CURE: Simulation-Augmented Auto-Tuning in Robotics
Md Abir Hossen, Sonam Kharade, Jason M. O\'Kane,
Bradley Schmerl,
David Garlan and
Pooyan Jamshidi.
2024. Submitted for publication.
Online links:
Abstract
| Robotic systems are typically composed of vari-
ous subsystems, such as localization and navigation, each en-
compassing numerous configurable components (e.g., selecting
different planning algorithms). Once an algorithm has been
selected for a component, its associated configuration options
must be set to the appropriate values. Configuration options
across the system stack interact non-trivially. Finding optimal
configurations for highly configurable robots to achieve desired
performance poses a significant challenge due to the interactions
between configuration options across software and hardware
that result in an exponentially large and complex configuration
space. These challenges are further compounded by the need
for transferability between different environments and robotic
platforms. Data efficient optimization algorithms (e.g., Bayesian
optimization) have been increasingly employed to automate the
tuning of configurable parameters in cyber-physical systems.
However, such optimization algorithms converge at later stages,
often after exhausting the allocated budget (e.g., optimization
steps, allotted time) and lacking transferability. This paper
proposes CURE—a method that identifies causally relevant con-
figuration options, enabling the optimization process to operate
in a reduced search space, thereby enabling faster optimization
of robot performance. CURE abstracts the causal relationships
between various configuration options and robot performance
objectives by learning a causal model in the source (a low-cost
environment such as the Gazebo simulator) and applying the
learned knowledge to perform optimization in the target (e.g.,
Turtlebot 3 physical robot). We demonstrate the effectiveness and
transferability of CURE by conducting experiments that involve
varying degrees of deployment changes in both physical robots
and simulation |
Keywords: Cyberphysical Systems, Machine Learning, Robot Adaptation, Self-adaptation.
@Unpublished{2024:Hossen:CURE,
AUTHOR = {Hossen, Md Abir and Kharade, Sonam and O\'Kane, Jason M. and Schmerl, Bradley and Garlan, David and Jamshidi, Pooyan},
TITLE = {CURE: Simulation-Augmented Auto-Tuning in Robotics},
YEAR = {2024},
ABSTRACT = {Robotic systems are typically composed of vari-
ous subsystems, such as localization and navigation, each en-
compassing numerous configurable components (e.g., selecting
different planning algorithms). Once an algorithm has been
selected for a component, its associated configuration options
must be set to the appropriate values. Configuration options
across the system stack interact non-trivially. Finding optimal
configurations for highly configurable robots to achieve desired
performance poses a significant challenge due to the interactions
between configuration options across software and hardware
that result in an exponentially large and complex configuration
space. These challenges are further compounded by the need
for transferability between different environments and robotic
platforms. Data efficient optimization algorithms (e.g., Bayesian
optimization) have been increasingly employed to automate the
tuning of configurable parameters in cyber-physical systems.
However, such optimization algorithms converge at later stages,
often after exhausting the allocated budget (e.g., optimization
steps, allotted time) and lacking transferability. This paper
proposes CURE—a method that identifies causally relevant con-
figuration options, enabling the optimization process to operate
in a reduced search space, thereby enabling faster optimization
of robot performance. CURE abstracts the causal relationships
between various configuration options and robot performance
objectives by learning a causal model in the source (a low-cost
environment such as the Gazebo simulator) and applying the
learned knowledge to perform optimization in the target (e.g.,
Turtlebot 3 physical robot). We demonstrate the effectiveness and
transferability of CURE by conducting experiments that involve
varying degrees of deployment changes in both physical robots
and simulation},
NOTE = {Submitted for publication},
KEYWORDS = {Cyberphysical Systems, Machine Learning, Robot Adaptation, Self-adaptation}
}
|
