Abstract We propose Text2Motion, a language-based planning framework enabling robots
to solve sequential manipulation tasks that require long-horizon reasoning. Given a natural …

Large language models (LLMs) have demonstrated impressive results in developing
generalist planning agents for diverse tasks. However, grounding these plans in expansive …

We present a method to learn compositional multi-object dynamics models from image
observations based on implicit object encoders, Neural Radiance Fields (NeRFs), and …

Robot manipulation of multiple objects is an important topic for applications including
warehouse automation, service robots performing cleaning, and large-scale object sorting …

Large language models (LLMs) have demonstrated impressive results in developing
generalist planning agents for diverse tasks. However, grounding these plans in expansive …

Humans excel in complex long-horizon soft body manipulation tasks via flexible tool use:
bread baking requires a knife to slice the dough and a rolling pin to flatten it. Often regarded …

Abstract 3D scene graphs (3DSGs) are an emerging description; unifying symbolic,
topological, and metric scene representations. However, typical 3DSGs contain hundreds of …

In robotic domains, learning and planning are complicated by continuous state spaces,
continuous action spaces, and long task horizons. In this work, we address these challenges …

We present PDDLGym, a framework that automatically constructs OpenAI Gym
environments from PDDL domains and problems. Observations and actions in PDDLGym …

Manipulation tasks, like loading a dishwasher, can be seen as a sequence of spatial
constraints and relationships between different objects. We aim to discover these rules from …