A
bar graph records the energy consumed by each node. The simulation results
show that:
??? Rerouting, forced after a limited number of iterations (No = 10), significantly
alters the network routes if the nodes are moving.
??? The nodes that are not involved with goal attainment or routing of
sensory data from the robots near the goal will drift towards the terminating
node (node 20 at located at (0,0)), unless they are assigned di?®erent sampling
missions.
??? The damping coe?±cient a?®ects goal attainment as well as the individual
energy consumption per node. Small damping values will lead to high energy
consumption rates, while high damping values will prevent nodes from
reaching the goals. A trade-o?® value of damping was found experimentally to
provide the final configuration in Figure 8(b).
58
Chapter 2 Algorithms for Robotic Deployment of WSN
(a)
(b)
Fig. 8. Motion of a 20 node network after 100 iterations, under a potential field
that optimizes the data rate through the network and repositions to sample at goal
location (7.8).
The algorithms described in this chapter, in particular the EKF (AS-3) and
potential fields (POT-1), provide a strategy for systematic exploration of unknown
environments using MWSN. The robot nodes are directed to sample
at locations that most reduce the uncertainty in our knowledge of the field
to maintain a network with optimized data rates. Future work includes experimental
work to validate the proposed algorithms on nonlinear and time-
59
distribution.
Pages:
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125