Datasets:
DexHeim
/
plaything

Dataset card Data Studio Files
xet
 Community
2
text
stringlengths
5
149
the first BVLOS application test was successfully
completed for inspecting power lines, by Delair -Tech who
flew a UAV for over 30 miles using a 3G wireless network
to guide the drone (Fig. 5). The company were granted a
specific flight corridor in which conducted the test flight.
Although the flight was conduct via autopilot, two pilots
were present at the start and two pilots were present at thelanding site. Using the 3G network allowed for real -time
communication from any distance as long as there was 3G
coverage [29].
2017 saw Israel step up its involvement in the BVLOS
UAS sector and has recently granted full permission for
BVLOS flights. The award was given by the Civil Aviation
Authority of Israel (CAAI), to Airobotics who have
developed a UAV that can achieve and execute missions
safely without the aid of a pilot (Fig. 6). On -board is
Airobotics own computer software which also incorporates
Artificial Intelligence (AI) which is programmed to make
decisions and execute actions that are usually performed by
a human pilot [30]. The BVLOS platform is based on three
parts. The first component was the UAV, named
“Optimus”, which is a drone that is capable of flying thirty -
minute missions whilst being equipped with a one -kilogram
payload. The second component is a completely unmanned,
automated airbase from which the UAV can be launchedfrom and also lands on. The third and final piece and the
most important is the software and the AI software, which
enables operators to use the software easily and manage
missions just with one click [31].
This may sound as though the problem that was once
facing the drone industry has been solved, however the use
of AI itself presents problems of its own. AI itself is a
controversial topic for both industry and politics. Keeping
AI, or narrow AI, which is purely focused on autonomous
drone navigation, at a level that is beneficial for the good of
mankind is hotly debated and motivates many research
areas although flight safety is always the key element to be considered. The goal for most research is to create general
AI that far outgrows the relative conformity if narrow AI
[31]. Currently the AI that we are living with are neural
networks and machine learning algorithms that are used in
everyday common devices [31]. A main concern is for ourown preservation as it is feared that AI could at some point
become intelligent enough to replace humans and become
part of a technological singularity. Indeed this is a situation
some may even welcome as they see AI as a panacea for
civilisation [32] even though AI might outperform humans
at every cognitive task and risks rendering us obsolete [31].
AI will undoubtedly have a major impact on people’s lives,
but the benefits are undeniable.
VI. UAS S ENSOR FUSION
Sensory communication with any UAS is paramount to
operating beyond the pilots’ field of vision. A study into
potential sensory appliances has been presented by Zhahir
et al. [34] and looked at the current development of UAV
sense and avoid systems. One possible theory presented as
a possible way to achieve safe BVLOS applications, was to
equip an UAV with electro -optical sensors combined with
radar and infrared sensory capabilities. However, badweather or overcast and cloudy conditions could affect the
performance at object and hazard identification as the
sensors rely on good light to be able to work at full
capacity. Another possibility discussed was ‘sensor fusion’,
enabling multiple sensory tasks on a UAV platform to be
performed simultaneously to enhance hazard detection and
minimise flight risks. Ramasamy et al. [35] details a
successful test using sensor fusion. The research
successfully produced a simulated study of sensor fusion
which combines natural inspired sensors and non -
cooperative sensors. The algorithm that was used by the
researchers to achieve this is known as track to track fusion
and is based on Boolean decision logic data structure that
can evaluate and solve issues such as limited information of
the environment or partial loss of transmitted information.
The UAV platform is essential for military applications
and its role in conflict and congested zones cannot beunderestimated. Small unmanned aircraft embark upon
intelligence gathering missions via reconnaissance and
surveillance and BVLOS is an essential component. One
example of the most state of the Art recognisance UAV that
has been developed for BVLOS missions is the military’s
Black Hornet Nano [36] (Fig. 7).
Fig. 5. Delair Tech BVLOS for power line inspection applications [29].
Fig. 6. Airobotics autonomous BVLOS system [33].This small, compact UAV is fitted with multi -sensory
capability and has an integrated video stream data ink
where images can be viewed in real time. Part of the UAS
ability to perform BVLOS missions is that has the capacity
to be programmed with a pre -planned route using GPS and
can also be used in FPV with a maximum range of 1.5km
distance between the operator and the UAV [36]. In
comparison larger military UAS rely on satellite
communication to operate rather than a direct flowing radio
link. The military have further developed a system that can
detect other aircraft so that they may be targeted by air to
air missiles. The system known as Active Electronically
Scanned Array (AESA) radar, which is also known as an
active phase array radar, which operates by emitting a pulse
signal from a transmitter that in turn is received by an
onboard antenna that receives amplified echoes of any
objects in the vicinity.
Texas Instruments, in 2016 discussed the benefits ofdeveloping a low latency design for video enabled drones
[37]. One of the main features that a piloted UAV requires
would be an onboard camera as well as a range of other
SAA instruments. The needs of the camera are directly
linked to the needs of the UAV. A low power consumption
rate is necessary so that it does not impact on the UAV’s