Aerospace and Electronic Systems - August 2018 - 58

Feature Article:

DOI. No. 10.1109/MAES.2018.170145

UAV System for Photovoltaic Plant Inspection
Pia Addabbo, Antonio Angrisano, Mario Luca Bernardi, Giustino Fortunato University,
Benevento, Italy
Graziano Gagliarde, Alberto Mennella, TopView S.r.l., Caserta, Italy
Marco Nisi, Sistematica S.p.a., Terni, Italy
Silvia Liberata Ullo, University of Sannio, Benevento, Italy

INTRODUCTION
In the last two decades, growing attention on climate issues has
caused the worldwide increase of Photovoltaic (PV) plant production and installation, and the consequent promotion of clean
energy policies, with large amounts of incentives and funding
made available in the specific sector by Governments and the European Economic Community itself. Increasing PV distribution
and installation has to ask for efficient and low-cost methods for
inspection to monitor functionality and guarantee performance.
A big concern of PV plant owners is to rely on efficient maintenance procedures. Recognizing degradation and defects of PV
cells is a very important issue to allow immediate intervention
and substitution of modules to avoid output power losses and performance degradation.
New systems, based on the use of Unmanned Aerial Vehicles
(UAVs), have been thought to substitute human workers inspecting the PV plants to reduce maintenance costs and intervention
times. In recent years, extended research has been conducted on
automated remotely controlled systems for PV fields inspection,
realized through the employment of UAVs to substitute human
intervention on the PVs and to reduce human activity of data collection and postprocessing [1]. UAVs allow realizing the detection
of defects in PV plants thanks to the fusion of computer vision
algorithms and high accuracy Global Navigation Satellite System
(GNSS) positioning techniques. Particularly, the use of UAVs in
PV plants monitoring together with the GNSS positioning techniques allows getting reliable information for the diagnosis of PV
systems being able to detect and tag anomalies on the defective
panel. The advantages of UAVs employment are several: low
Authors' current addresses: P. Addabbo, A. Angrisano, M.
L. Bernardi, Giustino Fortunato University, Viale Raffaele
Delcogliano, 12, Benevento (BN), 82100, Italy; G. Gagliarde,
A. Mennella, TopView S.r.l., Via Alessandro Pertini, 25D, San
Nicola La Strada, 81020 (CE), Italy; M. Nisi, Sistematica S.p.a.,
Via Bramante, 43, 05100, Terni (TR), Italy; Silvia Liberata Ullo,
University of Sannio, Piazza Roma 21, Benevento, (BN) 82100
Italy, E-mail: (p.addabbo@unifortunato.eu).
Manuscript received July 21, 2017, revised October 10, 2017,
and ready for publication November 17, 2017.
Review handled by W. Walsh.
0885/8985/18/$26.00 © 2018 IEEE
58

costs, automated detection, large area coverage, accuracy in defect detection through the combination of different sensors, and
cameras [2].
Furthermore, information acquired by UAVs can be transmitted to a remote console where subsequent interventions can
be activated by the detected anomalies. Maintenance companies
equipped with suitable tools can in this way conduct an a priori
analysis of the PV plant, which can be then compared with an
a posteriori PV plant performance evaluation based on the final
energy produced.
As already underlined above, within the employment of
UAVs for PV plant monitoring, a big step forward has been represented by the incorporation of GNSS information to solve problems related to accurate panel positioning and geo-referencing
over the PV plant. In this field of investigation, a low-cost GNSS
Real Time Kinematic (RTK) receiver has been proposed to be
mounted on board the UAV system for locating the defective
cells [3], [4].
The goal of the authors of this article is to describe some critical aspects of direct geo-referencing, image acquisition, panel recognition, and faulty module detection in the specific case of PV
plant inspection. Besides this, they will present interesting results
come out of a validation campaign performed taking into account
different datasets.

THE REMOTELY PILOTED AIRCRAFT SYSTEM (RPAS)
The model of proposed system architecture is represented in
Figure 1, where a UAV performs a mission flying over a PV
field to collect optical and thermal images of solar panels. The
data gathered by the UAV are processed through a computer
vision algorithm running on-board the vehicle in tight synergy
with a geo-software module, capable of tagging thermographic
images using fixed centimeter-level positions provided by the
commercial U-blox M8 series receivers. The accurate positioning provided by GNSS signals enables the automation of the
entire process allowing to correctly geo-referencing the defective panel inspected by the thermal camera on-board the UAV.
Finally, this information is provided to the remote service center
in charge of the defect identification and PV plant management.
We refer to the overall architecture as Easy-PV Remotely Piloted
Aircraft System (RPAS) Architecture, since the work presented

IEEE A&E SYSTEMS MAGAZINE

AUGUST 2018

Aerospace and Electronic Systems - August 2018

Table of Contents for the Digital Edition of Aerospace and Electronic Systems - August 2018