AI2-THOR: A virtual environment for training home assistant robots

September 8, 2022

AI2-THOR is a framework based on Unity to train robots for a variety of Embodied AI tasks. Get a look into the design of AI2-THOR and how it is used to train robots that navigate, follow language instructions and understand the world around them.

Training robots is a computationally intensive task, and it can take robots days to learn simple tasks such as navigation. See how Unity Simulation Pro, Unity’s new simulation-optimized runtime build, has helped AI2 significantly reduce the training time.

This session first aired at the Unity AI Summit 2021 on November 18, 2021.

Presenter: Roozbeh Mottaghi, Research Manager, PRIOR team, AI2

Download retail sample dataset today

Email Address

First Name

Last Name

Company

Job Title

Industry

Country

I understand that by checking this box, I agree to have Marketing Activities directed to me by Unity

I acknowledge the Unity Privacy Policy

I have read and agree to the Unity Terms of Service

Thank you!

Error - something went wrong!

Return to Home

Search Unity

AI2-THOR: A virtual environment for training home assistant robots

Previous Video

Next Video

AI2-THOR: A virtual environment for training home assistant robots

Previous Video

Next Video

Learn more about Unity

Motec shares how they mix real and synthetic data to enhance neural network training.

Learn how READY Robotics is making robots more accessible to everyone.

Featuring insights from Lockheed Martin Skunk Works and Unity’s aerospace experts, this whitepaper covers how real-time 3D, AI, and extended reality (XR) are transforming the aerospace industry.

During this webinar Unity’s computer vision team provided an overview of computer vision, walked through current real-world data workflows, and explained why companies are moving toward synthetically

Get training-ready, accurately labeled synthetic data.

Unity SVP of AI and Machine Learning Danny Lange shares how data-centric AI development can deliver successful AI applications.

Get a behind-the-scenes look at how Boeing is developing technology capabilities with long-term applications for transforming aircraft maintenance and inspection.

See how to use Algoryx’s AGX Dynamics for Unity and Unity ML-Agents to train an AI in a virtual environment to control a heavy vehicle.

See how to simulate scenarios with Unity SystemGraph and Ouster lidar, including sensor placement and objects of different reflective surfaces.

Get the first look at Unity’s new simulation-optimized runtime build and how it can significantly boost performance for your simulations with headless and multi-GPU distributed rendering.

From modeling robotic control to smart cameras, read how real-time 3D plays a big part in the future of simulation from our SVP of AI & Machine Learning, Danny Lange.

Find out how and why Formant used Unity to build their own robotics simulator and the ways in which it is saving robotics companies a significant amount of time and money.

Go behind the curtain at Volkswagen Group’s Innovation Center California.

Unlock the potential of your computer vision model. Work with Unity's expert team to generate custom synthetic datasets tailored to your specific requirements.

Learn how VW Group uses Unity’s tools to shape the future of mobility.

Get a free synthetic data sample, generated using Unity Computer Vision.

Get a free synthetic data sample, generated using Unity Computer Vision.

Gain insight into how Audere, a digital health nonprofit, is using synthetic data to confront challenges and drive positive social impact.

Every year hospitals spend $2 billion on avoidable hospital-acquired conditions. In this session, we will walk through how Unity's Computer Vision team collaborated with Ouva to use synthetic data for

Join our ROS World 2021 workshop covering robotics simulation using Unity and ROS2