Data processing: generic control systems or specific application – Generic control system – apparatus or process – Optimization or adaptive control
Reexamination Certificate
1999-03-04
2003-01-07
Grant, William (Department: 2121)
Data processing: generic control systems or specific application
Generic control system, apparatus or process
Optimization or adaptive control
C700S033000, C700S029000
Reexamination Certificate
active
06505085
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention is directed to a system for creating commands that increase the speed of system response. The invention can be applied to virtually any type of machine, process, or entity that can be modeled linearly. In particular, the invention deals with the creation of command profiles that are time-optimal, meaning commands that will move a system from one rest state to another rest state in the shortest amount of time. To this end, the invention provides a three-step procedure for generating time-optimal command profiles for all types of linear single-input, single-output systems. Constraint equations that result from this approach are both notationally and computationally simple. Consequently, the invention delivers results for complex systems faster and more reliably than traditional approaches. The invention furthermore incorporates additional practical constraints into its solution framework, thereby producing different types of time-efficient commands which satisfy a range of system performance requirements.
2. Description of the Related Art
With every passing year, the modern world demands that things change faster than they did before. In machines, processes, and attitudes, success is often defined by the speed with which these transitions can be made. In the information storage industry, for example, the speed of change is the primary performance metric that can make or break a product. In every computer disk drive or compact disk player, data storage and retrieval is performed by a read-write head that must scan rapidly from one disk track to another. The speed at which this read-write head can reposition itself between tracks is directly related to the speed of information access and the overall performance of the product. As another example of the importance of rapid change, the modern manufacturing environment contains many types of automated machines that are used for precision positioning, assembly, and inspection operations. The performance of these machines is measured largely by their ability to produce rapid and precise motions. Companies typically focus considerable attention toward improving the speed of these machines in order to maximize the throughput of the entire plant.
As with disk drives and manufacturing systems, it is not difficult to think of examples of automated machines that could benefit by an enhanced speed of operation. Furthermore, this rationale can be extended beyond mechanical hardware to many other kinds of systems. The pilot of a fighter jet, for example, would probably be very interested in the fastest way to change altitude if trying to avoid an oncoming missile. One could also imagine a stock broker working with a complicated financial model wondering how to quickly maximize profits by varying his investment portfolio. Alternately, the operators of a hydroelectric power plant might wish to increase the plant's power output as fast as possible to meet a sudden increase in demand. In all of these cases, as well as countless others, a knowledge of how to change things quickly would prove highly valuable.
Understanding how to move things faster can go a long way toward enhancing the performance of many systems. In some situations, however, this is not enough. Imagine a manufacturing robot moving so quickly that it tears itself from the factory floor. Or, how about a weather satellite that can scan the atmosphere with blazing speed but, in the process, exhausts its supply of fuel. One can imagine that the chairman of the Federal Reserve has more on his mind when adjusting interest rates than reducing inflation as rapidly as possible. Since he knows that his economic models are not perfect, he opts for a prudent schedule of interest rate adjustment in order to reliably and gradually improve economic indicators. In most applications, resting alongside the need for speed is such a set of practical issues that must also be addressed.
The modern world demands indeed that many systems change in a rapid and desirable manner. The question that still remains, however, is how to effect this change. As varied as the performance demands from system to system, so are the strategies for meeting these demands. A disk drive designer, for example, might choose to use a lighter material or a different type of actuator to reduce the access time of his drive. As an alternate approach, he might also decide to outfit his machine with sensors and develop a high-performance control system to meet more stringent performance specifications. These two strategies, hardware redesign and enhanced feedback control, are both viable methods for improving system performance. Many systems, however, do not tolerate these types of changes. An engineer at a hydroelectric plant, for example, would be foolhardy to consider building a new dam to better control the power output. Similarly, commissioning the Space Shuttle to install new sensors on an operating weather satellite might also prove equally absurd. In situations like these, engineers must focus their effort on improving performance by working with the system at hand. Even for systems that can be improved through hardware redesign and feedback control, the designer must then work with the improved system to further maximize performance.
The question, then, that remains is how to control the inputs to a given system to achieve a desired behavior at the system's output. For example, how can a crane operator best position a joystick to maneuver a cargo crate from a transport ship to a train platform as quickly as possible? What kind of torque profile should a designer specify in a disk drive actuator to most efficiently move the read-write head of the drive from track to track? How should the Fed vary the interest rate to quickly and reliably control inflation? The answer to each of these questions depends on many different factors. Understanding these factors and how they influence performance is the key to understanding how to answer these questions
1 The Feedforward Control Problem
The problem facing designers who wish to improve the performance of their systems, which can be called “the feedforward control problem”, addresses the question of how to select the input command for a given system that produces the most desirable output response. As illustrated in
FIG. 1
, the feedforward control problem can be stated as follows:
Given a known system, what is the input command that will move the output of the system from one state to another in the most desirable manner?
When stated in this form, it can be reasoned that the answer to this question will depend on four different factors. These four factors are as follows.
1. The first factor which influences the profile of the input is the nature of the change in the system output. For example, the actuator command that will best move a scanning sensor on a weather satellite over a ninety degree slew will probably have little resemblance to the command required to complete a three degree slew.
2. The second factor which influences the system response to an input command is the dynamic behavior of the system. In manufacturing systems, for example, this becomes apparent as machines are designed to be lighter and faster. As an unfortunate side-effect of these performance enhancements, undesirable machine dynamics often become the primary barrier to further performance improvements. For example, in many high-performance assembly systems, the speed with which a machine can operate is dictated not by the time required to execute positioning maneuvers, but rather by the time required for unwanted vibrations to settle. Consequently, the best input command profile for a given high-performance system should not ignore the dynamics of the system.
3. The third factor that should be considered when answering the foregoing question is the allowable level of the system inputs. One straightforward way to improve the speed of response in a manufacturing robot, for example, is to employ powerf
Seering Warren P.
Tuttle Timothy D.
Choate Hall & Stewart
Garland Steven R.
Grant William
Massachusetts Institute of Technology
LandOfFree
Method and apparatus for creating time-optimal commands for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for creating time-optimal commands for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for creating time-optimal commands for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3002423