Science of Au
481
Part B | 21
21. Science of Autonomy: Time-Optimal Path Planning
and Adaptive Sampling for Swarms of Ocean Vehicles
Pierre F.J. Lermusiaux, Tapovan Lolla, Patrick J. Haley Jr., Konuralp Yigit, Mattheus P. Ueckermann,
Thomas Sondergaard, Wayne G. Leslie
The science of autonomy is the systematic development of fundamental knowledge about
autonomous decision making and task completing in the form of testable autonomous methods,
models and systems. In ocean applications, it
involves varied disciplines that are not often connected. However, marine autonomy applications
are rapidly growing, both in numbers and in complexity. This new paradigm in ocean science and
operations motivates the need to carry out interdisciplinary research in the science of autonomy.
This chapter reviews some recent results and research directions in time-optimal path planning
and optimal adaptive sampling. The aim is to set
a basis for a large number of vehicles forming heterogeneous and collaborative underwater swarms
that are smart, i. e., knowledgeable about the
predicted environment and their uncertainties,
and about the predicted effects of autonomous
sensing on future operations. The methodologies are generic and applicable to any swarm
that moves and senses dynamic environmental fields. However, our focus is underwater path
planning and adaptive sampling with a range of
vehicles such as autonomous underwater ve21.1 Time-Optimal Path Planning
for Swarms of Ocean Vehicles ............... 482
21.1.1 Canonical Steady Flows .............. 483
21.1.2 Time-Dependent 2-D Flows........ 483
21.1.3 Maintain Swarms Formations ..... 485
21.1.4 Forbidden Regions .................... 485
21.1.5 Uncertain Flow Fields ................ 486
21.1.6 Realistic Ocean Conditions:
Three-Dimensional
Multi-Scale Flows ...................... 487
21.1.7 Realistic Ocean Conditions:
Swarms in Multi-Scale Flows
with Complex Geometries .......... 489
21.2 Adaptive Sampling for Swarms
of Ocean Vehicles ................................. 489
21.2.1 Early Results: Approximate
Schemes
for Realistic Applications ............ 490
21.2.2 Recent Progress: Towards
Rigorous Schemes
for Realistic Applications ............ 492
21.3 Conclusions and Outlook ...................... 494
References................................................... 496
hicles (AUVs), gliders, ships or remote sensing platforms.
Ocean observing is the process of collecting ocean
measurements to provide the most useful information
about the ocean systems and dynamics of interest. Such
marine sensing is expensive and challenging. In the
ocean, not everything can be measured on a sustained
basis. Ocean modeling is the process of developing
and utilizing theoretical and computational models for
the understanding and prediction of ocean dynamics.
Similar to ocean observing, ocean modeling is challenging, and not every naval and environmental process
can be modeled exactly. Stochastic ocean modeling explicitly acknowledges these uncertainties in modeling
and thus augments deterministic predictions with probabilistic information. Data assimilation is the process
of quantitatively estimating dynamically evolving fields
by combining information from observations with those
predicted by models, hence limiting uncertainties. With
stochastic predictive capabilities and data assimilation,
one can estimate future conditions and their uncertainties, and make informed decisions and complete desired
tasks. Systems that integrate some of these activities
without continuous human guidance are often referred
to as autonomous systems.
The science of autonomy is the systematic development of fundamental knowledge about autonomous
decision making and task completing in the form of
testable autonomous methods, models, and systems. In
ocean applications, it involves varied disciplines that
are not often connected. For example, ocean dynamics,
sensing, robotics, communications, predictions, and uncertainty, as well as varied types of human–computer–
vehicle interactions are inputs to ocean autonomy. There
are two important ocean modeling feedbacks to autonomy. The first one is the use of models to predict optimal
481
Part B | 21
21. Science of Autonomy: Time-Optimal Path Planning
and Adaptive Sampling for Swarms of Ocean Vehicles
Pierre F.J. Lermusiaux, Tapovan Lolla, Patrick J. Haley Jr., Konuralp Yigit, Mattheus P. Ueckermann,
Thomas Sondergaard, Wayne G. Leslie
The science of autonomy is the systematic development of fundamental knowledge about
autonomous decision making and task completing in the form of testable autonomous methods,
models and systems. In ocean applications, it
involves varied disciplines that are not often connected. However, marine autonomy applications
are rapidly growing, both in numbers and in complexity. This new paradigm in ocean science and
operations motivates the need to carry out interdisciplinary research in the science of autonomy.
This chapter reviews some recent results and research directions in time-optimal path planning
and optimal adaptive sampling. The aim is to set
a basis for a large number of vehicles forming heterogeneous and collaborative underwater swarms
that are smart, i. e., knowledgeable about the
predicted environment and their uncertainties,
and about the predicted effects of autonomous
sensing on future operations. The methodologies are generic and applicable to any swarm
that moves and senses dynamic environmental fields. However, our focus is underwater path
planning and adaptive sampling with a range of
vehicles such as autonomous underwater ve21.1 Time-Optimal Path Planning
for Swarms of Ocean Vehicles ............... 482
21.1.1 Canonical Steady Flows .............. 483
21.1.2 Time-Dependent 2-D Flows........ 483
21.1.3 Maintain Swarms Formations ..... 485
21.1.4 Forbidden Regions .................... 485
21.1.5 Uncertain Flow Fields ................ 486
21.1.6 Realistic Ocean Conditions:
Three-Dimensional
Multi-Scale Flows ...................... 487
21.1.7 Realistic Ocean Conditions:
Swarms in Multi-Scale Flows
with Complex Geometries .......... 489
21.2 Adaptive Sampling for Swarms
of Ocean Vehicles ................................. 489
21.2.1 Early Results: Approximate
Schemes
for Realistic Applications ............ 490
21.2.2 Recent Progress: Towards
Rigorous Schemes
for Realistic Applications ............ 492
21.3 Conclusions and Outlook ...................... 494
References................................................... 496
hicles (AUVs), gliders, ships or remote sensing platforms.
Ocean observing is the process of collecting ocean
measurements to provide the most useful information
about the ocean systems and dynamics of interest. Such
marine sensing is expensive and challenging. In the
ocean, not everything can be measured on a sustained
basis. Ocean modeling is the process of developing
and utilizing theoretical and computational models for
the understanding and prediction of ocean dynamics.
Similar to ocean observing, ocean modeling is challenging, and not every naval and environmental process
can be modeled exactly. Stochastic ocean modeling explicitly acknowledges these uncertainties in modeling
and thus augments deterministic predictions with probabilistic information. Data assimilation is the process
of quantitatively estimating dynamically evolving fields
by combining information from observations with those
predicted by models, hence limiting uncertainties. With
stochastic predictive capabilities and data assimilation,
one can estimate future conditions and their uncertainties, and make informed decisions and complete desired
tasks. Systems that integrate some of these activities
without continuous human guidance are often referred
to as autonomous systems.
The science of autonomy is the systematic development of fundamental knowledge about autonomous
decision making and task completing in the form of
testable autonomous methods, models, and systems. In
ocean applications, it involves varied disciplines that
are not often connected. For example, ocean dynamics,
sensing, robotics, communications, predictions, and uncertainty, as well as varied types of human–computer–
vehicle interactions are inputs to ocean autonomy. There
are two important ocean modeling feedbacks to autonomy. The first one is the use of models to predict optimal