SPIN: Sensor Placement for Indoor Navigation of Drones
Alireza Famili, Angelos Stavrou, Haining Wang, Jung‐Min Park
Abstract
Drones must operate permanently or temporarily autonomously for many applications. They rely on access to location services upon obtaining navigation commands and continually thereafter to enable completely or partially autonomous flight. Global Positioning System (GPS) is not always available, can be spoofed or jammed, and is particularly error-prone in indoor and underground settings. In this article, we present SPIN (Sensor Placement for Indoor Navigation of Drones), a sensor-assisted ranging system for drones that performs in GPS-deficient situations. SPIN employs a novel optimization technique for the deployment of indoor sensors in three-dimensional spaces. SPIN utilizes advancements in Evolutionary Algorithms to compute the smallest number of sensors and their ideal placement in order to minimize deployment costs and localization errors. This challenge is classified as NP-Hard and belongs to the class of Mixed Integer Programming (MIP) problems. SPIN can provide numerous optimal sensor configurations that decrease the number of deployed sensors, enabling autonomous navigation of drones in inside environments at a low cost.