Part B | 19.2
444 Part B Autonomous Ocean Vehicles, Subsystems and Control
and Seaglider [19.14], were designed to provide endurance [19.15, 16]. ABE was used to map the sea
bottom and to search out and study deep-sea hydrothermal vent sites and volcanoes [19.17–20]. The Odyssey
vehicle was used in experiments under the ice in the
Arctic [19.21]. REMUS was equipped with optical sensors and used to measure bioluminescence [19.22, 23].
Seagliders were sent on 5-month-long missions to measure physics, biology and chemistry off the coast of
Washington and in the Labrador Sea [19.15]. Slocum
gliders were introduced as an integral part of the longterm ecosystem observatory (LEO), an integrated observatory off the coast of New Jersey [19.24–26].
An early example of a control architecture design
for multiple AUVs was described in 1987 by Albus
and Blidberg in [19.27]: the architecture was designed
to enable two AUVs to perform cooperative search,
approach, and mapping using cooperative maneuvers
such as fly-formation, circle-split-and-rendezvous, and
leader-follower. In [19.28] the objective was to use
multiple AUVs as an imaging system. Virtual chains of
AUVs were considered by Triantafyllou and Streitlien
in [19.29]; a technology for one vehicle to track another in a chain-like fashion using an ultrashort baseline
(USBL) acoustic tracking system was demonstrated by
Singh et al. [19.30] in Buzzards Bay off Woods Hole,
MA in March 1996. In June 1996, two Odyssey vehicles were used along with an acoustic tomography
network for mapping in the Haro Strait region of British
Columbia [19.31]. Stilwell and Bishop presented a decentralized control framework for a cooperative platoon
of AUVs in [19.32]. Formation flying to map salinity
fronts was tested in the North River in North Carolina
using the Ranger micro-AUV [19.33].
A 1997 survey by Cao et al. [19.34] described cooperative mobile robotics as a still emerging field rich
with opportunities. The robotics community became
heavily engaged in the late 1980s first with a focus on
simulation [19.35, 36] and then physical implementations [19.37, 38]. Collective behavior in animal groups
was an early inspiration [19.39], with a behavioralbased approach becoming popular in the 1990s [19.40].
However, according to [19.34] in 1997, few applications of cooperative robotics had been reported, and
supporting theory was still in its formative stages.
Interest in very large-scale stationary sensor networks surged with advances in wireless communication
technology and microsensors, and environmental monitoring was an early, important application driver for
development of network architectures and algorithms
in this context [19.41–44]. Energy considerations were
used to justify the use of large numbers of stationary sensors over mobile sensors. However, the balance
tipped the other way in the case of undersea sensing due
to the challenges of undersea communication and the
emergence of relatively inexpensive, high-endurance
vehicles such as the autonomous underwater gliders.
Bretherton et al. [19.45] in the 1970s applied the
technique of objective analysis (OA) [19.46], which
uses classical linear estimation theory to compute objective maps, to address the problem of deployment
design for an array of stationary or passively drifting
sensors in the ocean where the aim was to provide
coverage and minimize uncertainty in the estimates
made from the data collected. Adaptive ocean sampling, as in [19.47], built on this concept to consider
the design of a trajectory for a mobile sensor platform,
complementing another sensor platform moving along
a predetermined track, in order to minimize uncertainty
in an ocean forecasting model. A performance metric
that accounts for both spatial and temporal sampling requirements was derived by Wilcox et al. [19.48]; it was
used to evaluate oceanographic survey performance
with AUVs in [19.49]. A methodology for control of
multiple sensor platforms based on information theory was presented in [19.50] and sampling strategies
driven by distributed parameter estimation were described in [19.51, 52].
Motivated by the many potential applications and
the rich theoretical possibilities, researchers in the control theory community began a significant effort in the
early 2000s to use systems theoretic approaches to design and study cooperative control. Artificial potentials
presented an attractive methodological basis for cooperative control of network formations [19.53–58] both
because convergence and performance could be proved
using Lyapunov stability theory (see early work on
robot navigation and obstacle avoidance [19.59, 60])
and because control laws derived from artificial potentials resembled the distributed, cohesive, and repulsive
forces used to model animals that move together [19.61,
62].
Artificial potential methods were also used to
design cooperative gradient climbing strategies for
a group of vehicles that could each only take a scalar
measurement at a time of the field of interest (e.g.,
ocean temperature) [19.63]. Capitalizing on this idea
and building on the methods of [19.55, 56], Ögren
et al. [19.4] developed a provable methodology to control the shape of the formation as well as the rotation,
translation, and expansion of the formation [19.64]; this
was used to design control strategies for a network of
vehicles to adaptively climb gradients in the sampled
field and thus robustly find peaks (Sect. 19.3.1 below
reviews the implementation of this methodology in the
field). These ideas were extended further by Zhang
and Leonard [19.65, 66] to design provable control
laws for cooperative level set tracking, whereby small
444 Part B Autonomous Ocean Vehicles, Subsystems and Control
and Seaglider [19.14], were designed to provide endurance [19.15, 16]. ABE was used to map the sea
bottom and to search out and study deep-sea hydrothermal vent sites and volcanoes [19.17–20]. The Odyssey
vehicle was used in experiments under the ice in the
Arctic [19.21]. REMUS was equipped with optical sensors and used to measure bioluminescence [19.22, 23].
Seagliders were sent on 5-month-long missions to measure physics, biology and chemistry off the coast of
Washington and in the Labrador Sea [19.15]. Slocum
gliders were introduced as an integral part of the longterm ecosystem observatory (LEO), an integrated observatory off the coast of New Jersey [19.24–26].
An early example of a control architecture design
for multiple AUVs was described in 1987 by Albus
and Blidberg in [19.27]: the architecture was designed
to enable two AUVs to perform cooperative search,
approach, and mapping using cooperative maneuvers
such as fly-formation, circle-split-and-rendezvous, and
leader-follower. In [19.28] the objective was to use
multiple AUVs as an imaging system. Virtual chains of
AUVs were considered by Triantafyllou and Streitlien
in [19.29]; a technology for one vehicle to track another in a chain-like fashion using an ultrashort baseline
(USBL) acoustic tracking system was demonstrated by
Singh et al. [19.30] in Buzzards Bay off Woods Hole,
MA in March 1996. In June 1996, two Odyssey vehicles were used along with an acoustic tomography
network for mapping in the Haro Strait region of British
Columbia [19.31]. Stilwell and Bishop presented a decentralized control framework for a cooperative platoon
of AUVs in [19.32]. Formation flying to map salinity
fronts was tested in the North River in North Carolina
using the Ranger micro-AUV [19.33].
A 1997 survey by Cao et al. [19.34] described cooperative mobile robotics as a still emerging field rich
with opportunities. The robotics community became
heavily engaged in the late 1980s first with a focus on
simulation [19.35, 36] and then physical implementations [19.37, 38]. Collective behavior in animal groups
was an early inspiration [19.39], with a behavioralbased approach becoming popular in the 1990s [19.40].
However, according to [19.34] in 1997, few applications of cooperative robotics had been reported, and
supporting theory was still in its formative stages.
Interest in very large-scale stationary sensor networks surged with advances in wireless communication
technology and microsensors, and environmental monitoring was an early, important application driver for
development of network architectures and algorithms
in this context [19.41–44]. Energy considerations were
used to justify the use of large numbers of stationary sensors over mobile sensors. However, the balance
tipped the other way in the case of undersea sensing due
to the challenges of undersea communication and the
emergence of relatively inexpensive, high-endurance
vehicles such as the autonomous underwater gliders.
Bretherton et al. [19.45] in the 1970s applied the
technique of objective analysis (OA) [19.46], which
uses classical linear estimation theory to compute objective maps, to address the problem of deployment
design for an array of stationary or passively drifting
sensors in the ocean where the aim was to provide
coverage and minimize uncertainty in the estimates
made from the data collected. Adaptive ocean sampling, as in [19.47], built on this concept to consider
the design of a trajectory for a mobile sensor platform,
complementing another sensor platform moving along
a predetermined track, in order to minimize uncertainty
in an ocean forecasting model. A performance metric
that accounts for both spatial and temporal sampling requirements was derived by Wilcox et al. [19.48]; it was
used to evaluate oceanographic survey performance
with AUVs in [19.49]. A methodology for control of
multiple sensor platforms based on information theory was presented in [19.50] and sampling strategies
driven by distributed parameter estimation were described in [19.51, 52].
Motivated by the many potential applications and
the rich theoretical possibilities, researchers in the control theory community began a significant effort in the
early 2000s to use systems theoretic approaches to design and study cooperative control. Artificial potentials
presented an attractive methodological basis for cooperative control of network formations [19.53–58] both
because convergence and performance could be proved
using Lyapunov stability theory (see early work on
robot navigation and obstacle avoidance [19.59, 60])
and because control laws derived from artificial potentials resembled the distributed, cohesive, and repulsive
forces used to model animals that move together [19.61,
62].
Artificial potential methods were also used to
design cooperative gradient climbing strategies for
a group of vehicles that could each only take a scalar
measurement at a time of the field of interest (e.g.,
ocean temperature) [19.63]. Capitalizing on this idea
and building on the methods of [19.55, 56], Ögren
et al. [19.4] developed a provable methodology to control the shape of the formation as well as the rotation,
translation, and expansion of the formation [19.64]; this
was used to design control strategies for a network of
vehicles to adaptively climb gradients in the sampled
field and thus robustly find peaks (Sect. 19.3.1 below
reviews the implementation of this methodology in the
field). These ideas were extended further by Zhang
and Leonard [19.65, 66] to design provable control
laws for cooperative level set tracking, whereby small