Towards learning through robotic interaction alone: the joint guided search task
This work addresses the challenge of enabling robots to learn through interaction alone, specifically for human-robot collaboration, but it appears incremental as it builds on existing concepts of attention and guided search.
The paper tackles the problem of achieving human-robot joint attention by proposing a biologically inspired approach that uses attention systems to constrain relevance and exchange nonverbal behavior during visual guided search, with the goal of sharing attention through synthetic foreground maps and human biological attention.
This work proposes a biologically inspired approach that focuses on attention systems that are able to inhibit or constrain what is relevant at any one moment. We propose a radically new approach to making progress in human-robot joint attention called "the joint guided search task". Visual guided search is the activity of the eye as it saccades from position to position recognizing objects in each fixation location until the target object is found. Our research focuses on the exchange of nonverbal behavior toward changing the fixation location while also performing object recognition. Our main goal is a very ambitious goal of sharing attention through probing synthetic foreground maps (i.e. what is being considered by the robotic agent) and the biological attention system of the human.