Abstract
〈Vol.2 No.1(2009.1)〉
Titles
▲
■ Computing the PageRank Variation for FragileWeb Data
Tokyo Institute of Technology・Hideaki ISHII,Politecnico di Torino・Roberto
TEMPO
Search engines utilize numerous measures to rank the webpages in the search
results. At Google, the so-called PageRank algorithm provides crucial information
by quantifying the importance of webpages. The PageRank values are considered
to be objective since the algorithm requires only the data of the web link
structure. In this paper, we first introduce the original PageRank algorithm
and give a brief overview of the distributed randomized approach recently
proposed by the authors. Then, we consider the problem of computing the
variation in the PageRank values that can be caused by erroneous data in
the web structure when some links are fragile. An efficient (centralized)
algorithm in the form of linear programming is proposed. This approach
employs results from the field of interval matrices. Numerical examples
are given to verify the effectiveness of the approach.
▲
■ A Realization-based Method for Identification of Closed-loop Systems using Measurable Disturbance
JFE Systems・Hajime Ase,Doshisha University・Tohru Katayama
We derive a closed-loop subspace identification method based on a realization
theory without using probing inputs, but we assume that there is a measurable
disturbance that can be used as a test input for closed-loop identification.
The orthogonal decomposition (ORT) technique is applied to obtain deterministic
and stochastic components of the joint input-output process.An ORT-based
closed-loop subspace identification method is then developed by combining
realizations of the deterministic and stochastic components, followed by
a model reduction procedure to obtain lower dimensional models. Some numerical
results are included to show the applicability of the present method.
▲
■ Scheduling Capacitated One-Way Vehicles on Paths with Deadlines
Kyoto Institute of Technology・Jun UCHIDA,Kyoto University・Yoshiyuki KARUNO,
and Kyoto University・Hiroshi NAGAMOCHI
In this paper, we deal with a scheduling problem of minimizing the number
of employed vehicles on paths. Let G=(V,E) be a path with a set V=\{v_{i}
\mid i=1,2,...,n} of vertices and a set E=\{\{v_{i},v_{i+1}\} \mid i=1,2,\ldots,n-1\}$
of edges. Vehicles with capacity $b$ are initially
situated at v_1. There is a job $i$ at each vertex $v_{i} \in V$, which
has its own handling time $h_{i}$ and deadline $d_{i}$. With each edge
$\{v_{i},v_{i+1}\} \in E$, a travel time $w_{i,i+1}$ is associated. Each
job is processed by exactly one vehicle, and the number of jobs processed
by a vehicle does not exceed the capacity $b$. A routing of a vehicle is
called {\em one-way} if the vehicle visits every edge $\{v_{i},v_{i+1}\}$
exactly once~(i.e., it simply moves from $v_{1}$ to $v_{n}$ on $G$). Any
vehicle is assumed to follow the one-way routing constraint. The problem
asks to find a schedule that minimizes the number of one-way vehicles,
meeting the deadline and capacity
constraints. A greedy heuristic is proposed, which repeats a dynamic programming
procedure for a single one-way vehicle problem of maximizing the number
of non-tardy jobs. We show that the greedy heuristic runs in $O(n^{3})$
time, and the approximation ratio is at most $\ln b + 1$.
▲
■ Precise Speed Tracking Control of Ultrasonic Motors via Sampled-Data
H∞ Control
Nagaoka University ofTechnology・Tharathip VONGSAROJ, Yasuhide KOBAYASHI,
Hisaya FUJIOKA, and Shigeo YANABE
In this paper, we consider a synthesis problem of digital controllerfor
speed tracking control systems of ultrasonic motors. In the previous work,
continuous-time H∞ control based synthesis involving Tustin transformation
has been utilized to solve the problem, where it has been empirically known
that over-estimation of modeling error is needed to avoid speed fluctuation
in high frequency range. In this paper, it is shown that desirable performance
can be obtained by sampled-data H∞ control synthesis without such conservativeness
in modelings.
▲
■ Mean Square Optimal Hedging with Non-Uniform Rebalancing Intervals
Kyoto University・Kenji Sato,Yuji Yamada,Hisaya Fujioka
This paper proposes a method of discrete hedging by extending the MSOH
(mean square optimal hedging) problem. In the proposed method, the risk
of option sellers is minimized in terms of the mean square hedging error
as in the MSOH problem,while it is allowed to rebalance with non-uniform
intervals as opposed to the uniform assumption in the MSOH problem. The
benefit of the extra freedom is demonstrated via numerical simulations.
▲
■ An iterative method for the output-feedback mixed H_2/H∞ control problem
Akashi National College of Technology・Yasushi Kami,
Kyushu Institute of Technology・Eitaku Nobuyama
This paper is concerned with the mixed H_2/H∞ control problem with output-feedback.
To propose a new iterative method for the problem we firstly construct
a controller sequence which converges to the unconstrained globally optimal
$H_2$ controller. After then, using this sequence, we propose an iterative
method for the mixed $H_2/H∞ control problem. The advantage of our method
is to guarantee that the obtained solution is either the globally optimal
solution or a feasible solution in the neighborhood of the boundary of
the H∞ norm constraint. A numerical example shows the efficiency of our
method.
▲ ■ Global Asymptotical Stabilization of Morse-Smale Systems using Weak
Control-Lyapunov Functions
RIKEN・Gou Nishida,Takayuki Tsuzuki,Hisakazu Nakamura,Yuh Yamashita
This paper proposes a method of constructing weak control-Lyapunov functions
for nonlinear systems by introducing a topological geometric assumption
called a Morse-Smale system.A Lyapunov function is one of the most important
tools to study stability and stabilization in nonlinear systems.
However, a general way of finding Lyapunov functions has not been found
yet.First, we confirm there is a weak Lyapunov function in Morse-Smale
systems.Next, we define the escapability for singular structures of the
weak Lyapunov function.If all singular structures are escapable, then the
Morse-Smale system is a global asymptotical stabilizable one.Finally, we
present the method of constructing a set of weak control-Lyapunov functions
to achieve global stabilization.The method is described in terms of a recursive
sequence of singular structures.We call the sequence a weak Lyapunov filtration.
▲
■ Motion Control of Inverted Pendulum Robots Using a Kalman Filter Based
Disturbance Observer}
Polytechnic University・Akira Shimada,Chaisamorn Yongyai
A high-speed motion control technique for inverted pendulum robots, utilizing
instability and a disturbance observer, based on the Kalman filtering technique,
is introduced. Inverted pendulums are basically controlled as they do not
topple. Shimada et al. developed a contrary idea and presented a controller
that deliberately off balanced the robot when it moved. To implement the
idea, a controller was designed using zero dynamics, which was derived
by partial feedback linearization. However, the control system was not
robust or sufficiently reliable. Although they presented a revised method
using H∞ control law, it was complex. Shimada et al. also presented a
design method for a disturbance observer using Kalman filtering. This paper
presents the latest control technique, combining both control laws to solve
the problem, and introduces an application with respect to inverted pendulum
robots. It further shows experimental results to confirm its validity.
▲ ■ Improved Particle Filter for Service Robot Self-Localization
Tokyo Institute of Technology・Guanghui Cen,Nobuto Matsuhira,Junko Hirokawa,Hideki
Ogawa,Ichiro Hagiwara,Yasumasa Fujisaki
Mobile robot localization is the problem of determining a robot'spose in
an environment, which is also the most basic problem in mobile robotics
application. Recently, particle filter becomes the most popular approach
in mobile robot localization and has been applied with great success to
a variety of state estimation problems. In this paper, the particle filter
is applied in our service robot position tracking and global localization.
Moreover, the posterior distribution of robot pose in global localization
is usually multi-model due to the symmetry of the environment and ambiguous
detected features. Considering these characteristics, we proposed an improved
cluster particle filter to increase the global localization robustness
and accuracy. On-line experiments based detailed analysis of coordinate
error and algorithm efficiency are given. On-line experiment results also
show the efficiency and robustness of our approach in our service robot
ApriAlpha$ ^{\textrm{TM}}$ Platform.
|
