Biomedical Engineering Reference
In-Depth Information
plication for sensor systems in supporting humans and extract the human sensing
capability. These operator-driven exterior sensor devices complemented with ad-
ditional external remote information sources, may even be acting and integrating
with mobile robot platforms performed in the search area.
The human-based sensor system is aimed to remotely interact with the envi-
ronment in order to achieve the best possible information flow from a distant dis-
aster area to the rescue workers. Then, based on the data received, it may perform
substantially safer search actions, and make more effective decisions than without
the complementary artificial information sources, e.g., Das (2003).
Consider a purely hypothetical case of an accident, when suddenly a disaster
situation occuring in a search and rescue mission, in which a heterogeneous
response team (humans, dogs, robots — movable and static sensors) enters an
unknown building. The different agents, via a network are connected to each other
and will each receive unique time-dependent information. Since different agents
have different specific tasks and conditions, then there is a need to provide each
agent with specific real-time amount of essential information. The communica-
tion between the agents is necessary for tasking the network, sharing information,
and for the overall situational management. The available performance of each
agent will be optimised in the sense that a camera will provide and share visual
information and an olfaction device distributing smell properties to the team. The
management, e.g., the team leader will then have access to all information and
based on the person's skill and earlier knowledge be able to extract essential in-
formation for decision-making purpose. Also, the tactical methodology has been
agreed upon. When they arrive at the scene, the tactical operation starts, and a
general strategic structure is considered to be valid. The strategic path is contin-
uously updated and refined as the procedure goes on. The advanced information
needed by the manager and the valid information provided to the different agents
is of most importance both in time and amount of data. In fact if human agents
are involved, then every agent in the team — artificial, human or dogs — on the
scene should be following the same structural level of semantic and organisation
when interacting in the network. The danger assessment strategy is estimated and
the different agents' performance will be validated. If there is a risk for humans
entering the disaster area, the artificial mobile and static sensors specific benefits
canbeused. Thenthedifferentagents,orsensing devices, can start producing data
that, e.g., may be used to establish a generic map of the scene. The map is a knowl-
edge estimation of gained information that is continuously updated and agents of
a characteristic static behaviour are placed in strategic places. The communication
is frequent and the agents are updated with essential activities about, for exam-
ple, injured people and events that may affect the rescue and search strategy. This
procedure continues until the coarse scene has been searched, and new tasks will
be followed. The hierarchy in the network may already be settled when humans
strategically manage the actions, the dogs are extremely useful to find people in
the debris but their communication may be limited at a distance and in a chaotic
environment. However, artificial sensor systems may be the solution in contami-
Search WWH ::




Custom Search