In fact, one would
like the sensor nodes with ???interesting??? sensor information to be allocated
more bandwidth than others. Defining what constitutes ???interesting??? sensor
information is entirely dependent on the sensor network task. For example, it
could be the detection of a single event that produces sensor readings above
a certain level, or it could be continuous high bandwidth video feed from a
camera that has captured an event of interest.
A number of fundamental issues arise in the coordination of multiple mobile
sensor nodes used to sample a space-time field distribution, subject to a set
of constraints. First, one would like to maximize the sensor information acquired
through planning and control of the vehicles??™ paths and their sampling
locations. At the same time, communications pathways are utilized and may
and also energy expenditure must be monitored in order to complete tasks
and plan reachable deployments.
These elements of distributed sensor coordination with mobile nodes are
depicted in Figure 1 for an underwater deployment scenario. Several prior papers
from Popa and Stephanou [39], Popa and Sanderson [40], and Popa and
Lewis [41] describe basic aspects of the deployment scenario using potential
fields for communication and navigation, and information measures for determining
the location of optimal sampling locations. This chapter combines the
following fundamental aspects of the deployment problem:
??? Combined navigation and estimation of robot nodes.
Pages:
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99