Schedule-based Motion Prediction for Human-Centric Autonomous Observation

As robot autonomy improves, robots are increasingly being considered in the role of autonomous observation systems — free-flying cameras capable of actively tracking human activity within some predefined area of interest. In this work, we formulate the autonomous observation problem through multi-objective optimization, presenting a novel SemiMDP formulation of the autonomous observation problem that maximizes observation rewards while accounting for both humanand robot-centric costs. We demonstrate how the problem can be solved for a known human trajectory using Constrained MDPs, and extend the approach to incorporate human motion prediction based on noisy rationality models, defined over a set of goals extracted from a task schedule.