Autonomous reuse of motor exploration trajectories

摘要

We present an algorithm for transferring exploration strategies between tasks that share a common motor space in the context of lifelong autonomous learning in robotics. The algorithm does not transfer observations, or make assumptions about how the learning is conducted. Instead, only selected motor commands are transferred between tasks, chosen autonomously according to an empirical measure of learning progress. We show that on a wide variety of variations from a source task, such as changing the object the robot is interacting with or altering the morphology of the robot, this simple and flexible transfer method increases early performance significantly in the new task. We also provide examples of situations where the transfer is not helpful.