“…For the learning aspect, there has been a growing interest in learning (inferring) dense-time temporal logic formulae from system trajectories [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14]. Such temporal logic formulae have been used as high-level knowledge or specifications in many applications in robotics [15], [16], [17], [18], power systems [19], [20], [21], [22], smart buildings [23], [24], agriculture [25], etc. We infer dense-time temporal logic formulae from the temperature and humidity sensor data as dense-time temporal logics can effectively capture the time-related features in the transient period when people enter a room.…”