
A UAV Sense And Avoid System Design Method Based On Software
Simulation
Yang Lyu
1
, Quan Pan
1
, Chunhui Zhao
1
, Changbin Yu
2
, Jinwen Hu
1
,
1
School of Automation, Northwestern Polytechnical University
2 College of Engineering & Computer Science, The Australian National University
1
lincoln1587@mail.nwpu.edu.cn;
2
brad.yu@anu.edu.au
Abstract— Sense and Avoid (SAA) has been identified as
one of the crucial technologies for UAV safety flight. The
SAA systematic design is to develop a system with sensor
configuration and SAA algorithms. Normally, the SAA system
integration and test can be both expensive and dangerous with-
out carefully system and algorithms development. In this paper,
we proposed a software based SAA simulation framework for
SAA system configuration and algorithms design. First, the
software structure and functional submodules are described.
Then the SAA function is modelled and optimization based
algorithms integrating environment sensing, collision evaluation
and avoidance. Simulation results and analysis are provided to
validate the simulation software and algorithms. By configuring
the sensors and adjusting the algorithms parameters, the
software based simulation framework can act as a guidance
to the development of practical SAA systems.
I. INTRODUCTION
Unmanned Aerial Vehicle (UAV) has shown great po-
tential in both military and civil applications[1][2][3].The
autonomous operation of UAVs requires that the UAVs must
demonstrate the safety flying capability interacting with other
air traffic, which is also recognized as Sense and Avoid
(SAA). The purpose of sense and avoid is to self-separate
from aircrafts and avoid collisions with other aircrafts[4].
The SAA function compose of two successive sub-functions
which are the environment sensing and collision avoidance
control. The environment sensing sub-function is to reliably
detect and declare potential collision threat by analysing the
measurement from multiple sensors, such as the ATC surveil-
lance system and airborne sensors. The collision avoidance
sub-function is to generate appropriate collision avoidance
path and executed the avoidance maneuver.
The effectiveness of collision avoidance highly depends
on the confidence of sensing results. Researches relating to
various SAA sensors configuration are proposed. Ground
based SAA (GBSAA) system is recognized as the most
accessible way to provide UAVs with the safety capabilities
as GBSAA relies on ground-based radar system to perform
the detection and tracking functions[5], which are already
validated in the civil airspace traffic surveillance system.
Due to the limited coverage of GBSAA system, it is suitable
for terminal area operation such as landing and taking off.
Comparing with GBSAA system, Airborne SAA (ABSAA)
system is preferable for future UAV applications as it is able
to provide environment sensing capability during the whole
UAV operation. The using of airborne T-CAS/ADS-B for
cooperative SAA applications are demonstrated in [6] and
[7] separately. For uncooperative intruders, various sensors
such as airborne radar [8] , Lidar [10], EO/IR [9] and so on
are frequently researched and related systems are developed.
However due to different sensor characteristic and function
limitation, the sensing data provided by a certain sensor
can hardly meet the requirement of SAA application. To
further improve the reliability of the sensing function, multi-
sensor based fusion architecture and algorithms are adopted
in many SAA systems such as [11]. The fusion frameworks
can efficiently take the advantages of various sensors to
obtain complete and reliable sensing information.
Besides the sensors configuration, the algorithms is an-
other key element to fulfill the SAA function. The algorithms
should guarantee that the civil airspace procedure and policy
are complied properly [17]. Normally the algorithms should
execute the environment sensing(aircraft detecting and track-
ing), collision threat evaluation (evaluation, prioritization,
and decision), and avoidance (path planning and maneuver
control) [4]. The design of algorithms and parameter setup
are highly dependent on the sensing mode. In cooperative
ways, such as ADS-B [7] or T-CAS [6], detailed air traf-
fic information are provided directly with high degree of
confidence, then the regulation can be directly modeled in
the collision threat process[18]. Due to the long sensing
range in cooperative ways (ADS-B more than 200nm), self-
separation based global path planning with slight maneuver
are preferred, such as the A* [19] or RRT [20]. While
in uncooperative ways, the environment sensing algorithms
can be more challenging. Data preprocessing, detection,
association and tracking are needed in order to efficient
obtain the status of intruders. Besides the policy and collision
avoid procedure, the sensing quality index of uncooperative
sensors such as statistical error or variance should be taken
into consideration when model the threat evaluation function
[21]. As the sensing range of uncooperative sensors various
from beyong-line-of-sight (BLOS) to very near, different
collision avoid resolution should be adopted, such as self-
separation with slight maneuver, or collision avoidance with
intense maneuver output [23], or reactive maneuver [22].
In addition to that, as for range-only sensors or angle only
sensors, special sensing algorithms should be designed to
obtain complete intruder status [24], or special collision
2016 International Conference on
Unmanned Aircraft Systems (ICUAS)
June 7-10, 2016. Arlington, VA USA
ThBTT4.6
978-1-4673-9333-1/16/$31.00 ©2016 IEEE 572