Guided Exploration of Human Intentions for Human-Robot Interaction
This addresses the challenge of fluid human-robot interaction, particularly in autonomous driving, but appears incremental as it builds on existing probabilistic and learning methods.
The paper tackles the problem of robots inferring human intentions for better interaction by learning a model of adaptive human behavior with intentions as latent variables and embedding it into a probabilistic decision model, with simulation experiments in autonomous driving showing improvements in efficiency and safety.
Robot understanding of human intentions is essential for fluid human-robot interaction. Intentions, however, cannot be directly observed and must be inferred from behaviors. We learn a model of adaptive human behavior conditioned on the intention as a latent variable. We then embed the human behavior model into a principled probabilistic decision model, which enables the robot to (i) explore actively in order to infer human intentions and (ii) choose actions that maximize its performance. Furthermore, the robot learns from the demonstrated actions of human experts to further improve exploration. Preliminary experiments in simulation indicate that our approach, when applied to autonomous driving, improves the efficiency and safety of driving in common interactive driving scenarios.