Toward robots that learn to summarize their actions in natural language: a set of tasks

Chad DeChant, Daniel Bauer

Keywords: RoboNLP, Explainable robotics, Long horizon tasks, Summarization

TL;DR: Robots should learn how to succinctly summarize their past actions in natural language and receive feedback based on that summary.

Abstract: Robots should be able to report in natural language what they have done. They should provide concise summaries, respond to questions about them, and be able to learn from the natural language responses they receive to their summaries. We propose that developing the capabilities for robots to summarize their actions is a new and necessary challenge which should be taken up by the robotic learning community. We propose an initial framework for robot action summarization, presented as a set of tasks which can serve as a target for research and a measure of progress.