droidlet: modular, heterogenous, multi-modal agents
This work addresses the challenge of creating multi-modal agents for robotics and AI, though it appears incremental as it integrates existing components rather than proposing a fundamentally new approach.
The paper tackles the problem of building integrated agents that combine perception, language, and action by introducing droidlet, a modular and heterogeneous architecture, which enables exploitation of large-scale datasets and robotics heuristics to facilitate learning from real-world interactions.
In recent years, there have been significant advances in building end-to-end Machine Learning (ML) systems that learn at scale. But most of these systems are: (a) isolated (perception, speech, or language only); (b) trained on static datasets. On the other hand, in the field of robotics, large-scale learning has always been difficult. Supervision is hard to gather and real world physical interactions are expensive. In this work we introduce and open-source droidlet, a modular, heterogeneous agent architecture and platform. It allows us to exploit both large-scale static datasets in perception and language and sophisticated heuristics often used in robotics; and provides tools for interactive annotation. Furthermore, it brings together perception, language and action onto one platform, providing a path towards agents that learn from the richness of real world interactions.