ALAN: Autonomously Exploring Robotic Agents in the Real World
This work addresses the challenge of autonomous robotic learning in unstructured real-world settings for applications like household assistance, though it appears incremental by building on existing exploration methods.
The paper tackles the problem of scaling autonomous robotic exploration to real-world environments with minimal human supervision, proposing ALAN which uses environment change measurement to enable efficient skill discovery and task execution, achieving tasks in real-world kitchen settings with reduced training and interaction time.
Robotic agents that operate autonomously in the real world need to continuously explore their environment and learn from the data collected, with minimal human supervision. While it is possible to build agents that can learn in such a manner without supervision, current methods struggle to scale to the real world. Thus, we propose ALAN, an autonomously exploring robotic agent, that can perform tasks in the real world with little training and interaction time. This is enabled by measuring environment change, which reflects object movement and ignores changes in the robot position. We use this metric directly as an environment-centric signal, and also maximize the uncertainty of predicted environment change, which provides agent-centric exploration signal. We evaluate our approach on two different real-world play kitchen settings, enabling a robot to efficiently explore and discover manipulation skills, and perform tasks specified via goal images. Website at https://robo-explorer.github.io/