843 
1. INTRODUCTION 
In the last years UAV (Unmanned Aerial Vehicle) systems 
became very actractive for various commercial, industrial, 
public, scientific and military operations. The tasks include 
pipeline inspection, dam surveillance, photogrammetric survey, 
infrastructure maintenance, inspection of flooded areas, fire 
fighting, terrain monitoring, volcano observations and so on. 
The name UAV denotes all vehicles, which are flying in the air 
with no pilot onboard and with capability for remote controlling 
the aircraft. Within this category, helicopters play an interesting 
role as they are suited for many applications for which fixed- 
wing aircraft show operational difficulties. Indeed such crafts 
offer more flexible manouvers, as they allow for vertical take 
off and landing, hovering and side flight. These impressive 
flying capabilities require a well trained pilot to be fully and 
effectively exploited; moreover the flight range of the piloted 
helicopter is limited to the line-of-sight or the skill of the pilot 
to detect and follow the orientation of the helicopter. Such 
issues have motivated the research and the design for 
autonomous system guidance which could both stabilize and 
also guide the helicopter precisely along a reference path. A 
number of works have been therefore published about the 
development of fully autonomous or partially-autonomous 
helicopters. Small autonomous helicopters have demonstrated to 
be a useful platform for a number of aerial applications such as 
aerial mapping and photography, surveillance (both military 
and civilian) and powerline inspection. 
The goal of this paper is to present the results of the simulation 
software implemented to evaluate in advance the issues related 
with the development of an autonomous control guidance 
system for our low-cost model helicopter, whose main 
ACCURACY ENHANCEMENT OF UNMANNED HELICOPTER POSITIONING 
WITH LOW COST SYSTEM 
U. Coppa a , A. Guamieri b , F. Pirotti b , A. Vettore b 
a Vesuvius Observatory - National Institute of Geophysics and Volcanology - Naples, Italy, coppa@ov.ingv.it 
b CIRGEO - Interdepartment Research Center for Geomatics, University of Padova - Italy, cirgeo@unipd.it 
Commission I, Working Group V/l 
KEY WORDS: UAV, Model helicopter, Kalman filter, MEMS, Autopilot 
ABSTRACT: 
In the last years UAV (Unmanned Aerial Vehicle) systems are become very actractive for various commercial, industrial, public, 
scientific and military operations. The tasks include pipeline inspection, dam surveillance, photogrammetric survey, infrastructure 
maintenance, inspection of flooded areas, fire fighting, terrain monitoring, volcano observations and so on. The impressive flying 
capabilities provided by UAVs require a well trained pilot to be fully and effectively exploited; moreover the flight range of the 
piloted helicopter is limited to the line-of-sight or the skill of the pilot to detect and follow the orientation of the helicopter. Such 
issues have motivated the research and the design for autonomous system guidance which could both stabilize and also guide the 
helicopter precisely along a reference path. The constant growth of research programs and the technological progress in the field of 
navigation systems, as denoted by the production of more and more performing GPS/INS integrated units, allowed a strong cost 
reduction and payload miniaturization, making the design of low cost UAV platforms more feasible and actractive.Small autonomous 
helicopters have demonstrated to be a useful platform for a number of airborne-based applications such as aerial mapping and 
photography, surveillance (both military and civilian), powerline inspection and agricolture monitoring. In this paper we present the 
results of a flight simulation system developed for the setup of the servos which our autonomous guidance system will be based on. 
Building a simulated environment allows, indeed, to evaluate in advance what are the main issues of a complex control system, 
avoiding to damage fragile and expensive instruments as the ones mounted on a model helicopter. 
components have been already described in (Guamieri et al, 
2006). 
2. SYSTEM OVERVIEW 
The primary goal of our model helicopter is to provide the user 
with a top view of the territory without resorting to more 
expensive classical aerial photogramme try. The system is 
designed to collect data for mapping and land monitoring 
purposes working on areas which represent a difficult task for 
already existing ground-based mobile mapping systems. From 
this viewpoint our system can be regarded as a lightweight and 
low cost complementary mapping tool to existing MMS. 
According to project specifications, the model helicopter will be 
used to survey areas of limited extent such as open mines, 
little rivers, cultivated fields, not only to monitor the land 
evolution and local changes in the terrain morphology but also 
to discover illegal uses of land resources. A further example of 
its application deals with the mapping of small water channels 
located along the Venice lagoon, which cannot be accurately 
mapped through classical aerial photogrammetry. These 
channels are of great interest for ongoing biological studies of 
the lagoon ecosystem. 
Several kinds of UAV-helicopters have been so far developed 
for photogrammetric data acquisition and terrain or object 
modeling. For example in (Nagai, 2004) the developed system 
integrates laser scanner and CCD-cameras with GPS/INS data 
for constructing digital surface models. This system uses a 
Subaru helicopter with a payload of 100 kg and diameter of the 
main rotor of 4.8 m. According to the range (3 Km) and altitude 
(2000), the helicopter can be defined as a mini or close range