I've been exploring the idea of using a multirotor 'drone' to improve the situational awareness of the command and operations officers on the scene of events like house fires.

Listen to the radio traffic from a typical residential house fire, and you'll find a non-trivial amount of the traffic is simply people trying to explain what they're seeing from their vantage point (e.g. changing smoke/fire conditions on the rear of the structure, imminent collapse of a portion of the roof, etc).

A multirotor with a gimbeled camera orbiting a few dozen meters above the scene could dramatically improve the chief officer's visibility of the event, increasing safety, and allowing for better tactical decisions.

This article lays out some of the largest issues I've run into.