Functional mapping for human-robot collaborative exploration

The actual structure of a disaster situation is typically unknown. We need to explore it to build up situation awareness, gathering information about structure and the presence of particular objects of interest like victims or threats. If this exploration involves robots, we are dealing with a collaboration between humans and robots, with both as problem-holders. What makes this collaboration complicated is that, from the viewpoint of task-work, there is time-pressure: The exploration needs to be done efficiently, and effectively. From the viewpoint of team-work, the robot needs to perform its tasks together with the human users such that it is apparent to the users why the robot is doing what it is doing. Without that, human users might fail to trust the robot, which can negatively impact overall team performance. In this paper, we present an approach to semantic mapping which aims to address both efficiency (task), and apparency (team). The approach models the environment from a geometrical-functional viewpoint, establishing where the robot needs to be, to be in an optimal position to gather particular information relative to a 3D-landmark in the environment. The approach combines top-down logical and probabilistic inferences about 3D-structure and robot morphology, with bottom-up quantitative maps. The inferences result in vantage positions for information gathering which are optimal in a quantitative sense (effectivity), and which mimic human spatial understanding (apparency). A quantitative evaluation shows that functional mapping leads to significantly better vantage points than a naive approach.