RoboTHOR is an environment within the AI2-THOR framework, designed to develop embodied AI agents. It consists of a series of scenes in simulation with counterparts in the physical world.

Key Features

Sim & Real Pairings

For each physical room, there is a corresponding synthetic equivalent which enables us to study the discrepancies in appearance and control.

Modular

The apartments are built in a modular fashion, drawing from an asset library.

Reconfigurable

The physical environments are built using modular and movable components, allowing us to host scenes with vastly different layouts within a single physical space.

Accessible to All

The simulation environment is open source. Researchers will be able to remotely deploy their models in the real environment.