User:Huber

From WWWwikiEn

E-Mail:

Academic Career

Research Interests

• Sensor-actuator-networks
• System and estimation theory
• Stochastic Control
• Machine learning and robotics.

DFG Research Training Group 1194 "Self-organizing Sensor-Actuator-Networks"

Teaching

• Tutor and coordinator of the lecture Lokalisierung mobiler Agenten (summer term 2007 and summer term 2008)
• Supervising tutor of the laboratory Praktikum: Forschungsprojekt "Intelligente Sensor-Aktor-Systeme" (summer term 2006 to winter term 2007/2008)

Awards

• Award for best tutorial in the summer term 2008 for the tutorial of the lecture Lokalisierung mobiler Agenten
• Award for best elective class in the summer term 2008 for the lecture Lokalisierung mobiler Agenten
• Award for best tutorial in the summer term 2007 for the tutorial of the lecture Lokalisierung mobiler Agenten
• Award for best laboratory experiment in the summer term 2007 for the laboratory Praktikum: Forschungsprojekt "Intelligente Sensor-Aktor-Systeme"
• Award for wasing among the best ten percent of graduates of the year 2005/2006

Publications

tm - Technisches Messen, Oldenbourg Verlag, 77(10):551-557, October, 2010.

URL

Author : Achim Kuwertz, Marco F. Huber, Felix Sawo, Uwe D. Hanebeck
Title : Modellbasierte Quellenverfolgung in räumlich ausgedehnten Phänomenen mittels Sensoreinsatzplanung
In : tm - Technisches Messen, Oldenbourg Verlag
Date : October 2010

Abstract

Bewegte Quellen können durch Emission räumlich
ausgedehnte Phänomene wie beispielsweise Schadstoff- oder
Temperaturverteilungen erzeugen. Zur Lokalisierung von Quellen
mit unbekannter Position stehen in vielen Aufgabenstellungen
Informationen nur indirekt durch die verteilte Vermessung des
induzierten Phänomens zur Verfügung - etwa unter Verwendung
stationärer oder mobiler Sensoren. Dieser Beitrag stellt
modellbasierte Verfahren für eine echtzeitfähige Lokalisierung
und Verfolgung von bewegten Quellen vor. Zur gezielten Maximierung
des Informationsgehalts der Messungen wird dabei eine vorausschauende
Sensoreinsatzplanung genutzt, welche eine hohe Lokalisierungsgüte bei
geringem Aufwand ermöglicht.

Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010), Taipei, Taiwan, October, 2010.

Author : Frederik Beutler, Marco F. Huber, Uwe D. Hanebeck
Title : Optimal Stochastic Linearization for Range-based Localization
In : Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)
Date : October 2010

Abstract

In range-based localization, the trajectory of a
mobile object is estimated based on noisy range measurements
between the object and known landmarks. In order to deal
with this uncertain information, a Bayesian state estimator
is presented, which exploits optimal stochastic linearization.
Compared to standard state estimators like the Extended
or Unscented Kalman Filter, where a point-based Gaussian
approximation is used, the proposed approach considers the
entire Gaussian density for linearization. By employing the common
assumption that the state and measurements are jointly
Gaussian, the linearization can be calculated in closed form
and thus analytic expressions for the range-based localization
problem can be derived.

Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010), Salt Lake City, Utah, USA, September, 2010.

Author : Frederik Beutler, Marco F. Huber, Uwe D. Hanebeck
Title : Semi-Analytic Stochastic Linearization for Range-Based Pose Tracking
In : Proceedings of the 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2010)
Date : September 2010

Abstract

In range-based pose tracking, the translation and
rotation of an object with respect to a global coordinate system
has to be estimated. The ranges are measured between the
target and the global frame. In this paper, an intelligent decomposition
is introduced in order to reduce the computational
effort for pose tracking. Usually, decomposition procedures only
exploit conditionally linear models. In this paper, this principle
is generalized to conditionally integrable substructures and
applied to pose tracking. Due to a modified measurement
equation, parts of the problem can even be solved analytically.

Proceedings of the 13th International Conference on Information Fusion (Fusion 2010), Edinburgh, United Kingdom, July, 2010.

Author : Peter Krauthausen, Marco F. Huber, Uwe D. Hanebeck
Title : Support-Vector Conditional Density Estimation for Nonlinear Filtering
In : Proceedings of the 13th International Conference on Information Fusion (Fusion 2010)
Date : July 2010

Abstract

A non-parametric conditional density
estimation algorithm for nonlinear stochastic dynamic
systems is proposed. The contributions are a novel sup-
port vector regression for estimating conditional den-
sities, modeled by Gaussian mixture densities, and an
algorithm based on cross-validation for automatically
determining hyper-parameters for the regression. The
conditional densities are employed with a modified axis-
aligned Gaussian mixture filter. The experimental va-
lidation shows the high quality of the conditional densi-
ties and good accuracy of the proposed filter.

Verteilte Messsysteme, pp. 179-191, KIT Scientific Publishing, March, 2010.

URL

Author : Achim Kuwertz, Marco F. Huber, Felix Sawo, Uwe D. Hanebeck
Title : Sensoreinsatzplanung zur Verfolgung von Quellen räumlich ausgedehnter Phänomene
In : Verteilte Messsysteme
Date : March 2010

Abstract

Räumlich ausgedehnte Phänomene wie Schadstoffverteilungen in Gewässern
oder Temperaturverteilungen in Räumen werden vielfach durch unbekannte, aber gegebenenfalls
sich bewegende Quellen erzeugt. Allerdings stehen in vielen praktisch relevanten Aufgabenstellungen
Informationen zur Lokalisierung einer derartigen Quelle nur indirekt durch eine Vermessung des
induzierten Phänomens zur Verfügung, welche den Einsatz eines verteilten Messsystems erfordert.
Die Messungen können dabei beispielsweise von einem stationären Sensornetz oder von mobilen
Sensoren stammen. In diesem Beitrag werden modellbasierte Verfahren zu echtzeitfähigen Lokalisierung
und schritthaltenden Verfolgung von Quellen vorgestellt, welche gezielt räumlich und zeitlich
verteilte Messungen einsetzen. Um den Informationsgewinn und somit den Nutzen verteilter Messungen zu
maximieren, spielt bei diesem Verfahren neben einer mathematischen Modellierung auch eine vorausschauende
Sensoreinsatzplanung eine zentrale Rolle. Das in diesem Beitrag vorgeschlagene Planungsverfahren
ermöglicht dabei die effiziente und ressourcenschonende Verfolgung beweglicher Quellen bei gleichzeitig
hoher Lokalisierungsgenauigkeit.

Proceedings of the 2009 IEEE Conference on Decision and Control (CDC 2009), Shanghai, China, December, 2009.

Author : Uwe D. Hanebeck, Marco F. Huber, Vesa Klumpp
Title : Dirac Mixture Approximation of Multivariate Gaussian Densities
In : Proceedings of the 2009 IEEE Conference on Decision and Control (CDC 2009)
Date : December 2009

Abstract

For the optimal approximation of multivariate
Gaussian densities by means of Dirac mixtures, i.e., by means of
a sum of weighted Dirac distributions on a continuous domain,
a novel systematic method is introduced. The parameters of
this approximate density are calculated by minimizing a global
distance measure, a generalization of the well–known Cramér–
von Mises distance to the multivariate case. This generalization
is obtained by defining an alternative to the classical cumulative
distribution, the Localized Cumulative Distribution (LCD). In
contrast to the cumulative distribution, the LCD is unique
and symmetric even in the multivariate case. The resulting
deterministic approximation of Gaussian densities by means of
discrete samples provides the basis for new types of Gaussian
filters for estimating the state of nonlinear dynamic systems
from noisy measurements.

Proceedings of the 12th International Conference on Information Fusion (Fusion 2009), Seattle, Washington, USA, July, 2009.

Author : Frederik Beutler, Marco F. Huber, Uwe D. Hanebeck
Title : Gaussian Filtering using State Decomposition Methods
In : Proceedings of the 12th International Conference on Information Fusion (Fusion 2009)
Date : July 2009

Abstract

State estimation for nonlinear systems generally
requires approximations of the system or the probability
densities, as the occurring prediction and filtering equations
cannot be solved in closed form. For instance, Linear Regression
Kalman Filters like the Unscented Kalman Filter
or the considered Gaussian Filter propagate a small set of
sample points through the system to approximate the posterior
mean and covariance matrix. To reduce the number of
sample points, special structures of the system and measurement
equation can be taken into account. In this paper, two
principles of system decomposition are considered and applied
to the Gaussian Filter. One principle exploits that only
a part of the state vector is directly observed by the measurement.
The second principle separates the system equations
into linear and nonlinear parts in order to merely approximate
the nonlinear part of the state. The benefits of both
decompositions are demonstrated on a real-world example.

Proceedings of the 12th International Conference on Information Fusion (Fusion 2009), Seattle, Washington, USA, July, 2009.

Author : Marco F. Huber, Achim Kuwertz, Felix Sawo, Uwe D. Hanebeck
Title : Distributed Greedy Sensor Scheduling for Model-based Reconstruction of Space-Time Continuous Physical Phenomena
In : Proceedings of the 12th International Conference on Information Fusion (Fusion 2009)
Date : July 2009

Abstract

A novel distributed sensor scheduling method for large-scale sensor
networks observing space-time continuous physical phenomena is
introduced. In a first step, the model of the distributed phenomenon
is spatially and temporally decomposed leading to a linear
probabilistic finite-dimensional model. Based on this representation,
the information gain of sensor measurements is evaluated by means of
the so-called covariance reduction function. For this reward function,
it is shown that the performance of the greedy sensor scheduling is at
least half that of the optimal scheduling considering long-term
effects. This finding is the key for distributed sensor scheduling,
where a central processing unit or fusion center is unnecessary, and
thus, scaling as well as reliability is ensured. Hence, greedy
scheduling in combination with a proposed hierarchical communication
scheme requires only local sensor information and communication.

Proceedings of the 12th International Conference on Information Fusion (Fusion 2009), Seattle, Washington, USA, July, 2009.

Author : Dennis Schieferdecker, Marco F. Huber
Title : Gaussian Mixture Reduction via Clustering
In : Proceedings of the 12th International Conference on Information Fusion (Fusion 2009)
Date : July 2009

Abstract

Recursive processing of Gaussian mixture functions inevitably leads to a
large number of mixture components. In order to keep the computational
complexity at a feasible level, the number of their components has to be
reduced periodically. There already exists a variety of algorithms for
this purpose, bottom-up and top-down approaches, methods that take the
global structure of the mixture into account or that work locally and
consider few mixture components at the same time. The mixture reduction
algorithm presented in this paper can be categorized as global top-down
approach. It takes a clustering algorithm originating from the field of
theoretical computer science and adapts it for the problem of Gaussian
mixture reduction. The achieved results are on the same scale as the
results of the current “state-of-the-art” algorithm PGMR, but, depending
on the input size, the whole procedure performs significantly faster.

26th International Conference on Machine Learning (ICML 2009) in Montreal, Canada, June, 2009.

Author : Marc P. Deisenroth, Marco F. Huber, Uwe D. Hanebeck
Title : Analytic Moment-based Gaussian Process Filtering
In : 26th International Conference on Machine Learning (ICML 2009) in Montreal, Canada
Date : June 2009

Abstract

We propose an analytic moment-based filter for nonlinear stochastic
dynamic systems modeled by Gaussian processes. Exact expressions for the
expected value and the covariance matrix are provided for both the
prediction step and the filter step, where an additional Gaussian
assumption is exploited in the latter case. Our filter does not require
further approximations. In particular, it avoids finite-sample
approximations. We compare the filter to a variety of Gaussian filters,
that is, the EKF, the UKF, and the recent GP-UKF proposed by Ko et al.
(2007).

IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2009) in Taipei, Taiwan, pp. 3413-3416, April, 2009.

Author : Frederik Beutler, Marco F. Huber, Uwe D. Hanebeck
Title : Instantaneous Pose Estimation using Rotation Vectors
In : IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2009) in Taipei, Taiwan
Date : April 2009

Abstract

An algorithm for estimating the pose, i.e., translation and rotation, of
an extended target object is introduced. Compared to conventional
methods, where pose estimation is performed on the basis of timeof-
flight (TOF) measurements between external sources and sensors
attached to the object, the proposed approach directly uses the amplitude
values measured at the sensors for estimation purposes without
an intermediate TOF estimation step. This is achieved by modeling
the wave propagation by a nonlinear dynamic system comprising a
system and a measurement equation. The nonlinear system equation
includes a model of the time-variant structure of the object rotation
based on rotation vectors. As a result, the measured amplitude values
at the sensors can be processed instantaneously in a recursive
fashion. Uncertainties in the measurement process are systematically
considered by employing a stochastic filter for estimating the
pose, i.e., the state of the nonlinear dynamic system.

Journal of Robotics and Autonomous Systems, Selected papers from 2006 IEEE International Conference on Multisensor Fusion and Integration (MFI 2006), 57(3):249-258, March, 2009.

URL

Author : Frederik Beutler, Marco F. Huber, Uwe D. Hanebeck
Title : Probabilistic Instantaneous Model-Based Signal Processing applied to Localization and Tracking
In : Journal of Robotics and Autonomous Systems, Selected papers from 2006 IEEE International Conference on Multisensor Fusion and Integration (MFI 2006)
Date : March 2009

Abstract

In this paper, a probabilistic approach for estimating time
and space-variant parameters of a system, based on sequentially received
discrete-time signal values, is presented. The system description is the
solution of a linear partial differential equation (PDE). The PDE describes
for example the wave propagation of an acoustic wave in a localization
system. The solution of the PDE is given by a time-variant and space-variant
impulse response. This impulse response is characterized by the time and
space-variant parameters in order to track an object, which emits for example
an acoustic signal. For estimating the position of the object in an
instantaneous way a Bayesian approach has to be used, which considers the
dynamic behavior of the parameters in a system model and uncertainties in a
stochastic manner by means of probability density functions. Hence, the new
approach provides a probabilistic instantaneous model-based signal processing,
where the sequentially measured signal values are processed directly and known
reference signal sequences are interpreted as part of a time-variant nonlinear
measurement equation.

Informatics in Control, Automation and Robotics -- Selected Papers from the International Conference on Informatics in Control, Automation and Robotics 2007, 24:239-252, Springer, September, 2008.

URL

Author : Florian Weissel, Marco F. Huber, Uwe D. Hanebeck
Title : Stochastic Nonlinear Model Predictive Control based on Gaussian Mixture Approximations
In : Informatics in Control, Automation and Robotics -- Selected Papers from the International Conference on Informatics in Control, Automation and Robotics 2007
Date : September 2008

Abstract

In this paper, a framework for stochastic Nonlinear
Model Predictive Control (NMPC) that explicitly incorporates the
noise influence on systems with continuous state spaces is introduced.
By the incorporation of noise, which results from uncertainties during
model identification and measurement, the quality of control can be
significantly increased. Since stochastic NMPC requires the prediction
of system states over a certain horizon, an efficient state prediction
technique for nonlinear noise-affected systems is required. This is
achieved by using transition densities approximated by axis-aligned
Gaussian mixtures together with methods to reduce the computational burden.
A versatile cost function representation also employing Gaussianmixtures
provides an increased freedom of modeling. Combining the rediction technique
with this value function representation allows closed-form calculation of
the necessary optimization problems arising from stochastic NMPC. The
capabilities of the framework and especially the benefits that can be
gained by considering the noise in the controller are illustrated by the
example of a mobile robot following a given path.

Proceedings of the 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2008), pp. 181-188, Seoul, Republic of Korea, August, 2008.

Author : Marco F. Huber, Tim Bailey, Hugh Durrant-Whyte, Uwe D. Hanebeck
Title : On Entropy Approximation for Gaussian Mixture Random Vectors
In : Proceedings of the 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2008)
Date : August 2008

Abstract

For many practical probability density representations
such as for the widely used Gaussian mixture densities, an
analytic evaluation of the differential entropy is not possible and
thus, approximate calculations are inevitable. For this purpose,
the first contribution of this paper deals with a novel entropy
approximation method for Gaussian mixture random vectors,
which is based on a component-wise Taylor-series expansion of
the logarithm of a Gaussian mixture and on a splitting method
of Gaussian mixture components. The employed order of the
Taylor-series expansion and the number of components used for
splitting allows balancing between accuracy and computational
demand. The second contribution is the determination of meaningful
and efficiently to calculate lower and upper bounds of the
entropy, which can be also used for approximation purposes.
In addition, a refinement method for the more important upper
bound is proposed in order to approach the true entropy value.

Proceedings of the 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2008), pp. 40-46, Seoul, Republic of Korea, August, 2008.

Author : Florian Weissel, Thomas Schreiter, Marco F. Huber, Uwe D. Hanebeck
Title : Stochastic Model Predictive Control of Time-Variant Nonlinear Systems with Imperfect State Information
In : Proceedings of the 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2008)
Date : August 2008

Abstract

In many technical systems, the system state, which
is to be controlled, is not directly accessible, but has to be
estimated from observations. Furthermore, the uncertainties
arising from this procedure are typically neglected in the
controller. To remedy this deficiency, in this paper, we present a
novel approach to stochastic nonlinear model predictive control
(NMPC) for heavily noise-affected systems with not directly
accessible, i.e., hidden states, extending the stochastic NMPCframework
presented in [1]. An important property of our novel
method is that, in contrast to classical approaches, time-variant
system and measurement equations as well as time-variant step
rewards can be considered. Extending the techniques from
[1] by introducing virtual future observations and combining
this with a novel tree search algorithm, called probabilistic
branch-and-bound search (PBAB), a solution with a feasible
computational demand of the challenging problem is possible.

Proceedings of the 17th IFAC World Congress (IFAC 2008), 17, Seoul, Republic of Korea, July, 2008.

Author : Marco F. Huber, Uwe D. Hanebeck
Title : Gaussian Filter based on Deterministic Sampling for High Quality Nonlinear Estimation
In : Proceedings of the 17th IFAC World Congress (IFAC 2008)
Date : July 2008

Abstract

In this paper, a Gaussian filter for nonlinear Bayesian estimation is introduced that is
based on a deterministic sample selection scheme. For an effective sample selection, a parametric
density function representation of the sample points is employed, which allows approximating the
cumulative distribution function of the prior Gaussian density. The computationally demanding
parts of the optimization problem formulated for approximation are carried out off-line for
obtaining an efficient filter, whose estimation quality can be altered by adjusting the number
of used sample points. The improved performance of the proposed Gaussian filter compared to
the well-known unscented Kalman fiter is demonstrated by means of two examples.

Proceedings of the 17th IFAC World Congress (IFAC 2008), 17, Seoul, Republic of Korea, July, 2008.

Author : Florian Weissel, Marco F. Huber, Dietrich Brunn, Uwe D. Hanebeck
Title : Stochastic Optimal Control based on Value-Function Approximation using Sinc Interpolation
In : Proceedings of the 17th IFAC World Congress (IFAC 2008)
Date : July 2008

Abstract

An effcient approach for solving stochastic optimal control problems is to employ
dynamic programming (DP). For continuous-valued nonlinear systems, the corresponding
DP recursion generally cannot be solved in closed form. Thus, a typical approach is to
discretize the DP value functions in order to be able to carry out the calculation. Especially
for multidimensional systems, either a large number of discretization points is necessary or
the quality of approximation degrades. This problem can be alleviated by interpolating the
discretized value function. In this paper, we present an approach based on optimal low-pass
interpolation employing sinc functions (sine cardinal). For the important case of systems with
Gaussian mixture noise (including the special case of Gaussian noise), we show how the
calculations required for this approach, especially the nontrivial calculation of an expected
value of a Gaussian mixture random variable transformed by a sinc function, can be carried out
analytically. We illustrate the effectiveness of the proposed interpolation scheme by an example
from the field of Stochastic Nonlinear Model Predictive Control (SNMPC).

Proceedings of the 11th International Conference on Information Fusion (Fusion 2008), pp. 1-8, Cologne, Germany, July, 2008.

Author : Marco F. Huber, Uwe D. Hanebeck
Title : Progressive Gaussian Mixture Reduction
In : Proceedings of the 11th International Conference on Information Fusion (Fusion 2008)
Date : July 2008

Abstract

For estimation and fusion tasks it is inevitable to
approximate a Gaussian mixture by one with fewer components
to keep the complexity bounded. Appropriate approximations
can be typically generated by exploiting the redundancy in
the shape description of the original mixture. In contrast to
the common approach of successively merging pairs of components
to maintain a desired complexity, the novel Gaussian
mixture reduction algorithm introduced in this paper avoids
to directly reduce the original Gaussian mixture. Instead, an
approximate mixture is generated from scratch by employing
homotopy continuation. This allows starting the approximation
with a single Gaussian, which is constantly adapted to the
progressively incorporated true Gaussian mixture. Whenever a
user-defined bound on the deviation of the approximation cannot
be maintained during the continuation, further components are
added to the approximation. This facilitates significantly reducing
the number of components even for complex Gaussian mixtures.

Proceedings of the 11th International Conference on Information Fusion (Fusion 2008), pp. 1-8, Cologne, Germany, July, 2008.

Author : Marco F. Huber, Uwe D. Hanebeck
Title : Priority List Sensor Scheduling using Optimal Pruning
In : Proceedings of the 11th International Conference on Information Fusion (Fusion 2008)
Date : July 2008

Abstract

State estimation and reconstruction quality of distributed
phenomena that are monitored by a network of distributed
sensors is strongly affected by communication failures,
which is a problem in real-world sensor networks. In this paper,
we propose a novel sensor scheduling approach named priority
list sensor scheduling (PLSS). This approach facilitates efficient
distributed estimation in sensor networks, even in case of unreliable
communication, by prioritizing the sensor nodes according
to local sensor schedules based on the predicted estimation
error. It is shown that PLSS minimizes the expected estimation
error for arbitrary packet-loss or transmission probabilities.
As prioritizing sensor nodes requires the calculation of several
sensor schedules, a novel pruning algorithm that preserves
optimal schedules is also derived in order to significantly reduce
the computational demand. This is accomplished by exploiting
the monotonicity of the Riccati equation and the information
contribution of individual sensor nodes in combination with a
branch-and-bound technique.

Proceedings of the 2007 IEEE Conference on Decision and Control (CDC 2007), pp. 3661-3666, New Orleans, Louisiana, USA, December, 2007.

Author : Florian Weissel, Marco F. Huber, Uwe D. Hanebeck
Title : A Nonlinear Model Predictive Control Framework Approximating Noise Corrupted Systems with Hybrid Transition Densities
In : Proceedings of the 2007 IEEE Conference on Decision and Control (CDC 2007)
Date : December 2007

Abstract

In this paper, a framework for Nonlinear Model Predictive Control
(NMPC) for heavily noise-affected systems is presented. Within this
framework, the noise influence, which originates from uncertainties
during model identification or measurement, is explicitly considered.
This leads to a significant increase in the control quality. One
part of the proposed framework is the efficient state prediction,
which is necessary for NMPC. It is based on transition density approximation
by hybrid transition densities, which allows efficient closed-form
state prediction of time-variant nonlinear systems with continuous
state spaces in discrete time. Another part of the framework is a
versatile value function representation using Gaussian mixtures,
Dirac mixtures, and even a combination of both. Based on these methods,
an efficient closed-form algorithm for calculating an approximate
value function of the NMPC optimal control problem employing dynamic
programming is presented. Thus, also very long optimization horizons
can be used and furthermore it is possible to calculate the value
function off-line, which reduces the on-line computational burden
significantly. The capabilities of the framework and especially the
benefits that can be gained by incorporating the noise in the controller
are illustrated by the example of a miniature walking robot following
a given path.

Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2007), pp. 2474-2479, San Diego, California, USA, November, 2007.

Author : Florian Weissel, Marco F. Huber, Uwe D. Hanebeck
Title : Test-Environment based on a Team of Miniature Walking Robots for Evaluation of Collaborative Control Methods
In : Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2007)
Date : November 2007

Abstract

For the collaborative control of a team of robots, a set of well-suited
high-level control algorithms, especially for path planning and measurement
scheduling, is essential. The quality of these control algorithms
can be significantly increased by considering uncertainties that
arise, e.g. from noisy measurements or system model abstraction,
by incorporating stochastic filters into the control. To develop
these kinds of algorithms and to prove their effectiveness, obviously
real-world experiments with real world uncertainties are mandatory.
Therefore, a test-environment for evaluating algorithms for collaborative
control of a team of robots is presented. This test-environment is
founded on miniature walking robots with six degrees of freedom.
Their novel locomotion concept not only allows them to move in a
wide variety of different motion patterns far beyond the possibilities
of traditionally employed wheel-based robots, but also to handle
real-world conditions like uneven ground or small obstacles. These
robots are embedded in a modular test-environment, comprising infrastructure
and simulation modules as well as a high-level control module with
submodules for pose estimation, path planning, and measurement scheduling.
The interaction of the individual modules of the introduced test-environment
is illustrated by an experiment from the field of cooperative localization
with focus on measurement scheduling, where the robots that perform
distance measurements are selected based on a novel criterion, the
normalized mutual Mahalanobis distance.

INFORMATIK 2007 - the 37th Annual Conference of the Gesellschaft für Informatik e.V. (GI), 3rd German Workshop Sensor Data Fusion: Trends, Solutions, Applications (SDF 2007), pp. 90-94, Bremen, Germany, September, 2007.

Author : Marco F. Huber, Eric Stiegeler, Uwe D. Hanebeck
Title : On Sensor Scheduling in Case of Unreliable Communication
In : INFORMATIK 2007 - the 37th Annual Conference of the Gesellschaft für Informatik e.V. (GI), 3rd German Workshop Sensor Data Fusion: Trends, Solutions, Applications (SDF 2007)
Date : September 2007

Abstract

This paper deals with the linear discrete-time sensor scheduling problem
in unreliable communication networks. In case of the common assumption
of an error-free communication, the sensor scheduling problem, where
one sensor from a sensor network is selected for measuring at a specific
time instant so that the estimation errors are minimized, can be
solved off-line by extensive tree search. For the more realistic
scenario, where communication is unreliable, a scheduling approach
using a prioritization list for the sensors is proposed that leads
to a minimization of the estimation error by selecting the most beneficial
sensor on-line. To lower the computational demand for the priority
list calculation, a novel optimal pruning approach is introduced.

Proceedings of the 2007 American Control Conference (ACC 2007), pp. 4425-4430, New York, New York, USA, July, 2007.

Author : Marco F. Huber, Dietrich Brunn, Uwe D. Hanebeck
Title : Efficient Nonlinear Measurement Updating based on Gaussian Mixture Approximation of Conditional Densities
In : Proceedings of the 2007 American Control Conference (ACC 2007)
Date : July 2007

Abstract

Filtering or measurement updating for nonlinear stochastic dynamic
systems requires approximate calculations, since an exact solution
is impossible to obtain in general. We propose a Gaussian mixture
approximation of the conditional density, which allows performing
measurement updating in closed form. The conditional density is a
probabilistic representation of the nonlinear system and depends
on the random variable of the measurement given the system state.
Unlike the likelihood, the conditional density is independent of
actual measurements, which permits determining its approximation
off-line. By treating the approximation task as an optimization problem,
we use progressive processing to achieve high quality results. Once
having calculated the conditional density, the likelihood can be
determined on-line, which, in turn, offers an efficient approximate
filter step. As result, a Gaussian mixture representation of the
posterior density is obtained. The exponential growth of Gaussian
mixture components resulting from repeated filtering is avoided implicitly
by the prediction step using the proposed techniques.

Proceedings of the 2007 American Control Conference (ACC 2007), pp. 3751-3756, New York, New York, USA, July, 2007.

Author : Florian Weissel, Marco F. Huber, Uwe D. Hanebeck
Title : Efficient Control of Nonlinear Noise-Corrupted Systems Using a Novel Model Predictive Control Framework
In : Proceedings of the 2007 American Control Conference (ACC 2007)
Date : July 2007

Abstract

Model identification and measurement acquisition is always to some
degree uncertain. Therefore, a framework for Nonlinear Model Predictive
Control (NMPC) is proposed that explicitly considers the noise influence
on nonlinear dynamic systems with continuous state spaces and a finite
set of control inputs in order to significantly increase the control
quality. Integral parts of NMPC are the prediction of system states
over a finite horizon as well as the problem specific modeling of
reward functions. For achieving an efficient and also accurate state
prediction, the introduced framework uses transition densities approximated
by means of axis-aligned Gaussian mixtures. The representation power
of Gaussian mixtures is also used to model versatile reward functions.
Thus, together with the prediction technique a closed-form calculation
of the optimization problems arising from NMPC is possible. Additionally,
an efficient algorithm for calculating an approximate value function
of the corresponding optimal control problem employing dynamic programming
is presented. Thus, the value function can be calculated off-line,
which reduces the on-line computational burden significantly and
also permits the use of long optimization horizons. The capabilities
of the framework and especially the benefits that can be gained by
incorporating the noise in the controller are illustrated by the
example of a two-wheeled differential-drive mobile robot following
a given path.

Proceedings of the 10th International Conference on Information Fusion (Fusion 2007), Quebec, Canada, July, 2007.

Author : Marco F. Huber, Uwe D. Hanebeck
Title : The Hybrid Density Filter for Nonlinear Estimation based on Hybrid Conditional Density Approximation
In : Proceedings of the 10th International Conference on Information Fusion (Fusion 2007)
Date : July 2007

Abstract

In nonlinear Bayesian estimation it is generally inevitable to incorporate
approximate descriptions of the exact estimation algorithm. There
are two possible ways to involve approximations: Approximating the
nonlinear stochastic system model or approximating the prior probability
density function. The key idea of the introduced novel estimator
called Hybrid Density Filter relies on approximating the nonlinear
system, thus approximating conditional densities. These densities
nonlinearly relate the current system state to the future system
state at predictions or to potential measurements at measurement
updates. A hybrid density consisting of both Dirac delta functions
and Gaussian densities is used for an optimal approximation. This
paper addresses the optimization problem for treating the conditional
density approximation. Furthermore, efficient estimation algorithms
are derived based upon the special structure of the hybrid density,
which yield a Gaussian mixture representation of the system state\'s
density.

Proceedings of the 10th International Conference on Information Fusion (Fusion 2007), Quebec, Canada, July, 2007.

Author : Felix Sawo, Marco F. Huber, Uwe D. Hanebeck
Title : Parameter Identification and Reconstruction for Distributed Phenomena Based on Hybrid Density Filter
In : Proceedings of the 10th International Conference on Information Fusion (Fusion 2007)
Date : July 2007

Abstract

This paper addresses the problem of model-based reconstruction and
parameter identification of distributed phenomena characterized by
partial differential equations. The novelty of the proposed method
is the systematic approach and the integrated treatment of uncertainties,
which naturally occur in the physical system and arise from noisy
measurements. The main challenge of accurate reconstruction is that
model parameters, i.e., diffusion coefficients, of the physical model
are not known in advance and usually need to be identified. Generally,
the problem of parameter identification leads to a nonlinear estimation
problem. Hence, a novel efficient recursive procedure is employed.
Unlike other estimators, the so-called Hybrid Density Filter not
only assures accurate estimation results for nonlinear systems, but
also offers an efficient processing. By this means it is possible
to reconstruct and identify distributed phenomena monitored by autonomous
wireless sensor networks. The performance of the proposed estimation
method is demonstrated by means of simulations.

Proceedings of the 4th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2007), SPSMC:62-69, Angers, France, May, 2007.

Author : Florian Weissel, Marco F. Huber, Uwe D. Hanebeck
Title : A Closed-Form Model Predictive Control Framework for Nonlinear Noise-Corrupted Systems
In : Proceedings of the 4th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2007)
Date : May 2007

Abstract

In this paper, a framework for Nonlinear Model Predictive Control
(NMPC) that explicitly incorporates the noise influence on systems
with continuous state spaces is introduced. By the incorporation
of noise, which results from uncertainties during model identification
and the measurement process, the quality of control can be significantly
increased. Since NMPC requires the prediction of system states over
a certain horizon, an efficient state prediction technique for nonlinear
noise-affected systems is required. This is achieved by using transition
densities approximated by axis-aligned Gaussian mixtures together
with methods to reduce the computational burden. A versatile cost
function representation also employing Gaussian mixtures provides
an increased freedom of modeling. Combining the prediction technique
with this value function representation allows closed-form calculation
of the necessary optimization problems arising from NMPC. The capabilities
of the framework and especially the benefits that can be gained by
considering the noise in the controller are illustrated by the example
of a mobile robot following a given path.

International Conference on Information Processing in Sensor Networks (IPSN 2007), pp. 283-292, Cambridge, Massachusetts, USA, April, 2007.

Author : Marco F. Huber, Uwe D. Hanebeck
Title : Hybrid Transition Density Approximation for Efficient Recursive Prediction of Nonlinear Dynamic Systems
In : International Conference on Information Processing in Sensor Networks (IPSN 2007)
Date : April 2007

Abstract

For several tasks in sensor networks, such as localization, information
fusion, or sensor scheduling, Bayesian estimation is of paramount
importance. Due to the limited computational and memory resources
of the nodes in a sensor network, evaluation of the prediction step
of the Bayesian estimator has to be performed very efficiently. An
exact and closed-form representation of the predicted probability
density function of the system state is typically impossible to obtain,
since exactly solving the prediction step for nonlinear discrete-time
dynamic systems in closed form is unfeasible. Assuming additive noise,
we propose an accurate approximation of the predicted density, that
can be calculated efficiently by optimally approximating the transition
density using a hybrid density. A hybrid density consists of two
different density types: Dirac delta functions that cover the domain
of the current density of the system state, and another density type,
e.g. Gaussian densities, that cover the domain of the predicted density.
The freely selectable, second density type of the hybrid density
depends strongly on the noise affecting the nonlinear system. So,
the proposed approximation framework for nonlinear prediction is
not restricted to a specific noise density. It further allows an
analytical evaluation of the Chapman-Kolmogorov prediction equation
and can be interpreted as a deterministic sampling estimation approach.
In contrast to methods using random sampling like particle filters,
a dramatic reduction in the number of components and a subsequent
decrease in computation time for approximating the predicted density
is gained.

Proceedings of the 2006 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2006), pp. 98-103, Heidelberg, Germany, September, 2006.

Author : Marco Huber, Dietrich Brunn, Uwe D. Hanebeck
Title : Closed-Form Prediction of Nonlinear Dynamic Systems by Means of Gaussian Mixture Approximation of the Transition Density
In : Proceedings of the 2006 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2006)
Date : September 2006

Abstract

Recursive prediction of the state of a nonlinear stochastic dynamic
system cannot be efficiently performed in general, since the complexity
of the probability density function characterizing the system state
increases with every prediction step. Thus, representing the density
in an exact closed-form manner is too complex or even impossible.
So, an appropriate approximation of the density is required. Instead
of directly approximating the predicted density, we propose the approximation
of the transition density by means of Gaussian mixtures. We treat
the approximation task as an optimization problem that is solved
offline via progressive processing to bypass initialization problems
and to achieve high quality approximations. Once having calculated
the transition density approximation offline, prediction can be performed
efficiently resulting in a closed-form density representation with
constant complexity.

Proceedings of the 8th International Conference on Climbing and Walking Robots (CLAWAR 2005), 1:115-122, London, United Kingdom, September, 2005.

Author : Bernd Gaßmann, Marco Huber, J. Marius Zöllner, Rüdiger Dillmann
Title : Navigation of Walking Robots: Path Planning
In : Proceedings of the 8th International Conference on Climbing and Walking Robots (CLAWAR 2005)
Date : September 2005

Abstract

Proper navigation of walking machines in unstructured terrain requires
a path planning algorithm that reflects the flexibility of the robots
movements. This paper discusses the problem and presents a real time
capable path planning algorithm for walking machines that also makes
use of their climbing abilities.

Selected Talks

2009

Marco Huber
Probabilistic Framework for Sensor Management
Report colloquium of the GRK 1194
Karlsruhe, May 27, 2009

Marco Huber
Probabilistic Framework for Sensor Management
Defense of phd thesis
Karlsruhe, April 30, 2009

Marco Huber
Quasi-linear Sensor Management
Computational and Biological Learning Lab (CBL), University of Cambridge
Cambridge, UK, March 6, 2009

2008

Marco Huber
Probabilistic Framework for Sensor Management
Concluding presentation to project I4 - Research Training Group 1194 "Self-organizing Sensor-Actuator-Networks"
Bad Herrenalb, September 30, 2008

Marco Huber
Research Stay at ACFR, University of Sydney - Report -
Meeting - Research Training Group 1194 "Self-organizing Sensor-Actuator-Networks"
Karlsruhe, July 21, 2008

Marco Huber, Uwe D. Hanebeck
Gaussian Filter based on Deterministic Sampling for High Quality Nonlinear Estimation
17th IFAC World Congress
Seoul, Republic of Korea, July 10, 2008

Florian Weißel, Marco Huber, Dietrich Brunn, Uwe D. Hanebeck
Stochastic Optimal Control based on Value-Function Approximation using Sinc Interpolation
17th IFAC World Congress
Seoul, Republic of Korea, July 9, 2008

Marco Huber, Uwe D. Hanebeck
Priority List Sensor Scheduling using Optimal Pruning
The 11th International Conference on Information Fusion (Fusion 2008)
Cologne, July 1, 2008

Marco Huber, Uwe D. Hanebeck
Progressive Gaussian Mixture Reduction
The 11th International Conference on Information Fusion (Fusion 2008)
Cologne, July 1, 2008

Marco Huber, Florian Weissel, Uwe D. Hanebeck
Nonlinear Estimation and Control
Australian Centre for Field Robotics (ACFR), University of Sydney
Sydney, Australia, February 14, 2008

2007

Marco Huber, Uwe D. Hanebeck
On Sensor Scheduling in Case of Unreliable Communication
Finnish-German Co-operative Graduate School Network
Karlsruhe, November 27, 2007

Marco Huber, Eric Stiegeler, Uwe D. Hanebeck
On Sensor Scheduling in Case of Unreliable Communication
3rd German Workshop Sensor Data Fusion (SDF), 37th Annual Conference of the Gesellschaft für Informatik e.V. (GI)
Bremen, Germany, September 27, 2007

Marco Huber, Uwe D. Hanebeck
Optimal Sensor Deployment in Sensor-Actuator-Networks (Translation)
Retreat - Research Training Group 1194 "Self-organizing Sensor-Actuator-Networks"
Dagstuhl, September 18, 2007

Marco Huber, Dietrich Brunn, Uwe D. Hanebeck
Efficient Nonlinear Measurement Updating Based on Gaussian Mixture Approximation of Conditional Densities
2007 American Control Conference (ACC)
New York, New York, July 13, 2007

Marco Huber, Uwe D. Hanebeck
The Hybrid Density Filter for Nonlinear Estimation Based on Hybrid Conditional Density Approximation
The 10th International Conference on Information Fusion (Fusion 2007)
Quebec, Canada, July 10, 2007

Marco Huber, Uwe D. Hanebeck
Hybrid Transition Density Approximation for Efficient Recursive Prediction of Nonlinear Dynamic Systems
International Conference on Information Processing in Sensor Networks (IPSN)
Cambridge, Massachusetts, April 26, 2007

Uwe D. Hanebeck, Felix Sawo, Marco Huber
Information Processing in Sensor Networks for Model-Based
Reconstruction and Identification of Distributed Phenomena
ICRA 2007 Tutorial
Università di Roma "La Sapienza", Rom, April 10, 2007

Marco Huber, Uwe D. Hanebeck
Hybrid Density Filter (HDF)
Meeting - Research Training Group 1194 "Self-organizing Sensor-Actuator-Networks"
Karlsruhe, March 19, 2007

Felix Sawo, Marco Huber, Heiko Hamann
Information Processing in Sensor-Actuator-Networks (Translation)
Cooperation-Seminar of the Research Training Groups 1194 and Zeuss
Karlsruhe, January 8, 2007

2006

Uwe D. Hanebeck, Felix Sawo, Marco Huber
Model-Based Decentral Information Processing in Sensor-Actuator-Networks (Translation)
Lecture Series - Research Training Group 1194 "Self-organizing Sensor-Actuator-Networks"
Karlsruhe, December 8, 2006

Marco Huber, Uwe D. Hanebeck
Optimal Sensor Deployment in Sensor-Actuator-Networks (Translation)
Workshop - Research Training Group 1194 "Self-organizing Sensor-Actuator-Networks"
Dagstuhl, October 10, 2006

Marco Huber, Dietrich Brunn, Uwe D. Hanebeck
Closed-Form Prediction of Nonlinear Dynamic Systems by Means of Gaussian Mixture Approximation of the Transition Density
IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI 2006)
Heidelberg, September 4, 2006