Page 118 - Innovations in Intelligent Machines

108 I.K. Nikolos et al.
The path planner was tested in a simulated environment, and the sim-
ulation results demonstrated the ability of the algorithm to produce near
optimal paths without violating the imposed constraints. The adoption of
the B-Spline formulation provided the ability to keep the number of design
variables as small as possible, and at the same time produce reasonable and
smooth paths, without abrupt turns.
Future work will be focused on the development of an on-line path planner
for coordination of a team of UAVs. The methodology that will be used for
the planner will be a combination of this work and the work presented in [23],
where an on-line path planner for a single UAV was presented, which is able
to gradually produce B-spline paths in an unknown 3D environment.

7.1 Trends and challenges
The military market for UAVs has demonstrated a strong positive trend dur-
ing the past decade, with the corresponding commercial market showing a
similar behavior, although not so strong [1]. This trend is expected to con-
tinue, as the technology provides new solutions to the problems of autonomous
navigation of UAVs, and new ideas are emerging about the roles and tasks
that can be assigned to UAVs. The trend is supported by the large num-
ber of research teams that are working in the field. In particular, the field of
UAV cooperation gained increased interest during the past years due to the
advantages of using a team of UAVs instead of a single one to accomplish a
complicated mission. However, because of the youth of the field, the research
has been “scenario” oriented, and a rigorous formalism is still missing. Work
is still needed in the direction of clarifying: a) the assumptions about the sys-
tems of cooperating UAVs, b) the terminology used to describe the various
problems under consideration, c) the different categories of working scenarios,
d) the objectives and constraints for each problem, e) the “best practices”
that can be adopted for specific problems or sub-problems.
The research in the field of cooperating UAVs is highly interdisciplinary,
and knowledge from different science and technology fields is needed, even
from the beginning of the formulation of the problem under consideration. It
would be highly beneficially for the researchers, especially for those working
in more theoretical fields, to collaborate with possible users of the systems
or methodologies under development. Some of the problems, relevant to UAV
cooperation, that gain high interest are: a) the assignment of multiple tasks
to a team of UAVs, b) the path planning problem of cooperating UAVs in
the presence of various cooperation and mission constraints, c) information
exchange and data fusion between the cooperating vehicles, d) cooperative
sensing of targets, e) cooperative sensing of the environment, f) centralized
versus distributed coordination methodologies, especially for cases with com-
munication problems between the vehicles, g) on-line mission rescheduling and
task reassignment for fault-tolerant systems.