Imitation learning is a powerful tool for training robot manipulation policies, allowing them to learn from expert demonstrations without manual programming or trial-and-error. However, common methods of data collection, such as human supervision, scale poorly, as they are time-consuming and labor-intensive. In contrast, Task and Motion Planning (TAMP) can autonomously generate large-scale datasets of diverse demonstrations. In this work, we show that the combination of large-scale datasets generated by TAMP supervisors and flexible Transformer models to fit them is a powerful paradigm for robot manipulation. To that end, we present a novel imitation learning system called OPTIMUS that trains large-scale visuomotor Transformer policies by imitating a TAMP agent. OPTIMUS introduces a pipeline for generating TAMP data that is specifically curated for imitation learning and can be used to train performant transformer-based policies. In this paper, we present a thorough study of the design decisions required to imitate TAMP and demonstrate that OPTIMUS can solve a wide variety of challenging vision-based manipulation tasks with over 70 different objects, ranging from long-horizon pick-and-place tasks, to shelf and articulated object manipulation, achieving 70 to 80% success rates. Video results and code at https://mihdalal.github.io/optimus/
Reinforcement learning (RL), imitation learning (IL), and task and motion planning (TAMP) have demonstrated impressive performance across various robotic manipulation tasks. However, these approaches have been limited to learning simple behaviors in current real-world manipulation benchmarks, such as pushing or pick-and-place. To enable more complex, long-horizon behaviors of an autonomous robot, we propose to focus on real-world furniture assembly, a complex, long-horizon robot manipulation task that requires addressing many current robotic manipulation challenges to solve. We present FurnitureBench, a reproducible real-world furniture assembly benchmark aimed at providing a low barrier for entry and being easily reproducible, so that researchers across the world can reliably test their algorithms and compare them against prior work. For ease of use, we provide 200+ hours of pre-collected data (5000+ demonstrations), 3D printable furniture models, a robotic environment setup guide, and systematic task initialization. Furthermore, we provide FurnitureSim, a fast and realistic simulator of FurnitureBench. We benchmark the performance of offline RL and IL algorithms on our assembly tasks and demonstrate the need to improve such algorithms to be able to solve our tasks in the real world, providing ample opportunities for future research.
Combining symbolic and geometric reasoning in multi-agent systems is a challenging task that involves planning, scheduling, and synchronization problems. Existing works overlooked the variability of task duration and geometric feasibility that is intrinsic to these systems because of the interaction between agents and the environment. We propose a combined task and motion planning approach to optimize sequencing, assignment, and execution of tasks under temporal and spatial variability. The framework relies on decoupling tasks and actions, where an action is one possible geometric realization of a symbolic task. At the task level, timeline-based planning deals with temporal constraints, duration variability, and synergic assignment of tasks. At the action level, online motion planning plans for the actual movements dealing with environmental changes. We demonstrate the approach effectiveness in a collaborative manufacturing scenario, in which a robotic arm and a human worker shall assemble a mosaic in the shortest time possible. Compared with existing works, our approach applies to a broader range of applications and reduces the execution time of the process.
To achieve optimal robot behavior in dynamic scenarios we need to consider complex dynamics in a predictive manner. In the vehicle dynamics community, it is well know that to achieve time-optimal driving on low surface, the vehicle should utilize drifting. Hence many authors have devised rules to split circuits and employ drifting on some segments. These rules are suboptimal and do not generalize to arbitrary circuit shapes (e.g., S-like curves). So, the question "When to go into which mode and how to drive in it?" remains unanswered. To choose the suitable mode (discrete decision), the algorithm needs information about the feasibility of the continuous motion in that mode. This makes it a class of Task and Motion Planning (TAMP) problems, which are known to be hard to solve optimally in real-time. In the AI planning community, search methods are commonly used. However, they cannot be directly applied to TAMP problems due to the continuous component. Here, we present a search-based method that effectively solves this problem and efficiently searches in a highly dimensional state space with nonlinear and unstable dynamics. The space of the possible trajectories is explored by sampling different combinations of motion primitives guided by the search. Our approach allows to use multiple locally approximated models to generate motion primitives (e.g., learned models of drifting) and effectively simplify the problem without losing accuracy. The algorithm performance is evaluated in simulated driving on a mixed-track with segments of different curvatures (right and left). Our code is available at https://git.io/JenvB
We present Orbit, a unified and modular framework for robot learning powered by NVIDIA Isaac Sim. It offers a modular design to easily and efficiently create robotic environments with photo-realistic scenes and high-fidelity rigid and deformable body simulation. With Orbit, we provide a suite of benchmark tasks of varying difficulty -- from single-stage cabinet opening and cloth folding to multi-stage tasks such as room reorganization. To support working with diverse observations and action spaces, we include fixed-arm and mobile manipulators with different physically-based sensors and motion generators. Orbit allows training reinforcement learning policies and collecting large demonstration datasets from hand-crafted or expert solutions in a matter of minutes by leveraging GPU-based parallelization. In summary, we offer an open-sourced framework that readily comes with 16 robotic platforms, 4 sensor modalities, 10 motion generators, more than 20 benchmark tasks, and wrappers to 4 learning libraries. With this framework, we aim to support various research areas, including representation learning, reinforcement learning, imitation learning, and task and motion planning. We hope it helps establish interdisciplinary collaborations in these communities, and its modularity makes it easily extensible for more tasks and applications in the future.
We present a learning-enabled Task and Motion Planning (TAMP) algorithm for solving mobile manipulation problems in environments with many articulated and movable obstacles. Our idea is to bias the search procedure of a traditional TAMP planner with a learned plan feasibility predictor. The core of our algorithm is PIGINet, a novel Transformer-based learning method that takes in a task plan, the goal, and the initial state, and predicts the probability of finding motion trajectories associated with the task plan. We integrate PIGINet within a TAMP planner that generates a diverse set of high-level task plans, sorts them by their predicted likelihood of feasibility, and refines them in that order. We evaluate the runtime of our TAMP algorithm on seven families of kitchen rearrangement problems, comparing its performance to that of non-learning baselines. Our experiments show that PIGINet substantially improves planning efficiency, cutting down runtime by 80% on problems with small state spaces and 10%-50% on larger ones, after being trained on only 150-600 problems. Finally, it also achieves zero-shot generalization to problems with unseen object categories thanks to its visual encoding of objects. Project page https://piginet.github.io/.
PDDLStream solvers have recently emerged as viable solutions for Task and Motion Planning (TAMP) problems, extending PDDL to problems with continuous action spaces. Prior work has shown how PDDLStream problems can be reduced to a sequence of PDDL planning problems, which can then be solved using off-the-shelf planners. However, this approach can suffer from long runtimes. In this paper we propose LAZY, a solver for PDDLStream problems that maintains a single integrated search over action skeletons, which gets progressively more geometrically informed, as samples of possible motions are lazily drawn during motion planning. We explore how learned models of goal-directed policies and current motion sampling data can be incorporated in LAZY to adaptively guide the task planner. We show that this leads to significant speed-ups in the search for a feasible solution evaluated over unseen test environments of varying numbers of objects, goals, and initial conditions. We evaluate our TAMP approach by comparing to existing solvers for PDDLStream problems on a range of simulated 7DoF rearrangement/manipulation problems.
Advances in robotic skill acquisition have made it possible to build general-purpose libraries of learned skills for downstream manipulation tasks. However, naively executing these skills one after the other is unlikely to succeed without accounting for dependencies between actions prevalent in long-horizon plans. We present Sequencing Task-Agnostic Policies (STAP), a scalable framework for training manipulation skills and coordinating their geometric dependencies at planning time to solve long-horizon tasks never seen by any skill during training. Given that Q-functions encode a measure of skill feasibility, we formulate an optimization problem to maximize the joint success of all skills sequenced in a plan, which we estimate by the product of their Q-values. Our experiments indicate that this objective function approximates ground truth plan feasibility and, when used as a planning objective, reduces myopic behavior and thereby promotes long-horizon task success. We further demonstrate how STAP can be used for task and motion planning by estimating the geometric feasibility of skill sequences provided by a task planner. We evaluate our approach in simulation and on a real robot. Qualitative results and code are made available at https://sites.google.com/stanford.edu/stap.
To assist with everyday human activities, robots must solve complex long-horizon tasks and generalize to new settings. Recent deep reinforcement learning (RL) methods show promise in fully autonomous learning, but they struggle to reach long-term goals in large environments. On the other hand, Task and Motion Planning (TAMP) approaches excel at solving and generalizing across long-horizon tasks, thanks to their powerful state and action abstractions. But they assume predefined skill sets, which limits their real-world applications. In this work, we combine the benefits of these two paradigms and propose an integrated task planning and skill learning framework named LEAGUE (Learning and Abstraction with Guidance). LEAGUE leverages the symbolic interface of a task planner to guide RL-based skill learning and creates abstract state space to enable skill reuse. More importantly, LEAGUE learns manipulation skills in-situ of the task planning system, continuously growing its capability and the set of tasks that it can solve. We evaluate LEAGUE on four challenging simulated task domains and show that LEAGUE outperforms baselines by large margins. We also show that the learned skills can be reused to accelerate learning in new tasks domains and transfer to a physical robot platform.
We present a multi-robot task and motion planning method that, when applied to the rearrangement of objects by manipulators, results in solution times up to three orders of magnitude faster than existing methods and successfully plans for problems with up to twenty objects, more than three times as many objects as comparable methods. We achieve this improvement by decomposing the planning space to consider manipulators alone, objects, and manipulators holding objects. We represent this decomposition with a hypergraph where vertices are decomposed elements of the planning spaces and hyperarcs are transitions between elements. Existing methods use graph-based representations where vertices are full composite spaces and edges are transitions between these. Using the hypergraph reduces the representation size of the planning space-for multi-manipulator object rearrangement, the number of hypergraph vertices scales linearly with the number of either robots or objects, while the number of hyperarcs scales quadratically with the number of robots and linearly with the number of objects. In contrast, the number of vertices and edges in graph-based representations scales exponentially in the number of robots and objects. We show that similar gains can be achieved for other multi-robot task and motion planning problems.
Decision-making is challenging in robotics environments with continuous object-centric states, continuous actions, long horizons, and sparse feedback. Hierarchical approaches, such as task and motion planning (TAMP), address these challenges by decomposing decision-making into two or more levels of abstraction. In a setting where demonstrations and symbolic predicates are given, prior work has shown how to learn symbolic operators and neural samplers for TAMP with manually designed parameterized policies. Our main contribution is a method for learning parameterized polices in combination with operators and samplers. These components are packaged into modular neuro-symbolic skills and sequenced together with search-then-sample TAMP to solve new tasks. In experiments in four robotics domains, we show that our approach -- bilevel planning with neuro-symbolic skills -- can solve a wide range of tasks with varying initial states, goals, and objects, outperforming six baselines and ablations. Video: https://youtu.be/PbFZP8rPuGg Code: https://tinyurl.com/skill-learning
The booming of electric vehicles demands efficient battery disassembly for recycling to be environment-friendly. Currently, battery disassembly is still primarily done by humans, probably assisted by robots, due to the unstructured environment and high uncertainties. It is highly desirable to design autonomous solutions to improve work efficiency and lower human risks in high voltage and toxic environments. This paper proposes a novel neurosymbolic method, which augments the traditional Variational Autoencoder (VAE) model to learn symbolic operators based on raw sensory inputs and their relationships. The symbolic operators include a probabilistic state symbol grounding model and a state transition matrix for predicting states after each execution to enable autonomous task and motion planning. At last, the method's feasibility is verified through test results.
As robots play an increasingly important role in the industrial, the expectations about their applications for everyday living tasks are getting higher. Robots need to perform long-horizon tasks that consist of several sub-tasks that need to be accomplished. Task and Motion Planning (TAMP) provides a hierarchical framework to handle the sequential nature of manipulation tasks by interleaving a symbolic task planner that generates a possible action sequence, with a motion planner that checks the kinematic feasibility in the geometric world, generating robot trajectories if several constraints are satisfied, e.g., a collision-free trajectory from one state to another. Hence, the reasoning about the task plan's geometric grounding is taken over by the motion planner. However, motion planning is computationally intense and is usability as feasibility checker casts TAMP methods inapplicable to real-world scenarios. In this paper, we introduce neural feasibility classifier (NFC), a simple yet effective visual heuristic for classifying the feasibility of proposed actions in TAMP. Namely, NFC will identify infeasible actions of the task planner without the need for costly motion planning, hence reducing planning time in multi-step manipulation tasks. NFC encodes the image of the robot's workspace into a feature map thanks to convolutional neural network (CNN). We train NFC using simulated data from TAMP problems and label the instances based on IK feasibility checking. Our empirical results in different simulated manipulation tasks show that our NFC generalizes to the entire robot workspace and has high prediction accuracy even in scenes with multiple obstructions. When combined with state-of-the-art integrated TAMP, our NFC enhances its performance while reducing its planning time.
We present a framework for learning to guide geometric task and motion planning (GTAMP). GTAMP is a subclass of task and motion planning in which the goal is to move multiple objects to target regions among movable obstacles. A standard graph search algorithm is not directly applicable, because GTAMP problems involve hybrid search spaces and expensive action feasibility checks. To handle this, we introduce a novel planner that extends basic heuristic search with random sampling and a heuristic function that prioritizes feasibility checking on promising state action pairs. The main drawback of such pure planners is that they lack the ability to learn from planning experience to improve their efficiency. We propose two learning algorithms to address this. The first is an algorithm for learning a rank function that guides the discrete task level search, and the second is an algorithm for learning a sampler that guides the continuous motionlevel search. We propose design principles for designing data efficient algorithms for learning from planning experience and representations for effective generalization. We evaluate our framework in challenging GTAMP problems, and show that we can improve both planning and data efficiency
Multi-robot assembly systems are becoming increasingly appealing in manufacturing due to their ability to automatically, flexibly, and quickly construct desired structural designs. However, effectively planning for these systems in a manner that ensures each robot is simultaneously productive, and not idle, is challenging due to (1) the close proximity that the robots must operate in to manipulate the structure and (2) the inherent structural partial orderings on when each part can be installed. In this paper, we present a task and motion planning framework that jointly plans safe, low-makespan plans for a team of robots to assemble complex spatial structures. Our framework takes a hierarchical approach that, at the high level, uses Mixed-integer Linear Programs to compute an abstract plan comprised of an allocation of robots to tasks subject to precedence constraints and, at the low level, builds on a state-of-the-art algorithm for Multi-Agent Path Finding to plan collision-free robot motions that realize this abstract plan. Critical to our approach is the inclusion of certain collision constraints and movement durations during high-level planning, which better informs the search for abstract plans that are likely to be both feasible and low-makespan while keeping the search tractable. We demonstrate our planning system on several challenging assembly domains with several (sometimes heterogeneous) robots with grippers or suction plates for assembling structures with up to 23 objects involving Lego bricks, bars, plates, or irregularly shaped blocks.
While modern policy optimization methods can do complex manipulation from sensory data, they struggle on problems with extended time horizons and multiple sub-goals. On the other hand, task and motion planning (TAMP) methods scale to long horizons but they are computationally expensive and need to precisely track world state. We propose a method that draws on the strength of both methods: we train a policy to imitate a TAMP solver's output. This produces a feed-forward policy that can accomplish multi-step tasks from sensory data. First, we build an asynchronous distributed TAMP solver that can produce supervision data fast enough for imitation learning. Then, we propose a hierarchical policy architecture that lets us use partially trained control policies to speed up the TAMP solver. In robotic manipulation tasks with 7-DoF joint control, the partially trained policies reduce the time needed for planning by a factor of up to 2.6. Among these tasks, we can learn a policy that solves the RoboSuite 4-object pick-place task 88% of the time from object pose observations and a policy that solves the RoboDesk 9-goal benchmark 79% of the time from RGB images (averaged across the 9 disparate tasks).
Task and motion planning problems in robotics combine symbolic planning over discrete task variables with motion optimization over continuous state and action variables. Recent works such as PDDLStream have focused on optimistic planning with an incrementally growing set of objects until a feasible trajectory is found. However, this set is exhaustively expanded in a breadth-first manner, regardless of the logical and geometric structure of the problem at hand, which makes long-horizon reasoning with large numbers of objects prohibitively time-consuming. To address this issue, we propose a geometrically informed symbolic planner that expands the set of objects and facts in a best-first manner, prioritized by a Graph Neural Network that is learned from prior search computations. We evaluate our approach on a diverse set of problems and demonstrate an improved ability to plan in difficult scenarios. We also apply our algorithm on a 7DOF robotic arm in block-stacking manipulation tasks.
This paper presents JEDAI, an AI system designed for outreach and educational efforts aimed at non-AI experts. JEDAI features a novel synthesis of research ideas from integrated task and motion planning and explainable AI. JEDAI helps users create high-level, intuitive plans while ensuring that they will be executable by the robot. It also provides users customized explanations about errors and helps improve their understanding of AI planning as well as the limits and capabilities of the underlying robot system.
This study proposes a hierarchically integrated framework for safe task and motion planning (TAMP) of bipedal locomotion in a partially observable environment with dynamic obstacles and uneven terrain. The high-level task planner employs linear temporal logic (LTL) for a reactive game synthesis between the robot and its environment and provides a formal guarantee on navigation safety and task completion. To address environmental partial observability, a belief abstraction is employed at the high-level navigation planner to estimate the dynamic obstacles' location. Accordingly, a synthesized action planner sends a set of locomotion actions to the middle-level motion planner, while incorporating safe locomotion specifications extracted from safety theorems based on a reduced-order model (ROM) of the locomotion process. The motion planner employs the ROM to design safety criteria and a sampling algorithm to generate non-periodic motion plans that accurately track high-level actions. At the low level, a foot placement controller based on an angular-momentum linear inverted pendulum model is implemented and integrated with an ankle-actuated passivity-based controller for full-body trajectory tracking. To address external perturbations, this study also investigates safe sequential composition of the keyframe locomotion state and achieves robust transitions against external perturbations through reachability analysis. The overall TAMP framework is validated with extensive simulations and hardware experiments on bipedal walking robots Cassie and Digit designed by Agility Robotics.
We present an AND/OR graph-based, integrated multi-robot task and motion planning approach which (i) performs task allocation coordinating the activity of a given number of robots, and (ii) is capable of handling tasks which involve an a priori unknown number of object re-arrangements, such as those involved in retrieving objects from cluttered workspaces. Such situations may arise, for example, in search and rescue scenarios, while locating/picking a cluttered object of interest. The corresponding problem falls under the category of planning in clutter. One of the challenges while planning in clutter is that the number of object re-arrangements required to pick the target object is not known beforehand, in general. Moreover, such tasks can be decomposed in a variety of ways, since different cluttering object re-arrangements are possible to reach the target object. In our approach, task allocation and decomposition is achieved by maximizing a combined utility function. The allocated tasks are performed by an integrated task and motion planner, which is robust to the requirement of an unknown number of re-arrangement tasks. We demonstrate our results with experiments in simulation on two Franka Emika manipulators.