Most existing benchmarks for grounding language in interactive environments either lack realistic linguistic elements, or prove difficult to scale up due to substantial human involvement in the collection of data or feedback signals. We develop WebShop – a simulated e-commerce website environment with 1.18 million real-world products and 12,087 crowd-sourced text instructions. In this environment, an agent needs to navigate multiple types of webpages and issue diverse actions to find, customize, and purchase a product given an instruction. WebShop provides several challenges including understanding compositional instructions, query (re-)formulation, dealing with noisy text in webpages, and performing strategic exploration. We collect over 1,600 human trajectories to first validate the benchmark, then train and evaluate a diverse range of agents using reinforcement learning, imitation learning, and pre-trained image and language models. Our best model achieves a task success rate of 29%, which significantly outperforms rule heuristics but is far lower than expert human performance (59%). We also analyze agent and human trajectories and ablate various model components to provide insights for developing future agents with stronger language understanding and decision making abilities. Finally, we show our agent trained on WebShop exhibits non-trivial sim-to-real transfer when evaluated on amazon.com and ebay.com, indicating the potential value of our benchmark for developing practical web agents that can operate in the wild.
more »
« less
This content will become publicly available on July 10, 2024
Energy-based Models are Zero-Shot Planners for Compositional Scene Rearrangement
Language is compositional; an instruction can ex-
press multiple relation constraints to hold among objects in a
scene that a robot is tasked to rearrange. Our focus in this
work is an instructable scene-rearranging framework that gen-
eralizes to longer instructions and to spatial concept compositions
never seen at training time. We propose to represent language-
instructed spatial concepts with energy functions over relative
object arrangements. A language parser maps instructions to
corresponding energy functions and an open-vocabulary visual-
language model grounds their arguments to relevant objects in
the scene. We generate goal scene configurations by gradient
descent on the sum of energy functions, one per language predi-
cate in the instruction. Local vision-based policies then re-locate
objects to the inferred goal locations. We test our model on es-
tablished instruction-guided manipulation benchmarks, as well as
benchmarks of compositional instructions we introduce. We show
our model can execute highly compositional instructions zero-shot
in simulation and in the real world. It outperforms language-
to-action reactive policies and Large Language Model planners
by a large margin, especially for long instructions that involve
compositions of multiple spatial concepts. Simulation and real-
world robot execution videos, as well as our code and datasets are
publicly available on our website: https://ebmplanner.github.io.
more »
« less
- Award ID(s):
- 1849287
- NSF-PAR ID:
- 10496019
- Publisher / Repository:
- Robotics: Science and Systems Foundation
- Date Published:
- Journal Name:
- Robotics: Science and Systems 2023
- ISSN:
- 2330-7668
- Format(s):
- Medium: X
- Location:
- Daegu, Republic of Korea
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Tan, Jie ; Toussaint, Marc (Ed.)With the advent of large language models and large-scale robotic datasets, there has been tremendous progress in high-level decision-making for object manipulation [1, 2, 3, 4]. These generic models are able to interpret complex tasks using language commands, but they often have difficulties generalizing to out-of-distribution objects due to the inability of low-level action primitives. In contrast, existing task-specific models [5, 6] excel in low-level manipulation of unknown objects, but only work for a single type of action. To bridge this gap, we present M2T2, a single model that supplies different types of low-level actions that work robustly on arbitrary objects in cluttered scenes. M2T2 is a transformer model which reasons about contact points and predicts valid gripper poses for different action modes given a raw point cloud of the scene. Trained on a large-scale synthetic dataset with 128K scenes, M2T2 achieves zero-shot sim2real transfer on the real robot, outperforming the baseline system with state- of-the-art task-specific models by about 19% in overall performance and 37.5% in challenging scenes where the object needs to be re-oriented for collision- free placement. M2T2 also achieves state-of-the-art results on a subset of language conditioned tasks in RLBench [7]. Videos of robot experiments on unseen objects in both real world and simulation are available on our project website https://m2-t2.github.io.more » « less
-
The goal of this article is to enable robots to perform robust task execution following human instructions in partially observable environments. A robot’s ability to interpret and execute commands is fundamentally tied to its semantic world knowledge. Commonly, robots use exteroceptive sensors, such as cameras or LiDAR, to detect entities in the workspace and infer their visual properties and spatial relationships. However, semantic world properties are often visually imperceptible. We posit the use of non-exteroceptive modalities including physical proprioception, factual descriptions, and domain knowledge as mechanisms for inferring semantic properties of objects. We introduce a probabilistic model that fuses linguistic knowledge with visual and haptic observations into a cumulative belief over latent world attributes to infer the meaning of instructions and execute the instructed tasks in a manner robust to erroneous, noisy, or contradictory evidence. In addition, we provide a method that allows the robot to communicate knowledge dissonance back to the human as a means of correcting errors in the operator’s world model. Finally, we propose an efficient framework that anticipates possible linguistic interactions and infers the associated groundings for the current world state, thereby bootstrapping both language understanding and generation. We present experiments on manipulators for tasks that require inference over partially observed semantic properties, and evaluate our framework’s ability to exploit expressed information and knowledge bases to facilitate convergence, and generate statements to correct declared facts that were observed to be inconsistent with the robot’s estimate of object properties.more » « less
-
Language understanding is essential for the navigation agent to follow instructions. We observe two kinds of issues in the instructions that can make the navigation task challenging: 1. The mentioned landmarks are not recognizable by the navigation agent due to the different vision abilities of the instructor and the modeled agent. 2. The mentioned landmarks are applicable to multiple targets, thus not distinctive for selecting the target among the candidate viewpoints. To deal with these issues, we design a translator module for the navigation agent to convert the original instructions into easy-to-follow sub-instruction representations at each step. The translator needs to focus on the recognizable and distinctive landmarks based on the agent’s visual abilities and the observed visual environment. To achieve this goal, we create a new synthetic sub-instruction dataset and design specific tasks to train the translator and the navigation agent. We evaluate our approach on Room2Room (R2R), Room4room (R4R), and Room2Room Last (R2R-Last) datasets and achieve state-of-the-art results on multiple benchmarks.more » « less
-
The utility of collaborative manipulators for shared tasks is highly dependent on the speed and accuracy of communication between the human and the robot. The run-time of recently developed probabilistic inference models for situated symbol grounding of natural language instructions depends on the complexity of the representation of the environment in which they reason. As we move towards more complex bi-directional interactions, tasks, and environments, we need intelligent perception models that can selectively infer precise pose, semantics, and affordances of the objects when inferring exhaustively detailed world models is inefficient and prohibits real-time interaction with these robots. In this paper we propose a model of language and perception for the problem of adapting the configuration of the robot perception pipeline for tasks where constructing exhaustively detailed models of the environment is inefficient and in- consequential for symbol grounding. We present experimental results from a synthetic corpus of natural language instructions for robot manipulation in example environments. The results demonstrate that by adapting perception we get significant gains in terms of run-time for perception and situated symbol grounding of the language instructions without a loss in the accuracy of the latter.more » « less
