Aerial View Based Guidance System

Abstract

With the limited amount of resources and roads available, traffic congestion seems to be the current leading problem for urban areas. Reliance on the traditional vehicular navigation systems seems to a huge contributing factor, which simply cannot meet the present demand for better traffic management. While the existent navigation system relies on the Global Navigation Satellite System (GNSS) and predefined historic navigation data, a real time optimal pathfinding navigation system has been lacking. We have proposed a close to a real-time system that calculates an optimal path from start to destination using floodfill algorithm and provides guidance to ground vehicles. A camera hovering over the top provides a bird’s eye view and via the Hyper Text Transfer Protocol (HTTP) on top of 802.11n protocol, transfers aerial view from the airborne device to Unmanned Ground Vehicle (UGV) where the actual computer vision processing is done in a local processor. The use of floodfill algorithm although requires an image that contains the whole maze but makes the system far more efficient. This makes it real time and efficient which may be further improved and integrated with actual traffic scenarios.