An official website of the United States government
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
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
- PAR ID:
- 10496019
- Publisher / Repository:
- Robotics: Science and Systems Foundation
- Date Published:
- Journal Name:
- Robotics: Science and Systems 2023
- ISSN:
- 2330-7668
- ISBN:
- 978-0-9923747-9-2
- Format(s):
- Medium: X
- Location:
- Daegu, Republic of Korea
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Humans often use natural language instructions to control and interact with robots for task execution. This poses a big challenge to robots that need to not only parse and understand human instructions but also realise semantic understanding of an unknown environment and its constituent elements. To address this challenge, this study presents a vision-language model (VLM)-driven approach to scene understanding of an unknown environment to enable robotic object manipulation. Given language instructions, a pretrained vision-language model built on open-sourced Llama2-chat (7B) as the language model backbone is adopted for image description and scene understanding, which translates visual information into text descriptions of the scene. Next, a zero-shot-based approach to fine-grained visual grounding and object detection is developed to extract and localise objects of interest from the scene task. Upon 3D reconstruction and pose estimate establishment of the object, a code-writing large language model (LLM) is adopted to generate high-level control codes and link language instructions with robot actions for downstream tasks. The performance of the developed approach is experimentally validated through table-top object manipulation by a robot.more » « less
-
We propose to decompose instruction execution to goal prediction and action generation. We design a model that maps raw visual observations to goals using LINGUNET, a language-conditioned image generation network, and then generates the actions required to complete them. Our model is trained from demonstration only without external resources. To evaluate our approach, we introduce two benchmarks for instruction following: LANI, a navigation task; and CHAI, where an agent executes household instructions. Our evaluation demonstrates the advantages of our model decomposition, and illustrates the challenges posed by our new benchmarks.more » « less
-
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
-
Inferring Compact Representations for Efficient Natural Language Understanding of Robot InstructionsThe speed and accuracy with which robots are able to interpret natural language is fundamental to realizing effective human-robot interaction. A great deal of attention has been paid to developing models and approximate inference algorithms that improve the efficiency of language understanding. However, existing methods still attempt to reason over a representation of the environment that is flat and unnecessarily detailed, which limits scalability. An open problem is then to develop methods capable of producing the most compact environment model sufficient for accurate and efficient natural language understanding. We propose a model that leverages environment-related information encoded within instructions to identify the subset of observations and perceptual classifiers necessary to perceive a succinct, instruction-specific environment representation. The framework uses three probabilistic graphical models trained from a corpus of annotated instructions to infer salient scene semantics, perceptual classifiers, and grounded symbols. Experimental results on two robots operating in different environments demonstrate that by exploiting the content and the structure of the instructions, our method learns compact environment representations that significantly improve the efficiency of natural language symbol grounding.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.15607/RSS.2023.XIX.030
