arXiv subjects

Jeremy A. Collins

Publications and source records attributed to Jeremy A. Collins.

9 recordsLinked to original sources

arXiv2026

Beyond Imitation: Self-Improving Robot Policies via Off-Policy Q-Planning

Behaviour Cloning (BC) has driven remarkable progress in robot manipulation, yet it is fundamentally limited by its inability to self-improve: a policy that fails cannot learn from that failure without additional human demonstrations. Reinforcement Learning fine-tuning offers a path to self-improvement but has proven difficult to scale to the multi-billion-parameter models underpinning modern robot policies. We propose Q-Planning, which equips a large visuomotor BC policy with a small off-policy Q-function. Because a Q-function estimates value rather than imitates actions, it can be trained on the same successful demonstrations as the BC policy and later absorb both successful and failed deployment rollouts, an asymmetry BC does not have. We exploit this asymmetry to enable value-guided action selection at inference (a single-step Q-weighted average over BC draws) and online self-improvement that fine-tunes only the Q-function, leaving the BC weights untouched. On LIBERO and bimanual RoboTwin, ten iterations of self-improvement lift every benchmark score we tested (LIBERO-10 93% to 99%, RoboTwin 83.8% to 91.4%) and shorten successful episodes on the near-ceiling suites (LIBERO-Object, LIBERO-Goal). On two contact-rich bimanual real-robot tasks, the same loop (BC frozen, no human intervention) improves purely from its own deployment rollouts: stack-cups 40% to 90% and insert-wallet 25% to 80% in five iterations, whereas SFT on successful rollouts alone stalls at 55% and 30%. Under an identical online budget Q-Planning is the only method, among Best-of-N, filtered SFT, IBRL, DSRL, and DAWR, that improves stably from failures without training an auxiliary actor.

cs.RO↗

arXiv2026

COBALT: Crowdsourcing Robot Learning via Cloud-Based Teleoperation with Smartphones

The scarcity of large-scale, high-quality demonstration data remains a bottleneck in scaling imitation learning for robotic manipulation. We present COBALT, a teleoperation platform designed to democratize robot learning at scale both in simulation and in the real world. By leveraging vectorized environments, our scalable, load-balanced infrastructure supports concurrent teleoperation by multiple users on a single GPU, yielding a significant reduction in teleoperation cost. Operators can connect from nearly anywhere on Earth using commonly available devices, including single or dual smartphones, VR headsets, 3D mice, and keyboards. An inmemory data cache and efficient video streaming keep control and rendering synchronous, sustaining dozens of concurrent users at 20 Hz with sub-100 ms end-to-end latency for up to 8 concurrent users per GPU. We also demonstrate stable operation supporting 256 simulated clients across 8 GPUs, underscoring the system's ability to scale across hardware and within individual servers. We perform a comprehensive user study showing that phone-based teleoperation performs comparably to or better than specialized hardware, enabling faster, more ergonomic data collection. To ensure data quality, COBALT logs a suite of real-time metrics to automatically filter suboptimal demonstrations. We further demonstrate that a structured user training curriculum significantly improves data collection quality. Guided by insights from our user study, we crowdsource the collection of a large-scale, high-quality pilot dataset with 7500+ demonstrations (50+ hours) collected with smartphones across nine countries over five days. We validate the dataset's quality by training state-of-the-art imitation learning algorithms. Please visit https://cobalt-teleop.github.io/ for more details.

cs.RO↗

arXiv2025

AMPLIFY: Actionless Motion Priors for Robot Learning from Videos

Action-labeled data for robotics is scarce and expensive, limiting the generalization of learned policies. In contrast, vast amounts of action-free video data are readily available, but translating these observations into effective policies remains a challenge. We introduce AMPLIFY, a novel framework that leverages large-scale video data by encoding visual dynamics into compact, discrete motion tokens derived from keypoint trajectories. Our modular approach separates visual motion prediction from action inference, decoupling the challenges of learning what motion defines a task from how robots can perform it. We train a forward dynamics model on abundant action-free videos and an inverse dynamics model on a limited set of action-labeled examples, allowing for independent scaling. Extensive evaluations demonstrate that the learned dynamics are both accurate, achieving up to 3.7x better MSE and over 2.5x better pixel prediction accuracy compared to prior approaches, and broadly useful. In downstream policy learning, our dynamics predictions enable a 1.2-2.2x improvement in low-data regimes, a 1.4x average improvement by learning from action-free human videos, and the first generalization to LIBERO tasks from zero in-distribution action data. Beyond robotic control, we find the dynamics learned by AMPLIFY to be a versatile latent world model, enhancing video prediction quality. Our results present a novel paradigm leveraging heterogeneous data sources to build efficient, generalizable world models. More information can be found at https://amplify-robotics.github.io/.

cs.RO↗

arXiv2025

RoCoDA: Counterfactual Data Augmentation for Data-Efficient Robot Learning from Demonstrations

Imitation learning in robotics faces significant challenges in generalization due to the complexity of robotic environments and the high cost of data collection. We introduce RoCoDA, a novel method that unifies the concepts of invariance, equivariance, and causality within a single framework to enhance data augmentation for imitation learning. RoCoDA leverages causal invariance by modifying task-irrelevant subsets of the environment state without affecting the policy's output. Simultaneously, we exploit SE(3) equivariance by applying rigid body transformations to object poses and adjusting corresponding actions to generate synthetic demonstrations. We validate RoCoDA through extensive experiments on five robotic manipulation tasks, demonstrating improvements in policy performance, generalization, and sample efficiency compared to state-of-the-art data augmentation methods. Our policies exhibit robust generalization to unseen object poses, textures, and the presence of distractors. Furthermore, we observe emergent behavior such as re-grasping, indicating policies trained with RoCoDA possess a deeper understanding of task dynamics. By leveraging invariance, equivariance, and causality, RoCoDA provides a principled approach to data augmentation in imitation learning, bridging the gap between geometric symmetries and causal reasoning. Project Page: https://rocoda.github.io

cs.RO↗

arXiv2024

PressureVision++: Estimating Fingertip Pressure from Diverse RGB Images

Touch plays a fundamental role in manipulation for humans; however, machine perception of contact and pressure typically requires invasive sensors. Recent research has shown that deep models can estimate hand pressure based on a single RGB image. However, evaluations have been limited to controlled settings since collecting diverse data with ground-truth pressure measurements is difficult. We present a novel approach that enables diverse data to be captured with only an RGB camera and a cooperative participant. Our key insight is that people can be prompted to apply pressure in a certain way, and this prompt can serve as a weak label to supervise models to perform well under varied conditions. We collect a novel dataset with 51 participants making fingertip contact with diverse objects. Our network, PressureVision++, outperforms human annotators and prior work. We also demonstrate an application of PressureVision++ to mixed reality where pressure estimation allows everyday surfaces to be used as arbitrary touch-sensitive interfaces. Code, data, and models are available online.

cs.CV↗

arXiv2023

Visual Contact Pressure Estimation for Grippers in the Wild

Sensing contact pressure applied by a gripper can benefit autonomous and teleoperated robotic manipulation, but adding tactile sensors to a gripper's surface can be difficult or impractical. If a gripper visibly deforms, contact pressure can be visually estimated using images from an external camera that observes the gripper. While researchers have demonstrated this capability in controlled laboratory settings, prior work has not addressed challenges associated with visual pressure estimation in the wild, where lighting, surfaces, and other factors vary widely. We present a model and associated methods that enable visual pressure estimation under widely varying conditions. Our model, Visual Pressure Estimation for Robots (ViPER), takes an image from an eye-in-hand camera as input and outputs an image representing the pressure applied by a soft gripper. Our key insight is that force/torque sensing can be used as a weak label to efficiently collect training data in settings where pressure measurements would be difficult to obtain. When trained on this weakly labeled data combined with fully labeled data that includes pressure measurements, ViPER outperforms prior methods, enables precision manipulation in cluttered settings, and provides accurate estimates for unseen conditions relevant to in-home use.

cs.RO↗

arXiv2023

ForceSight: Text-Guided Mobile Manipulation with Visual-Force Goals

We present ForceSight, a system for text-guided mobile manipulation that predicts visual-force goals using a deep neural network. Given a single RGBD image combined with a text prompt, ForceSight determines a target end-effector pose in the camera frame (kinematic goal) and the associated forces (force goal). Together, these two components form a visual-force goal. Prior work has demonstrated that deep models outputting human-interpretable kinematic goals can enable dexterous manipulation by real robots. Forces are critical to manipulation, yet have typically been relegated to lower-level execution in these systems. When deployed on a mobile manipulator equipped with an eye-in-hand RGBD camera, ForceSight performed tasks such as precision grasps, drawer opening, and object handovers with an 81% success rate in unseen environments with object instances that differed significantly from the training data. In a separate experiment, relying exclusively on visual servoing and ignoring force goals dropped the success rate from 90% to 45%, demonstrating that force goals can significantly enhance performance. The appendix, videos, code, and trained models are available at https://force-sight.github.io/.

cs.RO↗

arXiv2023

Force/Torque Sensing for Soft Grippers using an External Camera

Robotic manipulation can benefit from wrist-mounted force/torque (F/T) sensors, but conventional F/T sensors can be expensive, difficult to install, and damaged by high loads. We present Visual Force/Torque Sensing (VFTS), a method that visually estimates the 6-axis F/T measurement that would be reported by a conventional F/T sensor. In contrast to approaches that sense loads using internal cameras placed behind soft exterior surfaces, our approach uses an external camera with a fisheye lens that observes a soft gripper. VFTS includes a deep learning model that takes a single RGB image as input and outputs a 6-axis F/T estimate. We trained the model with sensor data collected while teleoperating a robot (Stretch RE1 from Hello Robot Inc.) to perform manipulation tasks. VFTS outperformed F/T estimates based on motor currents, generalized to a novel home environment, and supported three autonomous tasks relevant to healthcare: grasping a blanket, pulling a blanket over a manikin, and cleaning a manikin's limbs. VFTS also performed well with a manually operated pneumatic gripper. Overall, our results suggest that an external camera observing a soft gripper can perform useful visual force/torque sensing for a variety of manipulation tasks.

cs.RO↗

arXiv2022

Visual Pressure Estimation and Control for Soft Robotic Grippers

Soft robotic grippers facilitate contact-rich manipulation, including robust grasping of varied objects. Yet the beneficial compliance of a soft gripper also results in significant deformation that can make precision manipulation challenging. We present visual pressure estimation & control (VPEC), a method that infers pressure applied by a soft gripper using an RGB image from an external camera. We provide results for visual pressure inference when a pneumatic gripper and a tendon-actuated gripper make contact with a flat surface. We also show that VPEC enables precision manipulation via closed-loop control of inferred pressure images. In our evaluation, a mobile manipulator (Stretch RE1 from Hello Robot) uses visual servoing to make contact at a desired pressure; follow a spatial pressure trajectory; and grasp small low-profile objects, including a microSD card, a penny, and a pill. Overall, our results show that visual estimates of applied pressure can enable a soft gripper to perform precision manipulation.

cs.RO↗