Data processing: artificial intelligence – Fuzzy logic hardware
Reexamination Certificate
1999-02-23
2002-08-27
Starks, Jr., Wilbert L. (Department: 2121)
Data processing: artificial intelligence
Fuzzy logic hardware
C706S003000, C700S037000, C700S041000, C700S044000
Reexamination Certificate
active
06442534
ABSTRACT:
CROSS REFERENCE TO RELATED APPLICATIONS
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
BACKGROUND OF THE INVENTION
The invention relates to feedback controllers for hydraulic and pneumatic actuators and in particular to a control system employing multiple feedback controllers each tuned to different operating conditions of the actuator and combined using fuzzy logic.
Feedback controllers, such as those which may be used to control an hydraulic or pneumatic actuator, accept a reference signal (i.e., a position command indicating the desired position of the actuator) and receive a feedback signal (i.e., a signal indicating the actual position of the actuator). From these signals, the feedback controller produces a command signal to the actuator calculated to bring the feedback signal into closer agreement with the reference input. For hydraulic and pneumatic actuators, the reference signal may be either position or force.
In a “proportional-integral-derivative” (PID ) controller, a feedback signal is subtracted from the reference input to produce an error signal. Three signals are generated from the error signal: (1) a signal that is proportional to this error signal, (2) a signal that is the integral of this error signal, and (3) a signal that is the derivative of this error signal. Each of these three signals is given a different weight and then summed.
As is well understood in the art, the weighting of the proportional, integral and derivative signals is adjusted by the user to tune the PID controller to provide different control characteristics having benefits for different operating regimes of the actuator. For example, the controller may be tuned to produce reduced steady state error, reduced overshoot, reduced response time or the like. If the transfer function of the physical system to be controlled is well-known and may be approximated by a linear system, the approximate P, I and D gain factors may be calculated according to desired tradeoffs by a number of well-known methods. Inevitably, however, such tuned PID controllers compromise between the particular goals, for example, response time and overshoot.
Further, the transfer function of the physical system may not be well-known, as a result for example, of changing the load connected to the actuator. Practical tuning of the feedback controller in this case necessarily is limited to a typical or average transfer function.
Fuzzy logic is a well-known technique for controlling mathematically illunderstood processes. In fuzzy logic, input quantities are mapped to several fuzzy states defined by overlapping membership functions. If the input quantity is pressure, for example, the membership function might map it to the fuzzy states of: “low pressure”, “medium pressure” and “high pressure” to varying degrees. Fuzzy rules approximating rules applied by human experts may then be applied to fuzzy states to map the characterized inputs to output states. Competing output states resulting from the overlap of the membership functions and the rules are then combined in a third “defuzzifying” step according to one of several methodologies. A common method of combining output membership functions is to find their center of mass.
BRIEF SUMMARY OF THE INVENTION
The present invention employs the power of fuzzy logic in combining competing output values, to combine the output values of different feedback controllers each tuned to a different operating regime of an actuator. In this way separately tuned feedback controllers may address not only different operating regimes of the actuator but may also accommodate a range of operating conditions, for example, loads or the like. Individual and separately tunable feedback controllers allowed the overall control strategy to be built up using intuitive and well understood parts.
Specifically the present invention provides a control system for an actuator of a type receiving a command signal causing actuation of the hydraulic actuator and providing at least one feedback signal indicating resulting actuation of the hydraulic actuator. The control system includes a first feedback controller receiving a first feedback signal and tuned for first operating condition of the actuator and providing a first output signal, and a second feedback controller receiving a second feedback signal and tuned for a second operating condition of the actuator different from the first operating condition and providing a second output signal. A fuzzy logic circuit receives and combines the first and second output signals according to fuzzy logic rules to produce the command output for the hydraulic actuator.
Thus, it is one object of the invention to allow multiple, low-complexity feedback controllers to be combined to operate an actuator under different regimes for which it would be hard to tune a single feedback controller. The fuzzy logic rules may combine the first output signal and the second output signal based on their relative values. This principle can be extended to outputs of more than two controllers and/or combinations of outputs of individual controllers.
Thus it is another object of the invention to provide for a combination of outputs of multiple feedback controllers using rules based solely on the output from the feedback controller and not requiring access to other information.
The outputs of the first and second feedback controllers may be formed into a ratio used to combine the first and second outputs into a control output. More specifically, the combination may be such that the lesser of the first and second outputs has the greater weight in the combination.
Thus it is another object of the invention to incorporate in the combination an overarching rule that the feedback controller, providing for a lesser magnitude of control output to the actuator, predominates in the control process. This rule reflects the observation that the better-tuned feedback controller for the particular operating condition requires least perturbation of the actuator.
The first feedback controller and the second feedback controller may be proportional-integral-derivative controllers and the fuzzy logic rules may disable the integral portion of a given feedback controller when the contribution of the output of the given feedback controller is less than a predetermined amount of the command signal.
Thus it is another object of the invention to prevent “windup” of the controller whose output is not used such as would cause a continued error signal to increase the amount of the integral term of that controller. When the particular controller is de-emphasized in the control equation, its integral term is likewise disabled or reduced.
The foregoing and other objects and advantages of the invention will appear from the following description. In the description, reference is made to the accompanying drawings which form a part hereof and in which there is shown by way of illustration a preferred embodiment of the invention. Such embodiment does not necessary represent the full scope of the invention, however, and reference must be made to the claims herein for interpreting the scope of the invention.
REFERENCES:
patent: 4362978 (1982-12-01), Pollard et al.
patent: 5159547 (1992-10-01), Chand
patent: 5394678 (1995-03-01), Lonn et al.
patent: 5497063 (1996-03-01), Day et al.
patent: 5557154 (1996-09-01), Erhart
patent: 5566065 (1996-10-01), Hansen et al.
patent: 5570282 (1996-10-01), Hansen et al.
patent: 5822740 (1998-10-01), Haissig et al.
patent: 5867456 (1999-02-01), Tsai et al.
Jackson, A., Fuzzy logic vs traditional approaches to the design of microcontroller-based systems, Aerospace Applications Conference, 1994. Proceedings., 1994 IEEE, Feb. 5-12, 1994, pp.: 19-33.*
Malki, H.A.; Huaidong Li; Guanrong Chen, New design and stability analysis of fuzzy proportional-derivative control systems, Fuzzy Systems, IEEE Transactions on, vol.: 2 4, Nov. 1994, pp.: 245-254.*
Thaler, George J. et al, Servomechanism Analysis, Mcgraw-Hill Book Company, Inc., New York, 1953, pp. 337-349. Feb. 1994.
Au Burton K.
Horacek Petr
Kes Miroslav
Sladek Bohumir
Forman Adam J.
Gerasimow Alexander M.
Rockwell Automation Technologies Inc.
Starks, Jr. Wilbert L.
Walbrun William R.
LandOfFree
Hydraulic controller using multiple regime specific... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Hydraulic controller using multiple regime specific..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Hydraulic controller using multiple regime specific... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2973985