UniETP: Unifying Environments for Generalizable Embodied Task Planning
2026-07-20 • Robotics
RoboticsComputer Vision and Pattern Recognition
AI summaryⓘ
The authors address a challenge in Embodied Task Planning, where an agent needs to carry out step-by-step actions to complete a task in different virtual environments. They point out that existing simulators each use their own systems, making it hard to compare models or train them well. To fix this, the authors created UniETP, a unified setup that combines four popular simulators under one consistent framework with shared observations and actions. This setup also introduces a new, varied dataset to test how well models handle different task complexities. They run tests on recent models to find where these models struggle, aiming to improve future planning methods.
Embodied Task PlanningsimulatorAI2-THORVirtualHomeHabitatBEHAVIORobservation spaceaction spacetask diversitydataset
Authors
Peiran Xu, Jiaqi Zheng, Ziyou Wang, Yadong Mu
Abstract
This paper focuses on the problem of Embodied Task Planning, where an agent is required to execute a sequence of atomic actions within an interactive environment to complete a user-specified task. Though a variety of simulators and datasets have previously been built for this task, these efforts are largely isolated, with each using its own observation format, action type, and task domain. This fragmentation complicates comprehensive model evaluation and hinders the scalability of training data. As an effort towards generalizable embodied planning, we propose UniETP, a unified interface integrating four commonly-used simulators (AI2-THOR, VirtualHome, Habitat, BEHAVIOR). UniETP is characterized by both standardization and diversity. On one hand, it formalizes all the simulators into a consistent observation and action space, and builds an evaluation system to support complicated task goal. On the other hand, it enhances task diversity and complexity across dimensions like task logic, instance grounding, and instruction understanding, constructing a new dataset with varied levels of difficulty in an automatic manner. Extensive experiments on the proposed benchmark are conducted to evaluate the embodied planning capabilities of recent models and analyze the performance bottlenecks. Codes and data will be available at https://github.com/woyut/UniETP .