Search

In modern robotics, addressing the lack of accurate state space information in real-world scenarios has led to a significant focus on utilizing visuomotor observation to provide safety assurances. Although supervised learning methods, such as imitation learning, have demonstrated potential in synthesizing control policies based on visuomotor observations, they require ground truth safety labels for the complete dataset and do not provide formal safety assurances. On the other hand, traditional control-theoretic methods like Control Barrier Functions (CBFs) and Hamilton-Jacobi (HJ) Reachability provide formal safety guarantees but depend on accurate knowledge of system dynamics, which is often unavailable for high-dimensional visuomotor data. To overcome these limitations, we propose a novel approach to synthesize a semi-supervised safe visuomotor policy using barrier certificates that integrate the strengths of model-free supervised learning and model-based control methods. This framework synthesizes a provably safe controller without requiring safety labels for the complete dataset and ensures completeness guarantees for both the barrier certificate and the policy. We validate our approach through distinct case studies: an inverted pendulum system and the obstacle avoidance of an autonomous mobile robot., Comment: First two authors have contributed equally. 8 Pages, 3 figures

Recent advances in the fields of natural language processing and computer vision have shown great potential in understanding the underlying dynamics of the world from large-scale internet data. However, translating this knowledge into robotic systems remains an open challenge, given the scarcity of human-robot interactions and the lack of large-scale datasets of real-world robotic data. Previous robot learning approaches such as behavior cloning and reinforcement learning have shown great capabilities in learning robotic skills from human demonstrations or from scratch in specific environments. However, these approaches often require task-specific demonstrations or designing complex simulation environments, which limits the development of generalizable and robust policies for new settings. Aiming to address these limitations, we propose an agent-based framework for grounding robot policies to the current context, considering the constraints of a current robot and its environment using visuomotor-grounded language guidance. The proposed framework is composed of a set of conversational agents designed for specific roles -- namely, high-level advisor, visual grounding, monitoring, and robotic agents. Given a base policy, the agents collectively generate guidance at run time to shift the action distribution of the base policy towards more desirable future states. We demonstrate that our approach can effectively guide manipulation policies to achieve significantly higher success rates both in simulation and in real-world experiments without the need for additional human demonstrations or extensive exploration. Project videos at https://sites.google.com/view/motorcortex/home., Comment: 19 pages, 6 figures, 1 table

In contrast to single-skill tasks, long-horizon tasks play a crucial role in our daily life, e.g., a pouring task requires a proper concatenation of reaching, grasping and pouring subtasks. As an efficient solution for transferring human skills to robots, imitation learning has achieved great progress over the last two decades. However, when learning long-horizon visuomotor skills, imitation learning often demands a large amount of semantically segmented demonstrations. Moreover, the performance of imitation learning could be susceptible to external perturbation and visual occlusion. In this paper, we exploit dynamical movement primitives and meta-learning to provide a new framework for imitation learning, called Meta-Imitation Learning with Adaptive Dynamical Primitives (MiLa). MiLa allows for learning unsegmented long-horizon demonstrations and adapting to unseen tasks with a single demonstration. MiLa can also resist external disturbances and visual occlusion during task execution. Real-world robotic experiments demonstrate the superiority of MiLa, irrespective of visual occlusion and random perturbations on robots., Comment: 15 pages, 6 figures

An end-to-end (E2E) visuomotor policy is typically treated as a unified whole, but recent approaches using out-of-domain (OOD) data to pretrain the visual encoder have cleanly separated the visual encoder from the network, with the remainder referred to as the policy. We propose Visual Alignment Testing, an experimental framework designed to evaluate the validity of this functional separation. Our results indicate that in E2E-trained models, visual encoders actively contribute to decision-making resulting from motor data supervision, contradicting the assumed functional separation. In contrast, OOD-pretrained models, where encoders lack this capability, experience an average performance drop of 42% in our benchmark results, compared to the state-of-the-art performance achieved by E2E policies. We believe this initial exploration of visual encoders' role can provide a first step towards guiding future pretraining methods to address their decision-making ability, such as developing task-conditioned or context-aware encoders.

Suppose you are at your desk looking at some objects on it. You don't know the precise distance from your eye to any particular object in meters. However, you can immediately reach out and touch any of them. Instead of the meter, your knowledge of distance is encoded in unknown but embodied units of action. In contrast, standard approaches in robotics assume calibration to the meter, so that separated vision and control processes can be interfaced. Consequently, robots are precisely manufactured and calibrated, resulting in expensive systems available in only a few configurations. In response, we propose Embodied Visuomotor Representation, a framework that allows distance to be measured by a robot's own actions and thus minimizes dependence on calibrated 3D sensors and physical models. Using it, we demonstrate that a robot without knowledge of its size, environmental scale, or its own strength can become capable of touching and clearing obstacles after several seconds of operation. Similarly, we demonstrate in simulation that an agent, without knowledge of its mass or strength, can jump a gap of unknown size after performing a few test oscillations. These experiments parallel bee and gerbil behavior, respectively., Comment: 47 pages, 10 figures, 1 table, under review

Despite significant progress in robotics and embodied AI in recent years, deploying robots for long-horizon tasks remains a great challenge. Majority of prior arts adhere to an open-loop philosophy and lack real-time feedback, leading to error accumulation and undesirable robustness. A handful of approaches have endeavored to establish feedback mechanisms leveraging pixel-level differences or pre-trained visual representations, yet their efficacy and adaptability have been found to be constrained. Inspired by classic closed-loop control systems, we propose CLOVER, a closed-loop visuomotor control framework that incorporates feedback mechanisms to improve adaptive robotic control. CLOVER consists of a text-conditioned video diffusion model for generating visual plans as reference inputs, a measurable embedding space for accurate error quantification, and a feedback-driven controller that refines actions from feedback and initiates replans as needed. Our framework exhibits notable advancement in real-world robotic tasks and achieves state-of-the-art on CALVIN benchmark, improving by 8% over previous open-loop counterparts. Code and checkpoints are maintained at https://github.com/OpenDriveLab/CLOVER., Comment: Accepted at NeurIPS 2024. Code and models: https://github.com/OpenDriveLab/CLOVER

Vision-based imitation learning has shown promising capabilities of endowing robots with various motion skills given visual observation. However, current visuomotor policies fail to adapt to drastic changes in their visual observations. We present Perception Stitching that enables strong zero-shot adaptation to large visual changes by directly stitching novel combinations of visual encoders. Our key idea is to enforce modularity of visual encoders by aligning the latent visual features among different visuomotor policies. Our method disentangles the perceptual knowledge with the downstream motion skills and allows the reuse of the visual encoders by directly stitching them to a policy network trained with partially different visual conditions. We evaluate our method in various simulated and real-world manipulation tasks. While baseline methods failed at all attempts, our method could achieve zero-shot success in real-world visuomotor tasks. Our quantitative and qualitative analysis of the learned features of the policy network provides more insights into the high performance of our proposed method.

Many robotic systems, such as mobile manipulators or quadrotors, cannot be equipped with high-end GPUs due to space, weight, and power constraints. These constraints prevent these systems from leveraging recent developments in visuomotor policy architectures that require high-end GPUs to achieve fast policy inference. In this paper, we propose Consistency Policy, a faster and similarly powerful alternative to Diffusion Policy for learning visuomotor robot control. By virtue of its fast inference speed, Consistency Policy can enable low latency decision making in resource-constrained robotic setups. A Consistency Policy is distilled from a pretrained Diffusion Policy by enforcing self-consistency along the Diffusion Policy's learned trajectories. We compare Consistency Policy with Diffusion Policy and other related speed-up methods across 6 simulation tasks as well as three real-world tasks where we demonstrate inference on a laptop GPU. For all these tasks, Consistency Policy speeds up inference by an order of magnitude compared to the fastest alternative method and maintains competitive success rates. We also show that the Conistency Policy training procedure is robust to the pretrained Diffusion Policy's quality, a useful result that helps practioners avoid extensive testing of the pretrained model. Key design decisions that enabled this performance are the choice of consistency objective, reduced initial sample variance, and the choice of preset chaining steps., Comment: https://consistency-policy.github.io/

A key challenge in manipulation is learning a policy that can robustly generalize to diverse visual environments. A promising mechanism for learning robust policies is to leverage video generative models, which are pretrained on large-scale datasets of internet videos. In this paper, we propose a visuomotor policy learning framework that fine-tunes a video diffusion model on human demonstrations of a given task. At test time, we generate an example of an execution of the task conditioned on images of a novel scene, and use this synthesized execution directly to control the robot. Our key insight is that using common tools allows us to effortlessly bridge the embodiment gap between the human hand and the robot manipulator. We evaluate our approach on four tasks of increasing complexity and demonstrate that harnessing internet-scale generative models allows the learned policy to achieve a significantly higher degree of generalization than existing behavior cloning approaches., Comment: Project page: https://dreamitate.cs.columbia.edu/

Despite the significant advancements in Deep Reinforcement Learning (RL) observed in the last decade, the amount of training experience necessary to learn effective policies remains one of the primary concerns both in simulated and real environments. Looking to solve this issue, previous work has shown that improved training efficiency can be achieved by separately modeling agent and environment, but usually requiring a supervisory agent mask. In contrast to RL, humans can perfect a new skill from a very small number of trials and in most cases do so without a supervisory signal, making neuroscientific studies of human development a valuable source of inspiration for RL. In particular, we explore the idea of motor prediction, which states that humans develop an internal model of themselves and of the consequences that their motor commands have on the immediate sensory inputs. Our insight is that the movement of the agent provides a cue that allows the duality between agent and environment to be learned. To instantiate this idea, we present Ego-Foresight, a self-supervised method for disentangling agent and environment based on motion and prediction. Our main finding is that visuomotor prediction of the agent provides regularization to the RL algorithm, by encouraging the actions to stay within predictable bounds. To test our approach, we first study the ability of our model to visually predict agent movement irrespective of the environment, in real-world robotic interactions. Then, we integrate Ego-Foresight with a model-free RL algorithm to solve simulated robotic manipulation tasks, showing an average improvement of 23% in efficiency and 8% in performance., Comment: 9 pages, 9 figures, 3 tables, conference

Modeling a generalized visuomotor policy has been a longstanding challenge for both computer vision and robotics communities. Existing approaches often fail to efficiently leverage cross-dataset resources or rely on heavy Vision-Language models, which require substantial computational resources, thereby limiting their multi-task performance and application potential. In this paper, we introduce a novel paradigm that effectively utilizes latent modeling of manipulation skills and an efficient visuomotor latent diffusion policy, which enhances the utilizing of existing cross-embodiment and cross-environment datasets, thereby improving multi-task capabilities. Our methodology consists of two decoupled phases: action modeling and policy modeling. Firstly, we introduce a task-agnostic, embodiment-aware trajectory latent autoencoder for unified action skills modeling. This step condenses action data and observation into a condensed latent space, effectively benefiting from large-scale cross-datasets. Secondly, we propose to use a visuomotor latent diffusion policy that recovers target skill latent from noises for effective task execution. We conducted extensive experiments on two widely used benchmarks, and the results demonstrate the effectiveness of our proposed paradigms on multi-tasking and pre-training. Code is available at https://github.com/AlbertTan404/RoLD.

We investigated the human capacity to acquire multiple visuomotor mappings for de novo skills. Using a grid navigation paradigm, we tested whether contextual cues implemented as different "grid worlds", allow participants to learn two distinct key-mappings more efficiently. Our results indicate that when contextual information is provided, task performance is significantly better. The same held true for meta-reinforcement learning agents that differed in whether or not they receive contextual information when performing the task. We evaluated their accuracy in predicting human performance in the task and analyzed their internal representations. The results indicate that contextual cues allow the formation of separate representations in space and time when using different visuomotor mappings, whereas the absence of them favors sharing one representation. While both strategies can allow learning of multiple visuomotor mappings, we showed contextual cues provide a computational advantage in terms of how many mappings can be learned.

Imitation learning provides an efficient way to teach robots dexterous skills; however, learning complex skills robustly and generalizablely usually consumes large amounts of human demonstrations. To tackle this challenging problem, we present 3D Diffusion Policy (DP3), a novel visual imitation learning approach that incorporates the power of 3D visual representations into diffusion policies, a class of conditional action generative models. The core design of DP3 is the utilization of a compact 3D visual representation, extracted from sparse point clouds with an efficient point encoder. In our experiments involving 72 simulation tasks, DP3 successfully handles most tasks with just 10 demonstrations and surpasses baselines with a 24.2% relative improvement. In 4 real robot tasks, DP3 demonstrates precise control with a high success rate of 85%, given only 40 demonstrations of each task, and shows excellent generalization abilities in diverse aspects, including space, viewpoint, appearance, and instance. Interestingly, in real robot experiments, DP3 rarely violates safety requirements, in contrast to baseline methods which frequently do, necessitating human intervention. Our extensive evaluation highlights the critical importance of 3D representations in real-world robot learning. Videos, code, and data are available on https://3d-diffusion-policy.github.io ., Comment: Published at Robotics: Science and Systems (RSS) 2024. Videos, code, and data: https://3d-diffusion-policy.github.io

This paper introduces LeTO, a method for learning constrained visuomotor policy via differentiable trajectory optimization. Our approach uniquely integrates a differentiable optimization layer into the neural network. By formulating the optimization layer as a trajectory optimization problem, we enable the model to end-to-end generate actions in a safe and controlled fashion without extra modules. Our method allows for the introduction of constraints information during the training process, thereby balancing the training objectives of satisfying constraints, smoothing the trajectories, and minimizing errors with demonstrations. This "gray box" method marries the optimization-based safety and interpretability with the powerful representational abilities of neural networks. We quantitatively evaluate LeTO in simulation and on the real robot. In simulation, LeTO achieves a success rate comparable to state-of-the-art imitation learning methods, but the generated trajectories are of less uncertainty, higher quality, and smoother. In real-world experiments, we deployed LeTO to handle constraints-critical tasks. The results show the effectiveness of LeTO comparing with state-of-the-art imitation learning approaches. We release our code at https://github.com/ZhengtongXu/LeTO., Comment: 8 pages, 5 figures

Imitation learning (IL) can train computationally-efficient sensorimotor policies from a resource-intensive Model Predictive Controller (MPC), but it often requires many samples, leading to long training times or limited robustness. To address these issues, we combine IL with a variant of robust MPC that accounts for process and sensing uncertainties, and we design a data augmentation (DA) strategy that enables efficient learning of vision-based policies. The proposed DA method, named Tube-NeRF, leverages Neural Radiance Fields (NeRFs) to generate novel synthetic images, and uses properties of the robust MPC (the tube) to select relevant views and to efficiently compute the corresponding actions. We tailor our approach to the task of localization and trajectory tracking on a multirotor, by learning a visuomotor policy that generates control actions using images from the onboard camera as only source of horizontal position. Numerical evaluations show 80-fold increase in demonstration efficiency and a 50% reduction in training time over current IL methods. Additionally, our policies successfully transfer to a real multirotor, achieving low tracking errors despite large disturbances, with an onboard inference time of only 1.5 ms. Video: https://youtu.be/_W5z33ZK1m4, Comment: Video: https://youtu.be/_W5z33ZK1m4. Evolved paper from our previous work: arXiv:2210.10127

This paper introduces a new approach for quickly adapting a multi-view visuomotor system for robots to varying camera configurations from the baseline setup. It utilises meta-learning to fine-tune the perceptual network while keeping the policy network fixed. Experimental results demonstrate a significant reduction in the number of new training episodes needed to attain baseline performance., Comment: Change of authors since further experiments based on second and third authors comments will be added to a future version of this paper

If a robot masters folding a kitchen towel, we would expect it to master folding a large beach towel. However, existing policy learning methods that rely on data augmentation still don't guarantee such generalization. Our insight is to add equivariance to both the visual object representation and policy architecture. We propose EquivAct which utilizes SIM(3)-equivariant network structures that guarantee generalization across all possible object translations, 3D rotations, and scales by construction. EquivAct is trained in two phases. We first pre-train a SIM(3)-equivariant visual representation on simulated scene point clouds. Then, we learn a SIM(3)-equivariant visuomotor policy using a small amount of source task demonstrations. We show that the learned policy directly transfers to objects that substantially differ from demonstrations in scale, position, and orientation. We evaluate our method in three manipulation tasks involving deformable and articulated objects, going beyond typical rigid object manipulation tasks considered in prior work. We conduct experiments both in simulation and in reality. For real robot experiments, our method uses 20 human demonstrations of a tabletop task and transfers zero-shot to a mobile manipulation task in a much larger setup. Experiments confirm that our contrastive pre-training procedure and equivariant architecture offer significant improvements over prior work. Project website: https://equivact.github.io, Comment: ICRA 2024; The first two authors contributed equally

Large increases in visuomotor feedback gains occur during initial adaptation to novel dynamics, which we propose are due to increased internal model uncertainty. That is, large errors indicate increased uncertainty in our prediction of the environment, increasing feedback gains and co-contraction as a coping mechanism. Our previous work showed distinct patterns of visuomotor feedback gains during abrupt or gradual adaptation to a force field, suggesting two complementary processes: reactive feedback gains increasing with internal model uncertainty and the gradual learning of predictive feedback gains tuned to the environment. Here we further investigate what drives these changes visuomotor feedback gains in learning, by separating the effects of internal model uncertainty from visual error signal through removal of visual error information. Removing visual error information suppresses the visuomotor feedback gains in all conditions, but the pattern of modulation throughout adaptation is unaffected. Moreover, we find increased muscle co-contraction in both abrupt and gradual adaptation protocols, demonstrating that visuomotor feedback responses are independent from the level of co-contraction. Our result suggests that visual feedback benefits motor adaptation tasks through higher visuomotor feedback gains, but when it is not available participants adapt at a similar rate through increased co-contraction. We have demonstrated a direct connection between learning and predictive visuomotor feedback gains, independent from visual error signals. This further supports our hypothesis that internal model uncertainty drives initial increases in feedback gains., Comment: 29 pages, 11 figures. arXiv admin note: substantial text overlap with arXiv:2008.07574