J. D. Hernández, M. Moll, and L. E. Kavraki, “Lazy Evaluation of Goal Specifications Guided by Motion Planning,” in Proceedings of the IEEE International Conference on Robotics and Automation, 2019, pp. 944–950.
Nowadays robotic systems are expected to share workspaces and collaborate with humans. In such collaborative environments, an important challenge is to ground or establish the correct semantic interpretation of a human request. Once such an interpretation is available, the request must be translated into robot motion commands in order to complete the desired task. Nonetheless, there are some cases in which a human request cannot be grounded to a unique interpretation, thus leading to an ambiguous request. A simple example could be to ask a robot to “put a cup on the table,” where multiple cups are available. In order to deal with this kind of ambiguous request, and therefore, to make the human-robot interaction easy and as seamless as possible, we propose a delayed or lazy variable grounding. Our approach uses a motion planner, which considers and determines the feasibility of the different valid groundings by representing them with goal regions. This new approach also includes a reward-penalty strategy, which attempts to prioritize those goal regions that are more promising to provide a final solution. We validate our approach by solving requests with multiple valid alternatives in both simulation and real-world experiments.
PDF preprint: http://kavrakilab.org/publications/hernandez2019lazy-evaluation-of-goal-specifications.pdf