff fabric. It also feels different when touched. Using a collection of 118 fabric sample, we captured color and depth images of draped fabrics along with tactile data from a high resolution touch sensor. We then sought to associate the information from vision and touch by jointly training CNNs across the three modalities. Through the CNN, each input, regardless of the modality, generates an embedding vector that records the fabric's physical property. By comparing the embeddings, our system is able to look at a fabric image and predict how it will feel, and vice versa. We also show that a system jointly trained on vision and touch data can outperform a similar system trained only on visual data when tested purely with visual inputs.