arXiv subjects

Robert Platt

Publications and source records attributed to Robert Platt.

At least 55 records · Page 3Linked to original sources

arXiv2022

SEIL: Simulation-augmented Equivariant Imitation Learning

In robotic manipulation, acquiring samples is extremely expensive because it often requires interacting with the real world. Traditional image-level data augmentation has shown the potential to improve sample efficiency in various machine learning tasks. However, image-level data augmentation is insufficient for an imitation learning agent to learn good manipulation policies in a reasonable amount of demonstrations. We propose Simulation-augmented Equivariant Imitation Learning (SEIL), a method that combines a novel data augmentation strategy of supplementing expert trajectories with simulated transitions and an equivariant model that exploits the $\mathrm{O}(2)$ symmetry in robotic manipulation. Experimental evaluations demonstrate that our method can learn non-trivial manipulation tasks within ten demonstrations and outperforms the baselines with a significant margin.

cs.RO↗

arXiv2022

On-Robot Learning With Equivariant Models

Recently, equivariant neural network models have been shown to improve sample efficiency for tasks in computer vision and reinforcement learning. This paper explores this idea in the context of on-robot policy learning in which a policy must be learned entirely on a physical robotic system without reference to a model, a simulator, or an offline dataset. We focus on applications of Equivariant SAC to robotic manipulation and explore a number of variations of the algorithm. Ultimately, we demonstrate the ability to learn several non-trivial manipulation tasks completely through on-robot experiences in less than an hour or two of wall clock time.

cs.RO↗

arXiv2022

BulletArm: An Open-Source Robotic Manipulation Benchmark and Learning Framework

We present BulletArm, a novel benchmark and learning-environment for robotic manipulation. BulletArm is designed around two key principles: reproducibility and extensibility. We aim to encourage more direct comparisons between robotic learning methods by providing a set of standardized benchmark tasks in simulation alongside a collection of baseline algorithms. The framework consists of 31 different manipulation tasks of varying difficulty, ranging from simple reaching and picking tasks to more realistic tasks such as bin packing and pallet stacking. In addition to the provided tasks, BulletArm has been built to facilitate easy expansion and provides a suite of tools to assist users when adding new tasks to the framework. Moreover, we introduce a set of five benchmarks and evaluate them using a series of state-of-the-art baseline algorithms. By including these algorithms as part of our framework, we hope to encourage users to benchmark their work on any new tasks against these baselines.

cs.RO↗

arXiv2022

Equivariant Transporter Network

Transporter Net is a recently proposed framework for pick and place that is able to learn good manipulation policies from a very few expert demonstrations. A key reason why Transporter Net is so sample efficient is that the model incorporates rotational equivariance into the pick module, i.e. the model immediately generalizes learned pick knowledge to objects presented in different orientations. This paper proposes a novel version of Transporter Net that is equivariant to both pick and place orientation. As a result, our model immediately generalizes place knowledge to different place orientations in addition to generalizing pick knowledge as before. Ultimately, our new model is more sample efficient and achieves better pick and place success rates than the baseline Transporter Net model.

cs.RO↗

arXiv2022

Graph-Structured Policy Learning for Multi-Goal Manipulation Tasks

Multi-goal policy learning for robotic manipulation is challenging. Prior successes have used state-based representations of the objects or provided demonstration data to facilitate learning. In this paper, by hand-coding a high-level discrete representation of the domain, we show that policies to reach dozens of goals can be learned with a single network using Q-learning from pixels. The agent focuses learning on simpler, local policies which are sequenced together by planning in the abstract space. We compare our method against standard multi-goal RL baselines, as well as other methods that leverage the discrete representation, on a challenging block construction domain. We find that our method can build more than a hundred different block structures, and demonstrate forward transfer to structures with novel objects. Lastly, we deploy the policy learned in simulation on a real robot.

cs.RO↗

arXiv2022

Visual Foresight With a Local Dynamics Model

Model-free policy learning has been shown to be capable of learning manipulation policies which can solve long-time horizon tasks using single-step manipulation primitives. However, training these policies is a time-consuming process requiring large amounts of data. We propose the Local Dynamics Model (LDM) which efficiently learns the state-transition function for these manipulation primitives. By combining the LDM with model-free policy learning, we can learn policies which can solve complex manipulation tasks using one-step lookahead planning. We show that the LDM is both more sample-efficient and outperforms other model architectures. When combined with planning, we can outperform other model-based and model-free policies on several challenging manipulation tasks in simulation.

cs.RO↗

arXiv2022

Hierarchical Reinforcement Learning under Mixed Observability

The framework of mixed observable Markov decision processes (MOMDP) models many robotic domains in which some state variables are fully observable while others are not. In this work, we identify a significant subclass of MOMDPs defined by how actions influence the fully observable components of the state and how those, in turn, influence the partially observable components and the rewards. This unique property allows for a two-level hierarchical approach we call HIerarchical Reinforcement Learning under Mixed Observability (HILMO), which restricts partial observability to the top level while the bottom level remains fully observable, enabling higher learning efficiency. The top level produces desired goals to be reached by the bottom level until the task is solved. We further develop theoretical guarantees to show that our approach can achieve optimal and quasi-optimal behavior under mild assumptions. Empirical results on long-horizon continuous control tasks demonstrate the efficacy and efficiency of our approach in terms of improved success rate, sample efficiency, and wall-clock training time. We also deploy policies learned in simulation on a real robot.

cs.RO↗

arXiv2022

Binding Actions to Objects in World Models

We study the problem of binding actions to objects in object-factored world models using action-attention mechanisms. We propose two attention mechanisms for binding actions to objects, soft attention and hard attention, which we evaluate in the context of structured world models for five environments. Our experiments show that hard attention helps contrastively-trained structured world models to learn to separate individual objects in an object-based grid-world environment. Further, we show that soft attention increases performance of factored world models trained on a robotic manipulation task. The learned action attention weights can be used to interpret the factored world model as the attention focuses on the manipulated object in the environment.

cs.LG↗

arXiv2022

Efficient and Accurate Candidate Generation for Grasp Pose Detection in SE(3)

Grasp detection of novel objects in unstructured environments is a key capability in robotic manipulation. For 2D grasp detection problems where grasps are assumed to lie in the plane, it is common to design a fully convolutional neural network that predicts grasps over an entire image in one step. However, this is not possible for grasp pose detection where grasp poses are assumed to exist in SE(3). In this case, it is common to approach the problem in two steps: grasp candidate generation and candidate classification. Since grasp candidate classification is typically expensive, the problem becomes one of efficiently identifying high quality candidate grasps. This paper proposes a new grasp candidate generation method that significantly outperforms major 3D grasp detection baselines. Supplementary material is available at https://atenpas.github.io/psn/.

cs.RO↗

arXiv2022

Tactile Pose Estimation and Policy Learning for Unknown Object Manipulation

Object pose estimation methods allow finding locations of objects in unstructured environments. This is a highly desired skill for autonomous robot manipulation as robots need to estimate the precise poses of the objects in order to manipulate them. In this paper, we investigate the problems of tactile pose estimation and manipulation for category-level objects. Our proposed method uses a Bayes filter with a learned tactile observation model and a deterministic motion model. Later, we train policies using deep reinforcement learning where the agents use the belief estimation from the Bayes filter. Our models are trained in simulation and transferred to the real world. We analyze the reliability and the performance of our framework through a series of simulated and real-world experiments and compare our method to the baseline work. Our results show that the learned tactile observation model can localize the pose of novel objects at 2-mm and 1-degree resolution for position and orientation, respectively. Furthermore, we experiment on a bottle opening task where the gripper needs to reach the desired grasp state.

cs.RO↗

arXiv2022

$\mathrm{SO}(2)$-Equivariant Reinforcement Learning

Equivariant neural networks enforce symmetry within the structure of their convolutional layers, resulting in a substantial improvement in sample efficiency when learning an equivariant or invariant function. Such models are applicable to robotic manipulation learning which can often be formulated as a rotationally symmetric problem. This paper studies equivariant model architectures in the context of $Q$-learning and actor-critic reinforcement learning. We identify equivariant and invariant characteristics of the optimal $Q$-function and the optimal policy and propose equivariant DQN and SAC algorithms that leverage this structure. We present experiments that demonstrate that our equivariant versions of DQN and SAC can be significantly more sample efficient than competing algorithms on an important class of robotic manipulation problems.

cs.RO↗

arXiv2022

Sample Efficient Grasp Learning Using Equivariant Models

In planar grasp detection, the goal is to learn a function from an image of a scene onto a set of feasible grasp poses in $\mathrm{SE}(2)$. In this paper, we recognize that the optimal grasp function is $\mathrm{SE}(2)$-equivariant and can be modeled using an equivariant convolutional neural network. As a result, we are able to significantly improve the sample efficiency of grasp learning, obtaining a good approximation of the grasp function after only 600 grasp attempts. This is few enough that we can learn to grasp completely on a physical robot in about 1.5 hours.

cs.RO↗

arXiv2022

Factored World Models for Zero-Shot Generalization in Robotic Manipulation

World models for environments with many objects face a combinatorial explosion of states: as the number of objects increases, the number of possible arrangements grows exponentially. In this paper, we learn to generalize over robotic pick-and-place tasks using object-factored world models, which combat the combinatorial explosion by ensuring that predictions are equivariant to permutations of objects. Previous object-factored models were limited either by their inability to model actions, or by their inability to plan for complex manipulation tasks. We build on recent contrastive methods for training object-factored world models, which we extend to model continuous robot actions and to accurately predict the physics of robotic pick-and-place. To do so, we use a residual stack of graph neural networks that receive action information at multiple levels in both their node and edge neural networks. Crucially, our learned model can make predictions about tasks not represented in the training data. That is, we demonstrate successful zero-shot generalization to novel tasks, with only a minor decrease in model performance. Moreover, we show that an ensemble of our models can be used to plan for tasks involving up to 12 pick and place actions using heuristic search. We also demonstrate transfer to a physical robot.

cs.RO↗

arXiv2022

GASCN: Graph Attention Shape Completion Network

Shape completion, the problem of inferring the complete geometry of an object given a partial point cloud, is an important problem in robotics and computer vision. This paper proposes the Graph Attention Shape Completion Network (GASCN), a novel neural network model that solves this problem. This model combines a graph-based model for encoding local point cloud information with an MLP-based architecture for encoding global information. For each completed point, our model infers the normal and extent of the local surface patch which is used to produce dense yet precise shape completions. We report experiments that demonstrate that GASCN outperforms standard shape completion methods on a standard benchmark drawn from the Shapenet dataset.

cs.CV↗

arXiv2021

A Targeted Approach to Confounder Selection for High-Dimensional Data

We consider the problem of selecting confounders for adjustment from a potentially large set of covariates, when estimating a causal effect. Recently, the high-dimensional Propensity Score (hdPS) method was developed for this task; hdPS ranks potential confounders by estimating an importance score for each variable and selects the top few variables. However, this ranking procedure is limited: it requires all variables to be binary. We propose an extension of the hdPS to general types of response and confounder variables. We further develop a group importance score, allowing us to rank groups of potential confounders. The main challenge is that our parameter requires either the propensity score or response model; both vulnerable to model misspecification. We propose a targeted maximum likelihood estimator (TMLE) which allows the use of nonparametric, machine learning tools for fitting these intermediate models. We establish asymptotic normality of our estimator, which consequently allows constructing confidence intervals. We complement our work with numerical studies on simulated and real data.

stat.ME↗

arXiv2021

Equivariant $Q$ Learning in Spatial Action Spaces

Recently, a variety of new equivariant neural network model architectures have been proposed that generalize better over rotational and reflectional symmetries than standard models. These models are relevant to robotics because many robotics problems can be expressed in a rotationally symmetric way. This paper focuses on equivariance over a visual state space and a spatial action space -- the setting where the robot action space includes a subset of $\rm{SE}(2)$. In this situation, we know a priori that rotations and translations in the state image should result in the same rotations and translations in the spatial action dimensions of the optimal policy. Therefore, we can use equivariant model architectures to make $Q$ learning more sample efficient. This paper identifies when the optimal $Q$ function is equivariant and proposes $Q$ network architectures for this setting. We show experimentally that this approach outperforms standard methods in a set of challenging manipulation problems.

cs.RO↗

arXiv2021

Belief-Grounded Networks for Accelerated Robot Learning under Partial Observability

Many important robotics problems are partially observable in the sense that a single visual or force-feedback measurement is insufficient to reconstruct the state. Standard approaches involve learning a policy over beliefs or observation-action histories. However, both of these have drawbacks; it is expensive to track the belief online, and it is hard to learn policies directly over histories. We propose a method for policy learning under partial observability called the Belief-Grounded Network (BGN) in which an auxiliary belief-reconstruction loss incentivizes a neural network to concisely summarize its input history. Since the resulting policy is a function of the history rather than the belief, it can be executed easily at runtime. We compare BGN against several baselines on classic benchmark tasks as well as three novel robotic touch-sensing tasks. BGN outperforms all other tested methods and its learned policies work well when transferred onto a physical robot.

cs.RO↗

arXiv2021

Robotic Pick-and-Place With Uncertain Object Instance Segmentation and Shape Completion

We consider robotic pick-and-place of partially visible, novel objects, where goal placements are non-trivial, e.g., tightly packed into a bin. One approach is (a) use object instance segmentation and shape completion to model the objects and (b) use a regrasp planner to decide grasps and places displacing the models to their goals. However, it is critical for the planner to account for uncertainty in the perceived models, as object geometries in unobserved areas are just guesses. We account for perceptual uncertainty by incorporating it into the regrasp planner's cost function. We compare seven different costs. One of these, which uses neural networks to estimate probability of grasp and place stability, consistently outperforms uncertainty-unaware costs and evaluates faster than Monte Carlo sampling. On a real robot, the proposed cost results in successfully packing objects tightly into a bin 7.8% more often versus the commonly used minimum-number-of-grasps cost.

cs.RO↗