RoboCat: A Self-Improving Generalist Agent for Robotic Manipulation
This addresses the challenge of heterogeneous robotic learning for broader automation, though it builds incrementally on foundation model concepts.
The authors tackled the problem of enabling robots to quickly learn new skills and adapt to different embodiments by proposing RoboCat, a generalist agent for robotic manipulation that can generalize to new tasks and robots both zero-shot and with only 100-1000 examples, showing improved efficiency as training data diversifies.
The ability to leverage heterogeneous robotic experience from different robots and tasks to quickly master novel skills and embodiments has the potential to transform robot learning. Inspired by recent advances in foundation models for vision and language, we propose a multi-embodiment, multi-task generalist agent for robotic manipulation. This agent, named RoboCat, is a visual goal-conditioned decision transformer capable of consuming action-labelled visual experience. This data spans a large repertoire of motor control skills from simulated and real robotic arms with varying sets of observations and actions. With RoboCat, we demonstrate the ability to generalise to new tasks and robots, both zero-shot as well as through adaptation using only 100-1000 examples for the target task. We also show how a trained model itself can be used to generate data for subsequent training iterations, thus providing a basic building block for an autonomous improvement loop. We investigate the agent's capabilities, with large-scale evaluations both in simulation and on three different real robot embodiments. We find that as we grow and diversify its training data, RoboCat not only shows signs of cross-task transfer, but also becomes more efficient at adapting to new tasks.