MATH268 Hybrids

Arlene Grace B.
MATH268 Hybrids
Hybrid system theory, a field of applied mathematics abutting control theory and
computer science, has an enormous potential for impact on practical problems.
Hybrid
systems can be loosely defined as systems that allow the interaction of discrete
events
and continuous dynamics; hybrid system theory attempts to prove properties such
as
reachability and stability. Discrete event models naturally accommodate linguistic
and
qualitative information, and are used to model modes of operation of a single
system, for
example an aircraft or the interaction of several aircraft. The continuous dynamics in
a
hybrid system model physical processes, such as the continuous response of an
aircraft to
changes in the positions of aileron and throttle.
Hybrid systems are good models of complex reactive systems, in which physical
processes interact with man-made automated environments; algorithms developed
to
analyze and control the behavior of hybrid systems may therefore be used in the
design of
automatic controls for these systems. A common real-world example of a hybrid
system
arises when advanced automation is introduced into manually operated systems in
order
to enhance performance and flexibility while significantly reducing the workload of
human operators. Accompanying this increase in automation, however, is the
necessity
of ensuring that the automated system always performs as expected. This is
especially
crucial for safety-critical systems: if a telephone switch crashes or a power grid node
goes down, lives are usually not lost; if an error occurs in the automated avionics in a
commercial jet, the results could be disastrous.
Many of today's safety-critical systems are growing at a rate that will make their
manual operation extremely difficult if not impossible in the near future. The air
traffic
control system is an example of such a system. Air traffic in the United States is
expected to grow by 5% annually for the next 15 years, and rates of growth across
the
Pacific Rim are expected to be more than 15% a year. Even with today's traffic,
ground
holds and airborne delays in flights due to congestion have become so common that
airlines pad their flight times with built-in allowances. Aging air traffic control
equipment certainly contributes to these delays: the plan view displays used by
controllers to look at radar tracks and flight information are the very same ones that
were
installed in the early 1970s, and they fail regularly. The computer systems that
calculate
radar tracks and store flight plans were designed in the 1980s, using software written
in
1972.
The introduction of new computers, display units, and communication technologies
for air traffic controllers will help alleviate the problems caused by failing equipment,
yet
the Federal Aviation Administration admits that any significant improvement will
require
that many of the basic practices of air traffic control be automated. For example,
today's
airspace has a rigid route structure based on altitude and on ground-based
navigational
“fixes”. The current practice of air traffic controllers is to route aircraft along
predefined
paths connecting fixes, to manage the complexity of route planning for several
aircraft at
once. The rigid structure puts strict constraints on aircraft trajectories, which could
otherwise follow wind-optimal or “user-preferred” routes (routes that are shorter or
involve lower fuel consumption because of tailwinds). Also, while a data link
between
aircraft and ground is being considered as a replacement for the current voice
communication over radio channels between pilot and controller, there is a limit to
the
amount of information processing that a controller can perform with these data.
Recent
studies indicate that, if there is no change to the structure of air traffic control, then
by the
year 2015 there could be a major accident every 7 to 10 days; obviously this cannot
be
permitted to happen.
The main goal of air traffic control is to maintain safe separation between aircraft
while guiding them to their destinations. However, its tight control over the motion
of
every aircraft in the system frequently causes bottlenecks to develop. Uncertainties
in
positions, velocities, and wind speeds, along with the inability of a single controller
to
handle large numbers of aircraft at once, lead to overly conservative controller
actions
and procedures. An example is the set of methods used by air traffic controllers to
predict and avoid conflicts between aircraft. If a controller predicts that the
separation
between two aircraft will become less than the regulatory separation, the controller
will
issue a directive to one or both of the pilots to alter their paths, speed, or both.
Often the
resolution is not needed, and usually it is too drastic. Also, user-preferred routes are
disallowed because of the requirement that prescribed jetways be used.
As a result of all these difficulties, there is a widely perceived need in the air traffic,
airline, and avionics communities for an architecture that integrates data storage,
processing, communications, and display into a safe and efficient air traffic
management
system; a new air traffic system has been proposed that involves the Global
Positioning
System and a datalink communication protocol called Automatic Dependent
Surveillance
for aircraft-aircraft and aircraft-ground communication. While the degree of
decentralization and level of automation in such a system are still under debate, the
integrity of any automated functionality in a new air traffic management system
depends
on a provably safe design as well as high confidence that the control actions will not
fail.
This level of reliability requires accurate models, techniques for verifying that the
design is safe to within the accuracy of these models, and procedures for
synthesizing the
system's control actions. Hybrid system researchers have designed models and
control
laws for two systems: a provably safe algorithm for resolving trajectory conflicts
between
aircraft, and a provably safe algorithm for a single aircraft to switch between
different
flight modes. A rigorous notion of “safety” in each case is crucial. In the conflict
resolution problem, the system is safe if the aircraft always maintain minimum
separation
from each other. In the mode-switching problem, system safety means that the state
of
the aircraft remains within minimum and maximum bounds imposed on its
velocities,
angles, etc., so that the aircraft does not stall and plunge out of the skyave been
continuing, rapid
gains in the achievable spatial resolution---for example, an eight-fold improvement
between 1996 and 1997. Most of the gains in speed and accuracy are attributable to
better mathematical algorithms, not to increased computational power.
The cognitive science driving fMRI has also advanced; one interesting discovery was
that reading more complex sentences causes greater brain activity in precisely the
ways
predicted by theory and earlier, more primitive external measurements of eye
movements. In ongoing projects, fMRI is being used to study the cognitive and brain
activity characteristics of high-functioning autistic subjects, and to examine brain
plasticity and rehabilitation in aphasia therapy.
One final point of interest is that certain aspects of the statistical techniques
developed
in the context of fMRI generalize to analysis of seismic data collected by
geophysicists in
oil exploration.

MATH268 Hybrids

Uploaded by

Copyright:

Available Formats

MATH268 Hybrids

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MATH268 Hybrids

Uploaded by

Copyright:

Available Formats

Arlene Grace B.

You might also like