RSS 2024 Papers

Generating safe motion plans in real-time is necessary for the wide-scale deployment of robots in unstructured and human-centric environments. These motion plans must be safe to ensure humans are not harmed and nearby objects are not damaged. However, they must also be generated in real-time to ensure the robot can quickly adapt to changes in the environment. Many trajectory optimization methods introduce heuristics that trade-off safety and real-time performance, which can lead to potentially unsafe plans. This paper addresses this challenge by proposing Safe Planning for Articulated Robots Using Reachability-based Obstacle Avoidance With Spheres (SPARROWS). SPARROWS is a receding-horizon trajectory planner that utilizes the combination of a novel reachable set representation and an exact signed distance function to generate provably-safe motion plans. At runtime, SPARROWS uses parameterized trajectories to compute reachable sets composed entirely of spheres that overapproximate the swept volume of the robot's motion. SPARROWS then performs trajectory optimization to select a safe trajectory that is guaranteed to be collision-free. We demonstrate that SPARROWS' novel reachable set is significantly less conservative than previous approaches. We also demonstrate that SPARROWS outperforms a variety of state-of-the-art methods in solving challenging motion planning tasks in cluttered environments. Code will be released upon acceptance of this manuscript.

Operating robots precisely and at high speeds has been a long-standing goal of robotics research. Balancing these competing demands is key to enabling the seamless collaboration of robots and humans and increasing task performance. However, traditional motor-driven systems often fall short in this balancing act. Due to their rigid and often heavy design exacerbated by positioning the motors into the joints, faster motions of such robots transfer high forces at impact. To enable precise and safe dynamic motions, we introduce a four degree-of-freedom tendon-driven robot arm. Tendons allow placing the actuation at the base to reduce the robot's inertia, which we show significantly reduces peak collision forces compared to conventional motor-driven systems. Pairing our robot with pneumatic muscles allows generating high forces and highly accelerated motions, while benefiting from impact resilience through passive compliance. Since tendons are subject to additional friction and hence prone to tear, we validate the reliability of our robotic arm on various experiments, including long-term dynamic motions. We also demonstrate its ease of control by quantifying the nonlinearities of the system and the performance on a challenging dynamic table tennis task learned from scratch using reinforcement learning. We open-source the entire hardware design, which can be largely 3D printed, the control software, and a proprioceptive dataset of 25 days of diverse robot motions.

This paper addresses the problem of object-goal navigation in autonomous inspections in real-world environments. Object-goal navigation is crucial to enable effective inspections in various settings, often requiring the robot to identify the target object within a large search space. Current object inspection methods fall short of human efficiency because they typically cannot bootstrap prior and common sense knowledge as humans do. In this paper, we introduce a framework that enables robots to use semantic knowledge from prior spatial configurations of the environment and semantic common sense knowledge. We propose SEEK (Semantic Reasoning for Object Inspection Tasks) that combines semantic prior knowledge with the robot's observations to search for and navigate toward target objects more efficiently. SEEK maintains two representations: a Dynamic Scene Graph (DSG) and a Relational Semantic Network (RSN). The RSN is a compact and practical model that estimates the probability of finding the target object across spatial elements in the DSG. We propose a novel probabilistic planning framework to search for the object using relational semantic knowledge. Our simulation analyses demonstrate that SEEK outperforms the classical planning and Large Language Models (LLMs)-based methods that are examined in this study in terms of efficiency for object-goal inspection tasks. We validated our approach on a physical legged robot in urban environments, showcasing its practicality and effectiveness in real-world inspection scenarios.

To interact with daily-life articulated objects of diverse structures and functionalities, understanding the object parts plays a central role in both user instruction comprehension and task execution. However, the possible discordance between the semantic meaning and physics functionalities of the parts poses a challenge for designing a general system. To address this problem, we propose SAGE, a novel framework that bridges semantic and actionable parts of articulated objects to achieve generalizable manipulation under natural language instructions. More concretely, given an articulated object, we first observe all the semantic parts on it, conditioned on which an instruction interpreter proposes possible action programs that concretize the natural language instruction. Then, a part-grounding module maps the semantic parts into so-called Generalizable Actionable Parts (GAParts), which inherently carry information about part motion. End-effector trajectories are predicted on the GAParts, which, together with the action program, form an executable policy. Additionally, an interactive feedback module is incorporated to respond to failures, which closes the loop and increases the robustness of the overall framework. Key to the success of our framework is the joint proposal and knowledge fusion between a large vision-language model (VLM) and a small domain-specific model for both context comprehension and part perception, with the former providing general intuitions and the latter serving as expert facts. Both simulation and real-robot experiments show our effectiveness in handling a large variety of articulated objects with diverse language-instructed goals.

Tactile feedback is critical for understanding the dynamics of both rigid and deformable objects in many manipulation tasks, such as non-prehensile manipulation and dense packing. We introduce an approach that combines visual and tactile sensing for robotic manipulation by learning a neural, tactile-informed dynamics model. Our proposed framework, RoboPack, employs a recurrent graph neural network to estimate object states, including particles and object-level latent physics information, from historical visuo-tactile observations and to perform future state predictions. Our tactile-informed dynamics model, learned from real-world data, can solve downstream robotics tasks with model-predictive control. We demonstrate our approach on a real robot equipped with a compliant Soft-Bubble tactile sensor on non-prehensile manipulation and dense packing tasks, where the robot must infer the physics properties of objects from direct and indirect interactions. Trained on only an average of 30 minutes of real-world interaction data per task, our model can perform online adaptation and make touch-informed predictions. Through extensive evaluations in both long-horizon dynamics prediction and real-world manipulation, our method demonstrates superior effectiveness compared to previous learning-based and physics-based simulation systems.

Recent advancements in Artificial Intelligence (AI) have largely been propelled by scaling. In Robotics, scaling is hindered by the lack of access to massive robot datasets. We advocate using realistic physical simulation as a means to scale environments, tasks, and datasets for robot learning methods. We present RoboCasa, a large-scale simulation framework for training generalist robots in everyday environments. RoboCasa features realistic and diverse scenes focusing on kitchen environments. We provide thousands of 3D assets across over 150 object categories and dozens of interactable furniture and appliances. We enrich the realism and diversity of our simulation with generative AI tools, such as object assets from text-to-3D models and environment textures from text-to-image models. We design a set of 100 tasks for systematic evaluation, including composite tasks generated by the guidance of large language models. To facilitate learning, we provide high-quality human demonstrations and integrate automated trajectory generation methods to substantially enlarge our datasets with minimal human burden. Our experiments show a clear scaling trend in using synthetically generated robot data for large-scale imitation learning and show great promise in harnessing simulation data in real-world tasks. Videos and open-source code are available at https://robocasa.ai/.

Tasks where robots must anticipate human intent, such as navigating around a cluttered home or sorting everyday items, are challenging because they exhibit a wide range of valid actions that lead to similar outcomes. Moreover, zero-shot cooperation between human-robot partners is an especially challenging problem because it requires the robot to infer and adapt on the fly to a latent human intent, which could vary significantly from human to human. Recently, deep learned motion prediction models have shown promising results in predicting human intent but are prone to being confidently incorrect. In this work, we present Risk-Calibrated Interactive Planning (RCIP), which is a framework for measuring and calibrating risk associated with uncertain action selection in human-robot cooperation, with the fundamental idea that the robot should ask for human clarification when the risk associated with the uncertainty in the human's intent cannot be controlled. RCIP builds on the theory of set-valued risk calibration to provide a finite-sample statistical guarantee on the cumulative loss incurred by the robot while minimizing the cost of human clarification in complex multi-step settings. Our main insight is to frame the risk control problem as a sequence-level multi-hypothesis testing problem, allowing efficient calibration using a low-dimensional parameter that controls a pre-trained risk-aware policy. Experiments across a variety of simulated and real-world environments demonstrate RCIP's ability to predict and adapt to a diverse set of dynamic human intents.

Legged locomotion has recently achieved remarkable success with the progress of machine learning techniques, especially deep reinforcement learning (RL). Controllers employing neural networks have demonstrated empirical and qualitative robustness against real-world uncertainties, including sensor noise and external perturbations. However, formally investigating the vulnerabilities of these locomotion controllers remains a challenge. This difficulty arises from the requirement to pinpoint vulnerabilities across a long-tailed distribution within a high-dimensional, temporally sequential space. As a first step towards quantitative verification, we propose a computational method that leverages sequential adversarial attacks to identify weaknesses in learned locomotion controllers. Our research demonstrates that, even state-of-the-art robust controllers can fail significantly under well-designed, low-magnitude adversarial sequence. Through experiments in simulation and on the real robot, we validate our approach's effectiveness, and we illustrate how the results it generates can be used to robustify the original policy and offer valuable insights into the safety of these black-box policies.

In the field of Robot Learning, the complex mapping between high-dimensional observations such as RGB images and low-level robotic actions, two inherently very different spaces, constitutes a complex learning problem, especially with limited amounts of data. In this work, we introduce Render and Diffuse (R&D) a method that unifies low-level robot actions and RGB observations within the image space using virtual renders of the 3D model of the robot. Using this joint observation-action representation it computes low-level robot actions using a learnt diffusion process that iteratively updates the virtual renders of the robot. This space unification simplifies the learning problem and introduces inductive biases that are crucial for sample efficiency and spatial generalisation. We thoroughly evaluate several variants of R&D in simulation and showcase their applicability on six everyday tasks in the real world. Our results show that R&D exhibits strong spatial generalisation capabilities and is more sample efficient than more common image-to-action methods.

Imitation learning methods need significant human supervision to learn policies robust to changes in object poses, physical disturbances, and visual distractors. Reinforcement learning, on the other hand, can explore the environment autonomously to learn robust behaviors but may require impractical amounts of unsafe real-world data collection. To learn performant, robust policies without the burden of unsafe real-world data collection or extensive human supervision, we propose RialTo, a new system for robustifying real-world imitation learning policies via reinforcement learning in digital twin simulation environments constructed on the fly from small amounts of real-world data. To enable this real-to-sim-to-real pipeline, RialTo proposes an easy-to-use interface for quickly scanning and constructing digital twins of real-world environments. We also introduce a novel inverse distillation procedure for bringing real-world demonstrations into simulated environments for efficient fine-tuning, with minimal human intervention and engineering required. We evaluate RialTo across a variety of robotic manipulation problems in the real world, such as robustly stacking dishes on a rack, placing books on a shelf and four other tasks. RialTo increases (over 67%) in policy robustness without requiring extensive human data collection.

Foundation models, e.g., large language models (LLMs), trained on internet-scale data possess zero-shot generalization capabilities that make them a promising technology towards detecting and mitigating out-of-distribution failure modes of robotic systems. Fully realizing this promise, however, poses two challenges: (i) mitigating the considerable computational expense of these models such that they may be applied online, and (ii) incorporating their judgement regarding potential anomalies into a safe control framework. In this work, we present a two-stage reasoning framework: First is a fast binary anomaly classifier that analyzes observations in an LLM embedding space, which may trigger a slower fallback selection stage that utilizes the reasoning capabilities of generative LLMs. These stages correspond to branch points in a model predictive control strategy that maintains the joint feasibility of continuing along various fallback plans to account for the slow reasoner's latency as soon as an anomaly is detected, thus ensuring safety. We show that our fast anomaly classifier outperforms autoregressive reasoning with state-of-the-art GPT models, even when instantiated with relatively small language models. This enables our runtime monitor to improve the trustworthiness of dynamic robotic systems, such as quadrotors or autonomous vehicles, under resource and time constraints. Videos illustrating our approach in both simulation and real-world experiments are available on our project page: https://sites.google.com/view/aesop-llm.

In this work, we study how to build a robotic system that can solve multiple 3D manipulation tasks given language instructions. To be useful in industrial and household domains, such a system should be capable of learning new tasks with few demonstrations and solving them precisely. Prior works, like PerAct and RVT, have studied this problem, however, they often struggle with tasks requiring high precision. We study how to make them more effective, precise, and fast. Using a combination of architectural and system-level improvements, we propose RVT-2, a multitask 3D manipulation model that is 6X faster in training and 2X faster in inference than its predecessor RVT. RVT-2 achieves a new state-of-the-art on RLBench, improving the success rate from 65% to 82%. RVT-2 is also effective in the real world, where it can learn tasks requiring high precision, like picking up and inserting plugs, with just 10 demonstrations. Visual results, code, and trained model are provided at: https://robotic-view-transformer-2.github.io/.

Language provides a way to break down complex concepts into digestible pieces. Recent works in robot imitation learning have proposed learning language-conditioned policies that predict actions given visual observations and the high-level task specified in language. These methods leverage the structure of natural language to share data between semantically similar tasks (e.g., "pick coke can" and "pick an apple") in multi-task datasets. However, as tasks become more semantically diverse (e.g., "pick coke can" and "pour cup"), sharing data between tasks becomes harder and thus learning to map high-level tasks to actions requires substantially more demonstration data. To bridge this divide between tasks and actions, our insight is to teach the robot the language of actions, describing low-level motions with more fine-grained phrases like "move arm forward" or "close gripper". Predicting these *language motions* as an intermediate step between high-level tasks and actions forces the policy to learn the shared structure of low-level motions across seemingly disparate tasks. Furthermore, a policy that is conditioned on language motions can easily be *corrected* during execution through human-specified language motions. This enables a new paradigm for flexible policies that can learn from human intervention in language. Our method RT-H builds an *action hierarchy* using language motions: it first learns to predict language motions, and conditioned on this along with the high-level task, it then predicts actions, using visual context at all stages. Experimentally we show that RT-H leverages this language-action hierarchy to learn policies that are more robust and flexible by effectively tapping into multi-task datasets. We show that these policies not only allow for responding to language interventions, but can also learn from such interventions and outperform methods that learn from teleoperated interventions. Our website and videos are found at https://rt-hierarchy.github.io.

Dynamic fast adaptation is one of the basic capabilities that enables the animals to timely and properly adjust its locomotion reacting to the unpredictable changes. Such capability is also essential for the quadruped robot, when working in the unforseen environment. While reinforcement learning (RL) has achieved a significant progress in locomotion control, rapid adaptation to the model uncertainties remains a challenge. In this paper, we seek to ascertain the control mechanism behind the locomotion RL policy, from which we propose a new RL-based Rapid onLine Adaptive Control (RL2AC) algorithm to complementarily combine the RL policy and the adaptive control together. RL2AC is run at a frequency of 1000Hz without the need for simultaneous training with RL. It presents a strong capability against the external disturbances or the sim-to-real gap, resulting in a robust locomotion, which is achieved through proper torque compensation derived from a novel adaptive controller. Various simulation and experiments have demonstrated the effectiveness of the proposed RL2AC against the heavy load, disturbances acted on one leg, lateral torque, sim-to-real gap and various terrains.

We need to trust robots that use often opaque AI methods. They need to explain themselves to us, and we need to trust their explanation. In this regard, explainability plays a critical role in trustworthy autonomous decision-making to foster transparency and acceptance among end users, especially in complex autonomous driving. Recent advancements in Multi-Modal Large Language models (MLLMs) have shown promising potential in enhancing the explainability as a driving agent producing control predictions along with natural language explanations. However, severe data scarcity due to expensive annotation costs and significant domain gaps between different datasets makes the development of a robust and generalisable system an extremely challenging task. Moreover, the prohibitively expensive training requirements of MLLM and the unsolved problem of catastrophic forgetting further limit their generalisability post-deployment.To address these challenges, we present RAG-Driver, a novel retrieval-augmented multi-modal large language model that leverages in-context learning for high-performance, explainable, and generalisable autonomous driving. By grounding in retrieved expert demonstration, we empirically validate that RAG-Driver achieves state-of-the-art performance in producing driving action explanations, justifications, and control signal prediction. More importantly, it exhibits exceptional zero-shot generalisation capabilities to unseen environments without further training endeavours.

Reinforcement learning provides an appealing framework for robotic control due to its ability to learn expressive policies purely through real-world interaction. However, this requires addressing real-world constraints, including avoidance of catastrophic failures during training, which might severely impede both learning progress and the performance of the final policy. In many robotics settings, this amounts to avoiding certain "unsafe" states. The high-speed off-road driving task represents a particularly challenging instantiation of this problem: a high-return policy should drive as aggressively and as quickly as possible, which often requires getting close to the edge of the set of "safe" states, and therefore places a particular burden on the method to avoid frequent failures. To both learn highly performant policies and avoid excessive failures, we propose a reinforcement learning framework that combines risk-sensitive control with an adaptive action space curriculum. We propose a reinforcement learning objective that uses a risk-sensitive metric to jointly train a policy and iteratively expand action bounds during training, starting with a low-speed policy and slowly increasing the speed over time. Furthermore, we show that our risk-sensitive objective automatically avoids out-of-distribution states when equipped with an estimator for epistemic uncertainty. We implement our algorithm on a small-scale rally car and show that it is capable of learning high-speed policies for a real-world off-road driving task. We show that our method greatly reduces the number of safety violations during the training process, and actually leads to better final high-speed driving policies at the end of training.

Recent years in robotics and imitation learning have shown remarkable progress in training large-scale foundation models by leveraging data across a multitude of embodiments. The success of such policies might lead us to wonder: just how diverse can the robots in the training set be while still facilitating positive transfer? In this work, we study this question in the context of heterogeneous embodiments, examining how even seemingly very different domains such as robotic navigation and manipulation can provide benefits when included in the training data for the same model. We train a single goal-conditioned policy that is capable of controlling a robotic arm, quadcopter, quadruped, and mobile base. We then investigate the extent to which transfer can occur across navigation and manipulation by framing them as a single goal-reaching task. In particular, we find that co-training with navigation data can enhance robustness and performance in goal-conditioned manipulation with a wrist-mounted camera. We then deploy our policy trained only from navigation-only and static manipulation-only data on a mobile manipulator, showing that it can control a similar but novel embodiment in a zero-shot manner. These results provide evidence that large-scale robotic policies can benefit from data collected across a wide variety of embodiments.

One promising approach towards effective robot decision making in complex, long-horizon tasks is to sequence together *parameterized skills*. We consider a setting where a robot is initially equipped with (1) a library of parameterized skills, (2) an AI planner for sequencing together the skills given a goal, and (3) a very general prior distribution for selecting skill parameters. Once deployed, the robot should rapidly and autonomously learn to improve its performance by specializing its skill parameter selection policy to the particular objects, goals, and constraints in its environment. In this work, we focus on the active learning problem of choosing which skills to *practice* to maximize expected future task success. We propose that the robot should *estimate* the competence of each skill, *extrapolate* the competence (asking: "how much would the competence improve through practice?"), and *situate* the skill in the task distribution through competence-aware planning. This approach is implemented within a fully autonomous system where the robot repeatedly plans, practices, and learns without any environment resets. Through experiments in simulation, we find that our approach learns effective parameter policies more sample-efficiently than several baselines. Experiments in the real-world demonstrate our approach’s ability to handle noise from perception and control and improve the robot’s ability to solve two long-horizon mobile-manipulation tasks after a few hours of autonomous practice. Project website: http://ees.csail.mit.edu

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.