Information Technology Reference
In-Depth Information
at sea are caused by collisions [5]. In addition, it has been found that human error is a
major contributing factor to those incidents. This could be due to an ever-decreasing
number of crew members adding more responsibilities per person. In the first three
months of 2012 alone, a number of worldwide maritime incidents occur including the
grounding of Costa Concordia and the collision between a cargo vessel and a
passenger ferry in Belfast Lough [6]. It appears from initial investigations that most, if
not all, of these incidents occurred as a result of human error and were therefore
preventable. This provides a sound motivation in developing autonomous collision
detection and avoidance systems for both manned and unmanned vehicles. For
uninhabited craft in particular, this cannot be overlooked as their collision with other
manned ships could endanger human lives. Hence a human operator is always
required to maintain a constant lookout for any potential obstacles which is costly.
UUVs usually operate underwater and therefore do not pose a direct threat to the
ambient surface traffic. Even so, there is a great deal of research in sonar-based ODA
strategies developed for UUVs as compared to USVs which is primarily due to recent
surge in underwater exploration.
The autonomy of an unmanned vehicle depends on the design of a reliable NGC
system. Of these, the navigation system acquires and processes data so that the
guidance system can generate appropriate trajectories to be followed by the vehicle. A
well-designed control system or autopilot tracks the reference commands as closely as
possible. The most common form of guidance law used in unmanned vehicles (marine
or airborne) is the line-of-sight (LOS) guidance [7]. In this method, a LOS angle is
formed and followed between the vehicle's current position and the target location.
Several other guidance laws are also based on this methodology. In the absence of a
collision detection system, the unmanned vehicle will follow the reference path
regardless of the presence of any intermediate objects. This could lead to catastrophe
as the vehicle may run into an obstacle thus damaging the on-board components and
in the worst case, sink it. The presence of an on-board ODA system is thus extremely
important for the vehicle to become self-sufficient.
Unfortunately, the marine research community has been mainly focussed on
advanced navigation and control systems design and little attention is paid to the area
of collision avoidance. The usual way adopted to work around this problem is by
human intervention [3] through a radio control channel or a wireless link thus adding
to the operating cost in the form of a manned support boat. As a consequence, the
usability and extent of the vessel is severely constrained. In [1], it is argued that
although USVs provide an excellent platform for fast experimentation and
development of guidance and control algorithms, their use is limited due to the lack of
a reliable ODA system.
Motivated by the need of designing a reliable ODA system for USVs, this paper
describes such an experimental platform that can improve the USVs' efficiency and
safety. The proposed system employs a high definition video camera and a laser
sensor mounted on a pan and tilt device to provide the NGC system of the USVs with
a visual reference of the surrounding area. In addition, the proposed system is
integrated with a risk assessment module which identifies any potential threat and
take/recommend suitable actions in order to alleviate it.
Search WWH ::
Custom Search