Searcharxiv⌕ Search

arXiv subjects

Edward H. Adelson

Publications and source records attributed to Edward H. Adelson.

At least 19 recordsLinked to original sources

Reins: Frontier Multimodal Models Driving Real Robot Arms on Long-Horizon Tasks, Zero-Shot

Frontier models such as GPT-6.1 Sol, GPT-6 Astra and Claude Opus 5.5 can now plan and call tools, yet they still struggle to drive real robots through long-horizon, contact-rich manipulation, blind while the arm moves and while they think. We present Reins, a harness in which an off-the-shelf model operates a dual-arm robot with 3D perception tools and atomic actions that report contact and grip. The model chains actions in one call, each with a condition that must hold, stopping at the first that fails; a second instance, the narrator, watches the cameras continuously and reports what changed with every result. From images, Reins also builds a simulation-ready twin of the scene for continual policy learning. With natural-language prompts and no task-specific code, GPT-6.1 Sol succeeds in 88% of real episodes on five tasks; without the harness, it fails every shell game, taking 8 times as long. Three frontier models perform alike but differ in tracking a scene that changes while the actor thinks. Code will be released upon publication.

cs.RO↗

ACTS: Learning Visuo-Tactile World Models from Human Physical Interaction

Abstract Action-conditioned world models for contact-rich manipulation require data that captures how contact forms, slides, and releases under varying force. Such data are slow to collect through robot teleoperation, which often lacks force feedback and is constrained by gripper geometry. We ask whether these dynamics can instead be learned from human physical interaction and transferred to a robot. We present ACTS, a visuo-tactile world model trained only on human interaction data. Operators manipulate handheld GelSight sensors while we record third-person and wrist video, tactile images, 6-DoF sensor motion, and tactile-derived normal force, yielding React, a 4.9-hour bimanual dataset across three objects. We represent actions at the tactile sensor as motion augmented by force change, providing a shared action space for human collection and robot execution. ACTS jointly predicts future scene, wrist, and tactile observations. On held-out human data, ACTS improves scene-view prediction by 0.7-2.0 dB PSNR over the stronger of two visuo-tactile world models trained from scratch on React. Without fine-tuning, in the robot teleoperation rollouts we inspected, predicted tactile imprints still appear and disappear with recorded contact. Visual transfer is less robust: camera predictions exceed frame copying on PushT, are mixed on Rope, and fall below it on Motherboard, while wrist views can blur or hallucinate structure. These results suggest that contact dynamics transfer more reliably than visual appearance from human to robot.

cs.RO↗

A Durable Vision-Based Tactile Fingertip for Robotic Manipulation

Currently available commercial vision-based tactile sensors provide rich contact information but remain vulnerable to abrasion and repeated concentrated loading, limiting their use in demanding robotic applications. This work presents a durable tactile fingertip comprising a soft silicone gel with a nonpigmented, textured, thin thermoplastic-polyurethane protective film and a replaceable sensing cartridge. Durability was evaluated using two accelerated laboratory procedures: a rotating-drum sanding test and a repetitive probe test applying 39.2 N (4.0 kgf) at 45 cycles per minute. Under the defined sanding conditions, the developed sensor reached the protective-film rupture endpoint after approximately 2-3 hours. During repetitive probe testing, all nine developed sensors remained functionally usable when testing was discontinued: seven after 5 days, one after 6 days, and one after 8 days. Commercial GelSight Mini and DIGIT specimens exhibited initial surface-film rupture after approximately 24-30 seconds of sanding and 25-35 minutes of repetitive loading. Damage to the developed sensor progressed gradually and produced little interference with tactile imaging at the test endpoints. These observations establish durability improvements of more than two orders of magnitude under the defined accelerated conditions. Combining increased durability, gradual degradation, and rapid cartridge replacement offers a practical approach to maintainable vision-based tactile sensing for demanding robotic applications.

cs.RO↗

CoStream: Composing Simple Behaviors for Generalizable Complex Manipulation

Long-horizon, contact-rich complex manipulation tasks, such as seating a GPU into a PCIe slot, demand both millimeter high precision and out-of-the-box generalization to new tasks. Existing paradigms struggle to satisfy both: classical pipelines use brittle, task-specific interfaces to achieve high-precision control but require costly pipeline redesigns to adapt to new tasks, whereas monolithic end-to-end policies provide better generalization but lack high precision on complex, out-of-distribution tasks unless retrained with new data. Both paradigms share an implicit assumption: once a manipulation capability is acquired, it must be deployed as a rigid pipeline or monolithic whole, rather than being freely decomposed and recomposed. In this paper, we show that complex manipulation capabilities can emerge naturally from the composition of simple, independent behaviors. Rather than deploying a monolithic policy or a rigid pipeline, we propose CoStream, a framework orchestrating foundation models and diverse sensing modalities into multiple composable core behaviors: a semantic behavior extracting spatial constraints via foundation models; a predictive behavior forecasting trajectories by tracking keypoints in imagined videos; and a reactive behavior providing high-frequency tactile and force corrections. On a shared $SE(3)$ interface, these outputs compose by right-multiplication into a single pose command at each control step, executed by a compliant controller. We demonstrate CoStream on 8 real-world tasks spanning everyday manipulation and precision assembly, with the strongest gains in contact-rich assembly and object transfer, and show robust recovery from manual perturbations during execution. Website: https://costream-simple.github.io

cs.RO↗

2D and 3D Grasp Planners for the GET Asymmetrical Gripper

In this paper, we introduce GET-2D-1.0, a fast grasp planner for the GET asymmetrical gripper that operates from a single-view RGB-D image, using the Ferrari-Canny metric and a novel sampling strategy, and GET-3D-1.0, a mesh-based method using a 3D gripper model and ray-tracing. We evaluate both grasp planners against baselines with physical experiments, which suggest that GET-2D-1.0 can improve over a bounding box baseline by over 40% in lift success, shake survival, and force resistance. Experiments with GET-3D-1.0 suggest slight improvement compared to GET-2D-1.0 on lift success and shake survival, but are more computationally expensive, averaging 17 seconds of planning compared to 683 ms for GET-2D-1.0.

cs.RO↗

ManipulationNet: An Infrastructure for Benchmarking Real-World Robot Manipulation with Physical Skill Challenges and Embodied Multimodal Reasoning

Dexterous manipulation enables robots to purposefully alter the physical world, transforming them from passive observers into active agents in unstructured environments. This capability is the cornerstone of physical artificial intelligence. Despite decades of advances in hardware, perception, control, and learning, progress toward general manipulation systems remains fragmented due to the absence of widely adopted standard benchmarks. The central challenge lies in reconciling the variability of the real world with the reproducibility and authenticity required for rigorous scientific evaluation. To address this, we introduce ManipulationNet, a global infrastructure that hosts real-world benchmark tasks for robotic manipulation. ManipulationNet delivers reproducible task setups through standardized hardware kits, and enables distributed performance evaluation via a unified software client that delivers real-time task instructions and collects benchmarking results. As a persistent and scalable infrastructure, ManipulationNet organizes benchmark tasks into two complementary tracks: 1) the Physical Skills Track, which evaluates low-level physical interaction skills, and 2) the Embodied Reasoning Track, which tests high-level reasoning and multimodal grounding abilities. This design fosters the systematic growth of an interconnected network of real-world abilities and skills, paving the path toward general robotic manipulation. By enabling comparable manipulation research in the real world at scale, this infrastructure establishes a sustainable foundation for measuring long-term scientific progress and identifying capabilities ready for real-world deployment.

cs.RO↗

Sashimi-Bot: Autonomous Tri-manual Advanced Manipulation and Cutting of Deformable Objects

Advanced robotic manipulation of deformable, volumetric objects remains one of the greatest challenges due to their pliancy, frailness, variability, and uncertainties during interaction. Motivated by these challenges, this article introduces Sashimi-Bot, an autonomous multi-robotic system for advanced manipulation and cutting, specifically the preparation of sashimi. The objects that we manipulate, salmon loins, are natural in origin and vary in size and shape, they are limp and deformable with poorly characterized elastoplastic parameters, while also being slippery and hard to hold. The three robots straighten the loin; grasp and hold the knife; cut with the knife in a slicing motion while cooperatively stabilizing the loin during cutting; and pick up the thin slices from the cutting board or knife blade. Our system combines deep reinforcement learning with in-hand tool shape manipulation, in-hand tool cutting, and feedback of visual and tactile information to achieve robustness to the variabilities inherent in this task. This work represents a milestone in robotic manipulation of deformable, volumetric objects that may inspire and enable a wide range of other real-world applications.

cs.RO↗

Object Recognition and Force Estimation with the GelSight Baby Fin Ray

Recent advances in soft robotic hands and tactile sensing have enabled both to perform an increasing number of complex tasks with the aid of machine learning. In particular, we presented the GelSight Baby Fin Ray in our previous work, which integrates a camera with a soft, compliant Fin Ray structure. Camera-based tactile sensing gives the GelSight Baby Fin Ray the ability to capture rich contact information like forces, object geometries, and textures. Moreover, our previous work showed that the GelSight Baby Fin Ray can dig through clutter, and classify in-shell nuts. To further examine the potential of the GelSight Baby Fin Ray, we leverage learning to distinguish nut-in-shell textures and to perform force and position estimation. We implement ablation studies with popular neural network structures, including ResNet50, GoogLeNet, and 3- and 5-layer convolutional neural network (CNN) structures. We conclude that machine learning is a promising technique to extract useful information from high-resolution tactile images and empower soft robotics to better understand and interact with the environments.

cs.RO↗

Grasp EveryThing (GET): 1-DoF, 3-Fingered Gripper with Tactile Sensing for Robust Grasping

We introduce the Grasp EveryThing (GET) gripper, a novel 1-DoF, 3-finger design for securely grasping objects of many shapes and sizes. Mounted on a standard parallel jaw actuator, the design features three narrow, tapered fingers arranged in a two-against-one configuration, where the two fingers converge into a V-shape. The GET gripper is more capable of conforming to object geometries and forming secure grasps than traditional designs with two flat fingers. Inspired by the principle of self-similarity, these V-shaped fingers enable secure grasping across a wide range of object sizes. Further to this end, fingers are parametrically designed for convenient resizing and interchangeability across robotic embodiments with a parallel jaw gripper. Additionally, we incorporate a rigid fingernail for ease in manipulating small objects. Tactile sensing can be integrated into the standalone finger via an externally-mounted camera. A neural network was trained to estimate normal force from tactile images with an average validation error of 1.3 N across a diverse set of geometries. In grasping 15 objects and performing 3 tasks via teleoperation, the GET fingers consistently outperformed standard flat fingers. All finger designs, compatible with multiple robotic embodiments, both incorporating and lacking tactile sensing, are available on GitHub.

cs.RO↗

The Feeling of Success: Does Touch Sensing Help Predict Grasp Outcomes?

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.

cs.RO↗

PoCo: Policy Composition from and for Heterogeneous Robot Learning

Training general robotic policies from heterogeneous data for different tasks is a significant challenge. Existing robotic datasets vary in different modalities such as color, depth, tactile, and proprioceptive information, and collected in different domains such as simulation, real robots, and human videos. Current methods usually collect and pool all data from one domain to train a single policy to handle such heterogeneity in tasks and domains, which is prohibitively expensive and difficult. In this work, we present a flexible approach, dubbed Policy Composition, to combine information across such diverse modalities and domains for learning scene-level and task-level generalized manipulation skills, by composing different data distributions represented with diffusion models. Our method can use task-level composition for multi-task manipulation and be composed with analytic cost functions to adapt policy behaviors at inference time. We train our method on simulation, human, and real robot data and evaluate in tool-use tasks. The composed policy achieves robust and dexterous performance under varying scenes and tasks and outperforms baselines from a single data source in both simulation and real-world experiments. See https://liruiw.github.io/policycomp for more details .

cs.RO↗

Transferable Tactile Transformers for Representation Learning Across Diverse Sensors and Tasks

This paper presents T3: Transferable Tactile Transformers, a framework for tactile representation learning that scales across multi-sensors and multi-tasks. T3 is designed to overcome the contemporary issue that camera-based tactile sensing is extremely heterogeneous, i.e. sensors are built into different form factors, and existing datasets were collected for disparate tasks. T3 captures the shared latent information across different sensor-task pairings by constructing a shared trunk transformer with sensor-specific encoders and task-specific decoders. The pre-training of T3 utilizes a novel Foundation Tactile (FoTa) dataset, which is aggregated from several open-sourced datasets and it contains over 3 million data points gathered from 13 sensors and 11 tasks. FoTa is the largest and most diverse dataset in tactile sensing to date and it is made publicly available in a unified format. Across various sensors and tasks, experiments show that T3 pre-trained with FoTa achieved zero-shot transferability in certain sensor-task pairings, can be further fine-tuned with small amounts of domain-specific data, and its performance scales with bigger network sizes. T3 is also effective as a tactile encoder for long horizon contact-rich manipulation. Results from sub-millimeter multi-pin electronics insertion tasks show that T3 achieved a task success rate 25% higher than that of policies trained with tactile encoders trained from scratch, or 53% higher than without tactile sensing. Data, code, and model checkpoints are open-sourced at https://t3.alanz.info

cs.RO↗

RainbowSight: A Family of Generalizable, Curved, Camera-Based Tactile Sensors For Shape Reconstruction

Camera-based tactile sensors can provide high resolution positional and local geometry information for robotic manipulation. Curved and rounded fingers are often advantageous, but it can be difficult to derive illumination systems that work well within curved geometries. To address this issue, we introduce RainbowSight, a family of curved, compact, camera-based tactile sensors which use addressable RGB LEDs illuminated in a novel rainbow spectrum pattern. In addition to being able to scale the illumination scheme to different sensor sizes and shapes to fit on a variety of end effector configurations, the sensors can be easily manufactured and require minimal optical tuning to obtain high resolution depth reconstructions of an object deforming the sensor's soft elastomer surface. Additionally, we show the advantages of our new hardware design and improvements in calibration methods for accurate depth map generation when compared to alternative lighting methods commonly implemented in previous camera-based tactile sensors. With these advancements, we make the integration of tactile sensors more accessible to roboticists by allowing them the flexibility to easily customize, fabricate, and calibrate camera-based tactile sensors to best fit the needs of their robotic systems.

cs.RO↗

A Passively Bendable, Compliant Tactile Palm with RObotic Modular Endoskeleton Optical (ROMEO) Fingers

Many robotic hands currently rely on extremely dexterous robotic fingers and a thumb joint to envelop themselves around an object. Few hands focus on the palm even though human hands greatly benefit from their central fold and soft surface. As such, we develop a novel structurally compliant soft palm, which enables more surface area contact for the objects that are pressed into it. Moreover, this design, along with the development of a new low-cost, flexible illumination system, is able to incorporate a high-resolution tactile sensing system inspired by the GelSight sensors. Concurrently, we design RObotic Modular Endoskeleton Optical (ROMEO) fingers, which are underactuated two-segment soft fingers that are able to house the new illumination system, and we integrate them into these various palm configurations. The resulting robotic hand is slightly bigger than a baseball and represents one of the first soft robotic hands with actuated fingers and a passively compliant palm, all of which have high-resolution tactile sensing. This design also potentially helps researchers discover and explore more soft-rigid tactile robotic hand designs with greater capabilities in the future. The supplementary video can be found here: https://youtu.be/RKfIFiewqsg

cs.RO↗

Scalable, Simulation-Guided Compliant Tactile Finger Design

Compliant grippers enable robots to work with humans in unstructured environments. In general, these grippers can improve with tactile sensing to estimate the state of objects around them to precisely manipulate objects. However, co-designing compliant structures with high-resolution tactile sensing is a challenging task. We propose a simulation framework for the end-to-end forward design of GelSight Fin Ray sensors. Our simulation framework consists of mechanical simulation using the finite element method (FEM) and optical simulation including physically based rendering (PBR). To simulate the fluorescent paint used in these GelSight Fin Rays, we propose an efficient method that can be directly integrated in PBR. Using the simulation framework, we investigate design choices available in the compliant grippers, namely gel pad shapes, illumination conditions, Fin Ray gripper sizes, and Fin Ray stiffness. This infrastructure enables faster design and prototype time frames of new Fin Ray sensors that have various sensing areas, ranging from 48 mm $\times$ \18 mm to 70 mm $\times$ 35 mm. Given the parameters we choose, we can thus optimize different Fin Ray designs and show their utility in grasping day-to-day objects.

cs.RO↗

GelSight Svelte: A Human Finger-shaped Single-camera Tactile Robot Finger with Large Sensing Coverage and Proprioceptive Sensing

Camera-based tactile sensing is a low-cost, popular approach to obtain highly detailed contact geometry information. However, most existing camera-based tactile sensors are fingertip sensors, and longer fingers often require extraneous elements to obtain an extended sensing area similar to the full length of a human finger. Moreover, existing methods to estimate proprioceptive information such as total forces and torques applied on the finger from camera-based tactile sensors are not effective when the contact geometry is complex. We introduce GelSight Svelte, a curved, human finger-sized, single-camera tactile sensor that is capable of both tactile and proprioceptive sensing over a large area. GelSight Svelte uses curved mirrors to achieve the desired shape and sensing coverage. Proprioceptive information, such as the total bending and twisting torques applied on the finger, is reflected as deformations on the flexible backbone of GelSight Svelte, which are also captured by the camera. We train a convolutional neural network to estimate the bending and twisting torques from the captured images. We conduct gel deformation experiments at various locations of the finger to evaluate the tactile sensing capability and proprioceptive sensing accuracy. To demonstrate the capability and potential uses of GelSight Svelte, we conduct an object holding task with three different grasping modes that utilize different areas of the finger. More information is available on our website: https://gelsight-svelte.alanz.info

cs.RO↗

GelSight Svelte Hand: A Three-finger, Two-DoF, Tactile-rich, Low-cost Robot Hand for Dexterous Manipulation

This paper presents GelSight Svelte Hand, a novel 3-finger 2-DoF tactile robotic hand that is capable of performing precision grasps, power grasps, and intermediate grasps. Rich tactile signals are obtained from one camera on each finger, with an extended sensing area similar to the full length of a human finger. Each finger of GelSight Svelte Hand is supported by a semi-rigid endoskeleton and covered with soft silicone materials, which provide both rigidity and compliance. We describe the design, fabrication, functionalities, and tactile sensing capability of GelSight Svelte Hand in this paper. More information is available on our website: \url{https://gelsight-svelte.alanz.info}.

cs.RO↗

GelSight360: An Omnidirectional Camera-Based Tactile Sensor for Dexterous Robotic Manipulation

Camera-based tactile sensors have shown great promise in enhancing a robot's ability to perform a variety of dexterous manipulation tasks. Advantages of their use can be attributed to the high resolution tactile data and 3D depth map reconstructions they can provide. Unfortunately, many of these tactile sensors use either a flat sensing surface, sense on only one side of the sensor's body, or have a bulky form-factor, making it difficult to integrate the sensors with a variety of robotic grippers. Of the camera-based sensors that do have all-around, curved sensing surfaces, many cannot provide 3D depth maps; those that do often require optical designs specified to a particular sensor geometry. In this work, we introduce GelSight360, a fingertip-like, omnidirectional, camera-based tactile sensor capable of producing depth maps of objects deforming the sensor's surface. In addition, we introduce a novel cross-LED lighting scheme that can be implemented in different all-around sensor geometries and sizes, allowing the sensor to easily be reconfigured and attached to different grippers of varying DOFs. With this work, we enable roboticists to quickly and easily customize high resolution tactile sensors to fit their robotic system's needs.

cs.RO↗