Commonsense Reasoning for Identifying and Understanding the Implicit Need of Help and Synthesizing Assistive Actions

Human-Robot Interaction (HRI) is an emerging subfield of service robotics.\nWhile most existing approaches rely on explicit signals (i.e. voice, gesture)\nto engage, current literature is lacking solutions to address implicit user\nneeds. In this paper, we present an architecture to (a) detect user implicit\nneed of help and (b) generate a set of assistive actions without prior\nlearning. Task (a) will be performed using state-of-the-art solutions for Scene\nGraph Generation coupled to the use of commonsense knowledge; whereas, task (b)\nwill be performed using additional commonsense knowledge as well as a sentiment\nanalysis on graph structure. Finally, we propose an evaluation of our solution\nusing established benchmarks (e.g. ActionGenome dataset) along with human\nexperiments. The main motivation of our approach is the embedding of the\nperception-decision-action loop in a single architecture.\n