Part B | 20
478 Part B Autonomous Ocean Vehicles, Subsystems and Control
AUV on Adaptive Path
In the previous section, the bistatic tracking results were
obtained with the AUV on a preplanned path. The AUV
was not reacting to the target state estimate obtained
from the automated perception. The target speed was
adjusted to keep the target away from the endfires of
the receiver. This was done to ensure good tracking results on the target. Operationally, this is not realistic
since the target is never cooperative. In this section,
we will implement the integrated perception, modeling, and control paradigm on the AUV. The synthetic
data, generated from a high-fidelity simulator, is used
here with the same bistatic assets as the experimental data in Fig. 20.23. The simulated tactical scenario
is shown in Fig. 20.25. For this scenario, the stationary DEMUS source was simulated to transmit sonar
pulse signals. The ER, towed by the CRV Leonardo,
was used to simulate a noncooperative target cruising
eastward at a fixed speed of 2 m=s from ping k D 1 to
650. The OEX AUV, towing the BENS array, was commanded to move along the fixed path initially. With the
new paradigm implemented on the AUV, the target state
estimate of the ER from the automated perception is
now used to drive both the perception-driven control
and unified model.
The simulated results of the tactical scenario at different pings k are obtained in Fig. 20.26. The target
state estimates of the confirmed track #27 due to ER
from the automated perception were used to drive the
perception-driven control and unified model. The AUV
was initially moving on a fixed path, but switched to an
adaptive path upon tracking the entry of the ER. This
adaptive path was planned with the bistatic behavior
in the perception-driven control, with the objective of
maximizing the information and acoustic performance
of the target state estimate. With the adaptive path taken
by the AUV, good target tracking results are obtained
for track #27.
From the results obtained in Figs. 20.24 and 20.26,
it is clear that the integrated perception, modeling, and
control paradigm allows the path of the AUV to be
adaptively adjusted in reaction to the target state estimated from the tactical situation. The path has been
planned with the objective of optimizing the target
tracking performance. Such methodology can be easily
extended for distributed nodes in an AUV network.
20.7 Conclusion
Being dependent on acoustic communication with
a channel capacity many orders of magnitude smaller
than the air- and land-based equivalents, the operation of distributed ocean sensing networks requires
a much higher level of autonomous, distributed data
processing, and control than land- and air-based equivalents. The nested autonomy paradigm and its underlying
principle of integrated sensing, modeling and control
described here are inherently suited for the layered
communication infrastructure provided by the lowbandwidth underwater acoustic communication and the
intermittent RF connectivity. Implemented using the
open-source MOOS-IvP behavior-based, autonomous
command, and control architecture, nested autonomy
allows each platform to autonomously detect, classify,
localize and track (DCLT) an episodic ocean event
without depending on operator command and control.
The prosecution of a particular event may be initiated
by the operators through an acoustically transmitted
compact command message, cued by other sensing systems, such as satellite remote sensing, for example,
or fully autonomously by an on-board detection algorithm, but the fundamental principle of the paradigm
is that each network node shall be capable of completing its missions objectives, irrespective of communication connectivity. The inferred properties of the
event are reported back to the operators by transmitting a compact event report, using a dedicated command and control language. Collaborative processing
and control may be exploited when the communication
channel allows, for example, for collaborative tracking of extended features such as plumes and fronts.
The nested command and control paradigm has here
been demonstrated for the fully autonomous, adaptive tracking of the space- and time-varying coastal
thermocline and acousticline, using relatively simple
onboard signal processing, but the highly modular
MOOS-IvP architecture allows the event characterization and tracking algorithms to be exchanged with
customized and advanced processing software without
altering the operational functionality of the autonomy
system.
478 Part B Autonomous Ocean Vehicles, Subsystems and Control
AUV on Adaptive Path
In the previous section, the bistatic tracking results were
obtained with the AUV on a preplanned path. The AUV
was not reacting to the target state estimate obtained
from the automated perception. The target speed was
adjusted to keep the target away from the endfires of
the receiver. This was done to ensure good tracking results on the target. Operationally, this is not realistic
since the target is never cooperative. In this section,
we will implement the integrated perception, modeling, and control paradigm on the AUV. The synthetic
data, generated from a high-fidelity simulator, is used
here with the same bistatic assets as the experimental data in Fig. 20.23. The simulated tactical scenario
is shown in Fig. 20.25. For this scenario, the stationary DEMUS source was simulated to transmit sonar
pulse signals. The ER, towed by the CRV Leonardo,
was used to simulate a noncooperative target cruising
eastward at a fixed speed of 2 m=s from ping k D 1 to
650. The OEX AUV, towing the BENS array, was commanded to move along the fixed path initially. With the
new paradigm implemented on the AUV, the target state
estimate of the ER from the automated perception is
now used to drive both the perception-driven control
and unified model.
The simulated results of the tactical scenario at different pings k are obtained in Fig. 20.26. The target
state estimates of the confirmed track #27 due to ER
from the automated perception were used to drive the
perception-driven control and unified model. The AUV
was initially moving on a fixed path, but switched to an
adaptive path upon tracking the entry of the ER. This
adaptive path was planned with the bistatic behavior
in the perception-driven control, with the objective of
maximizing the information and acoustic performance
of the target state estimate. With the adaptive path taken
by the AUV, good target tracking results are obtained
for track #27.
From the results obtained in Figs. 20.24 and 20.26,
it is clear that the integrated perception, modeling, and
control paradigm allows the path of the AUV to be
adaptively adjusted in reaction to the target state estimated from the tactical situation. The path has been
planned with the objective of optimizing the target
tracking performance. Such methodology can be easily
extended for distributed nodes in an AUV network.
20.7 Conclusion
Being dependent on acoustic communication with
a channel capacity many orders of magnitude smaller
than the air- and land-based equivalents, the operation of distributed ocean sensing networks requires
a much higher level of autonomous, distributed data
processing, and control than land- and air-based equivalents. The nested autonomy paradigm and its underlying
principle of integrated sensing, modeling and control
described here are inherently suited for the layered
communication infrastructure provided by the lowbandwidth underwater acoustic communication and the
intermittent RF connectivity. Implemented using the
open-source MOOS-IvP behavior-based, autonomous
command, and control architecture, nested autonomy
allows each platform to autonomously detect, classify,
localize and track (DCLT) an episodic ocean event
without depending on operator command and control.
The prosecution of a particular event may be initiated
by the operators through an acoustically transmitted
compact command message, cued by other sensing systems, such as satellite remote sensing, for example,
or fully autonomously by an on-board detection algorithm, but the fundamental principle of the paradigm
is that each network node shall be capable of completing its missions objectives, irrespective of communication connectivity. The inferred properties of the
event are reported back to the operators by transmitting a compact event report, using a dedicated command and control language. Collaborative processing
and control may be exploited when the communication
channel allows, for example, for collaborative tracking of extended features such as plumes and fronts.
The nested command and control paradigm has here
been demonstrated for the fully autonomous, adaptive tracking of the space- and time-varying coastal
thermocline and acousticline, using relatively simple
onboard signal processing, but the highly modular
MOOS-IvP architecture allows the event characterization and tracking algorithms to be exchanged with
customized and advanced processing software without
altering the operational functionality of the autonomy
system.