=Paper=
{{Paper
|id=Vol-2207/IWSM_Mensura_2018_paper_4
|storemode=property
|title=Functional Size Measurement for Energy Needs early Estimation in Autonomous Drones
|pdfUrl=https://ceur-ws.org/Vol-2207/IWSM_Mensura_2018_paper_4.pdf
|volume=Vol-2207
|authors=Hassan Soubra,Toufik Azib
|dblpUrl=https://dblp.org/rec/conf/iwsm/SoubraA18
}}
==Functional Size Measurement for Energy Needs early Estimation in Autonomous Drones==
https://ceur-ws.org/Vol-2207/IWSM_Mensura_2018_paper_4.pdf
Functional Size Measurement for Energy Needs early
Estimation in Autonomous Drones
Hassan Soubra1 and Toufik Azib1
1 ESTACA, Montigny-le-Bretonneux 78180, France
hassan.soubra@estaca.fr
Abstract. A drone, or more formally known as an unmanned aerial vehicle
(UAV), is a type of aircraft that is not designed to carry humans. Where in their
early ver-sions, drones were often controlled remotely by a human operator; they
have re-cently become fully or partially autonomous. Today, there is massive
growth in the use of drones in a wide range of applications: military, aerial pho-
tography, search and rescue operations, in the field of agriculture, to name a few.
The em-bedded software used in autonomous drones must perform in real-time,
allowing the drone to interact with its environment. The Internet of drones (IoD)
was born when drones were interconnected via a network allowing them to coor-
dinate. Functional size measurement (FSM), a powerful tool for managing soft-
ware pro-jects, provides an objective and quantitative base for management de-
cisions. In this paper, we discuss work in progress on COSMIC functional size
for estimat-ing energy consumption of drones as an illustration of the potential
benefit of us-ing FSM in the context of IoD.
Keywords: COSMIC, Unmanned Aerial Vehicle, Internet of Drones (IoD),
FSM, Energy consumption, Pixhawk Autopilot.
1 Introduction
A new era of UAV is emerging today: autonomous drones. These autonomous drones
are no longer controlled by a human operator and they have a growing number of ap-
plications: in defense, transportation, photography, communication, agriculture, disas-
ter mitigation and environment preservation [1]. Real-time embedded software (RTES)
and embedded energy sources are what makes autonomous drones possible. RTES al-
lows the drones to perform certain missions and coordinate in the context of IoD. Nat-
urally, the capacity of the source of energy should be sufficient for the drone to operate
a particular mission. Different types of missions have different power con-sumptions.
Energy constraints are very important in the context of UAVs [2]. COSMIC Software
functional size can be used for a number of purposes in domains where real-time em-
bedded software is omnipresent [3]: to obtain system related technical indicators early
in the design phase (e.g. processor load, network load, etc.). It can also be used to esti-
mate development effort, manage project scope changes, measure productivity, bench-
mark and normalize quality and maintenance ratios. The COSMIC method is accepted
48
H. Soubra, T. Azib
as an International Standard: ISO/IEC 19761 Software Engineering – COSMIC – A
functional size measurement method.
Among the many challenges facing autonomous drones today is energy consumption
[4]. In fact, there are several types of energy sources that can be embedded in an auton-
omous drone.
This paper presents a work in progress on an estimation process, using COSMIC
FSM, to predict energy consumption early in the project specification requirements
phase.
The paper is organized as follows. Section 2 presents a literature review of energy
consumption estimation. Section 3 presents an overview of COSMIC. Section 4 pre-
sents our work in progress including the experimental set-up. Section 5 presents our
conclusions and a discussion of our work in progress.
2 Related Work on Energy Consumption Estimation
This section presents related work on methods or processes tackling energy estimation
in the context of autonomous drones and the RTES domain generally.
In [5] an energy-aware path planning algorithm for UAV is proposed. The algorithm
is based on an energy model derived from real measurements, and it is aimed to mini-
mize energy consumption while satisfying coverage and resolution.
In [6] a study on flight mission planning of battery-operated autonomous drones,
considering autonomous recharging is presented.
In [7], a review various aspects of work on Li-ion battery prognostics and health
monitoring is presented. The study summarizes the techniques, algorithms and models
used for state-of-charge (SOC) estimation, current/voltage estimation, capacity estima-
tion and remaining-useful-life (RUL) prediction.
In [8] an electromotive force (EMF) model is proposed to estimate state of charge
(SOC) of a power battery.
In [9] an architecture of a framework which enables secure end-to-end and reliable
data transfer for heterogeneous mobile terminals by also describing and modeling its
power demand, with the aim of achieving a robust and reliable ubiquitous data transfer
service also minimizing the overall battery consumption in such devices has been pro-
posed.
In [10] a research study evaluates complexity of proper security implementation in
embedded devices possessing different constrains: limited memory, limited processing
power and limited energy consumption. It has identified a relation between additional
instructions processing against energy consumption.
In [11], energy and time related characteristics of widely used cryptographic algo-
rithms for data-sensitive embedded real-time systems are analyzed. In addition, a multi-
dimensional analysis framework that reveals power, speed, and unit energy cost of
cryptographic algorithms is proposed.
The literature review shows no work proposed on energy consumption estimation
early in the design phase.
49
IWSM/Mensura’18, September 18–20, 2018, Beijing, China
3 COSMIC Overview
COSMIC [12] provides a standardized method for measuring the functional size of
software from both Management Information Systems and real-time domains.
COSMIC is considered a second-generation FSM and has been accepted as an Inter-
national Standard (ISO/IEC 19761, Software Engineering – COSMIC – A functional
size measurement method).
In COSMIC, a functional process is a set of data movements representing an elemen-
tary part of FUR for the software being measured. The set is unique within the FUR
and is defined independently of any other functional process. A functional process may
have only one triggering Entry. Each functional process begins processing on receipt
of a data group moved by the triggering Entry data movement of the func-tional pro-
cess. The set of all data movements of a functional process is the set need-ed to meet
its FUR for all possible responses to the triggering Entry. According to COSMIC, soft-
ware functionality is embedded within the functional flows of data groups. Data flows
can be characterized by four distinct types of movement. Two types (Entries (E) and
Exits X)) between the functional user FU and a COSMIC func-tional process allow the
exchange of data with a functional user across a boundary. Two other types (Reads (R)
and Writes (W) between a COSMIC functional process and the persistent storage allow
exchange of data with the persistent storage hard-ware. The measurement result corre-
sponds to the functional size of the FUR of the software measured, and is expressed in
COSMIC Function Points (or CFP).
4 Work in Progress
4.1 Approach
To estimate energy consumption early in the design phase, our research is divided into
five steps:
1. Hardware: Design and assemble an autonomous drone platform. The
drone must be able to support different types of energy sources (e.g. bat-
teries, supercapacitors, etc.).
2. Software: Create different mission profiles for the autonomous drone and
implement them.
3. FSM: Measure the COSMIC functional size of each mission profile.
4. Energy needs: Observe the energy consumption for each mission profile
and for each energy source.
5. Estimation: Use the measurement results in step 3 and 4 to build energy
consumption estimation models for new mission profiles.
Fig.1 shows the steps of our work in progress. Today, only step 1 has been realized.
Steps 2 to 5 are still in progress.
50
H. Soubra, T. Azib
Fig. 1. Towards an energy consumption estimation model.
4.2 Experimental set-up
Step 1 of our research has been realized. An autonomous drone has been assembled at
ESTACA (Fig.2) using the following modules:
DJI F550 Flame Wheel E305 Kit ARF V2 + NAZA M-LITE GPS + Landing
Legs: Chassis and GPS.
Pixhawk PX4 2.4.6: autopilote module.
Turnigy Graphene 2700mAh 4S 65C Lipo Pack w/ XT60U: energy source
FrSky X8R 8/16CH Full Duplex: Telemetry (for safety reasons).
As shown in Fig.2, the drone is designed to support different energy sources. Fig.2
also shows the drone in the langing process.
The drone has been tested using a pilot mission programmed using QGroundControl
[13]. QGroundControl provides flight control and vehicle setup for PX4 powered vehi-
cles. QGroundControl enables mission planning for autonomous flight. QGroundCon-
trol will be used to define the mission profiles in step 2 of our methodology.
Fig. 2. The autonomous drone assembled at ESTACA. The drone in the landing process.
51
IWSM/Mensura’18, September 18–20, 2018, Beijing, China
Fig. 3 shows the mission planned using QGroundControl. The mission that the drone
has been tested with is as follows:
1. Takeoff.
2. Move forward 88 meters.
3. Land.
Fig. 3. The first mission plan of the autonomous drone.
5 Conclusion
Today, there is massive growth in the use of autonomous drones for different appli-
cations. Energy constraints are very important in the context of UAVs.
This paper described work in progress towards energy needs estimation early in the
mission-planning phase for UAV or drones.
The steps of a method for estimation energy needs using FSM has been described.
Then, a drone has been designed and assembled at ESTACA in order to verify the fea-
sibility of the method proposed. We are currently continuing to work on defining new
mission profiles for the drone, before starting both the FSM and the energy con-sump-
tion measurement processes. Later we will map the results and create estimation models
for power consumption for future missions.
Acknowledgements
This work would not be possible without the contribution of two ESTACA stu-dents:
Nathan BAILET and Antoine CASABIANCA. We are grateful to them for designing,
assembling and testing the drone.
52
H. Soubra, T. Azib
References
1. FLOREANO, Dario et WOOD, Robert J. Science, technology and the future of small auton-
omous drones. Nature, 2015, vol. 521, no 7553, p. 460
2. Di Franco, Carmelo, and Giorgio Buttazzo. "Energy-aware coverage path planning of
UAVs." Autonomous Robot Systems and Competitions (ICARSC), 2015 IEEE International
Confer-ence on. IEEE, 2015.
3. Soubra, Hassan, and Alain Abran. "Functional size measurement for the internet of things
(IoT): an example using COSMIC and the arduino open-source platform." Proceedings of
the 27th International Workshop on Software Measurement and 12th International Confer-
ence on Software Process and Product Measurement. ACM, 2017.
4. Choi, Youngmin, and Paul M. Schonfeld. Drone Deliveries Optimization with Battery En-
ergy Constraints. No. 18-02726. 2018.
5. Di Franco, Carmelo, and Giorgio Buttazzo. "Energy-aware coverage path planning of
UAVs." Autonomous Robot Systems and Competitions (ICARSC), 2015 IEEE International
Confer-ence on. IEEE, 2015.
6. TSENG, Chien-Ming, CHAU, Chi-Kin, ELBASSIONI, Khaled, et al. Flight tour planning
with recharging optimization for battery-operated autonomous drones. arXiv preprint
arXiv:1703.10049, 2017.
7. Zhang, Jingliang, and Jay Lee. "A review on prognostics and health monitoring of Li-ion
battery." Journal of Power Sources 196.15 (2011): 6007-6014.
8. WU, Tie-zhou et HU, Li-ping. Study of SOC estimation algorithm of power battery based
on UKF. Power Electronics, 2014, vol. 48, no 4, p. 23-26.
9. Castiglione, Arcangelo, et al. "An energy-aware framework for reliable and secure end-to-
end ubiquitous data communications." Intelligent Networking and Collaborative Systems
(IN-CoS), 2013 5th International Conference on. IEEE, 2013.
10. Krumins, Ingus. "Security Impact on Embedded System Performance." (2014).
11. Jiang, Wei, et al. "Measurement-based research on cryptographic algorithms for em-bedded
real-time systems." Journal of Systems Architecture 59.10 (2013): 1394-1404.
12. COSMIC Measurement Manual Version 4.0.1, Common Software Measurement In-terna-
tional Consortium, 2015. http://cosmic-sizing.org
13. http://qgroundcontrol.com/
53