Unmanned Aerial Vehicles (UAV) require for low-level
flights activities, sensors that can effectively operate in areas where natural
and man-made obstacles are present. The sensors must be reliable even when
meteorological conditions do not make possible for the human eye to
differentiate between obstacles or to even identify the obstacle, being this
the ultimate purpose of the sensor. The sensor(s) to be used under such
circumstances have to be able to detect all types of hazardous obstacles,
including topographic features, vegetation, buildings, poles/masts, towers,
cables and transmission
lines. It must count with satisfactory technological readiness
levels. A high range and bearing resolution will be indispensable and high
minimum detection range will be adequate for the platform velocity and dynamic
performances.
A sensor that can meet such requirement could be the LIDAR
Obstacle Warning and Avoidance System (LOWAS), which has become a mature
technology that small-to- medium size UAV’s resource to when their operations
require close proximity to the ground. LOWAS is of preference when
warning and avoidance of obstacles is necessary. It is important to mention that LIDAR is an
exteroceptive sensor. When we talk about
sensors, it is essential to differentiate between proprioceptive and
exteroceptive sensors. Proprioceptive sensors are those that measure internal state of
a system, it could be battery level or even the performance of gyroscopes. On
the other hand, exteroceptive sensors measure the external state of systems,
this could be temperature, meteorological conditions, or detection of objects
between others.
LOAWS is suitable for this mission because it able to
detect obstacles placed in the UAV route of flight or nearby the operation
area. LOWAS has the capability to classify and the same time prioritize the
obstacles that were detected. It can also provide visual and aural warnings and
information to the crew (Ramasamy, 2016). Therefore, LOWAS counts with three key
algorithms: prediction of the future platform trajectory; calculation of the
potential collisions with the detected obstacles; and generation of a set of
optimal avoidance trajectories (in case a risk of collision is determined)
(Ramasamy,2016). These algorithms are the key to how LOWAS will perform and
also help to kind of predict how the UAV will behave in circumstances not
favorable for the flight.
LOWAS goes beyond meeting the requirements. It counts
now with a cognitive remote pilot–aircraft interface that is being developed to
dynamically assist remote pilots based on their physiological and cognitive
states detected in real-time (Ramasamy, 2016). The purpose of this is to create
a cooperation between pilots and the systems. I definitely recommend this
system for warning and avoidance of obstacles, specially after working on the
human and machine cooperation that will benefit the result of the mission to be
carried by the UAV.
LOWAS also offers three levels of alerts. They are:
warning, caution, and advisory. These alerts help the remote pilot to prepare
or be aware of the upcoming obstacles that will have to be evaded. These alerts
come in the form of digital voice outputs and tone or they can also be
displayed like is done in the chart presented below.
https://www-sciencedirect-com.ezproxy.libproxy.db.erau.edu/science/article/pii/S1270963816301900
References
Kohanbash, D. (2014). Sensor types
(modalities) for robots to experience the world. Robots for Roboticits.
Retrieved from
http://robotsforroboticists.com/sensor-modalities/
Ramasamy, S. (2016). LIDAR obstacle
warning and avoidance system for unmanned aerial vehicle sense-and-avoid. Science
Direct. Retrieved from
https://www-sciencedirect-com.ezproxy.libproxy.db.erau.edu/science/article/pii/S1270963816301900
Newcomb, D. (2017). Velodyne LIDAR
for Safety's Sake. Cthreegroup. Retrieved from
http://cthreegroup.com/velodyne-lidar-for-safetys-sake/