Controller reconfiguration for nonlinear systems using composite observer

Thomas, Suba; Chang, B.C.; Kwatny, Harry G.

doi:10.1109/acc.2003.1242478

Cited by 7 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Assuming the possible faults are known a priori , the passive method takes into account of all these possible faults in the design stage and does not change the controller when the fault occurs 2–7. An active method usually uses a fault detection and isolation (FDI) unit to collect the fault information and changes the controller according to the fault 8–28 . Intuitively, the active method uses more information, thus, can achieve better performance than the passive method.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fault tolerant control of feedback linearizable systems with stuck actuators

Wang

Zhou

2008

Asian Journal of Control

View full text Add to dashboard Cite

An ordinary Petri-net (PN) based approach is proposed to design the programmable logical controller (PLC) for preventing collisions between vehicles in an automated guided vehicles (AGV) system. First, method is proposed to model an AGV system as an ordinary PN, called the plant PN. Second, for collision prevention in an AGV system, module control methods are proposed to design the PN supervisor (the closed-loop PN) by augmenting the plant PN. In detail, three modules are defined the line, divide, and merge modules, in the plant PN, and the control module methods are presented for each of these three modules. As a result, the closed-loop PN of an AGV system is obtained and can be analyzed using the PN theory. Finally, method is proposed to translate a closed-loop PN into a ladder diagram. Consequently, the PN supervisor is implemented by a PLC using these proposed methods. A laboratory AGV system is used to illustrate this approach.

show abstract

Section: Introductionmentioning

confidence: 99%

“…Intuitively, the active method uses more information, thus, can achieve better performance than the passive method. One typical active FTC system widely used in the literature contains a normal controller, a FDI unit, and a fault estimation and controller redesign module 8, 9, 11–15, 19, 20, 23–28 as illustrated in Fig. 1.…”

Section: Introductionmentioning

confidence: 99%

Fault tolerant control of feedback linearizable systems with stuck actuators

Wang

Zhou

2008

Asian Journal of Control

View full text Add to dashboard Cite

show abstract

“…However, they assume the existence of an FDI mechanism which manages to identify the failures. Thomas et al 10 have developed a reconfigurable controller based on nonlinear output regulation for nonlinear systems. They have used a composite observer which estimates the states as well as failed actuator signal in the event of actuator failures.…”

Section: Introductionmentioning

confidence: 99%

Online Sensor/Actuator Failure Isolation and Reconfigurable Control Using the Generalized Likelihood Ratio Method

Deshpande

Zamad

Patwardhan

2008

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

In processing plants, sensor and/or actuator failures can have considerable deteriorating effect on the closedloop performance. Such failures have to be diagnosed online, as quickly as possible, and actively accommodated to arrest the performance degradation. Active failure tolerance can be achieved by employing model-based failure diagnosis techniques and redesigning/restructuring controller online. In this work, a sensor/actuator failure isolation strategy has been developed under the linear generalized likelihood ratio (GLR) framework. The strategy is then extended to isolation of sensor and actuator failures in nonlinear systems. The infomation on sensor/actuator failures is further used for online reconfiguration of the state estimator and the controller/ control scheme. In case of sensor failure, the state estimator is reconfigured by removing the measurement of failed sensor from the measurement vector. If an observability property is preserved after sensor failure, then an inferential control scheme is employed subsequent to the failure. When an actuator failure is isolated, it is proposed to make modifications in the controller objectives or switch to a new controller to account for the loss of a degree of freedom. The efficacy of the proposed failure diagnosis and control structure reconfiguration schemes is demonstrated by conducting experimental studies on a benchmark heater mixer set up. Analysis of these results reveals that the proposed strategies are able to isolate the failures accurately and recover the closed-loop performance by online reconfiguration of the controller/control scheme.

show abstract

“…We first presented the idea of using multiple switching controllers and employing the servomechanism 15-19 and 2 H control theory [20][21] to address the actuator jam problem in Chang, Bajpai, and Kwatny 22 , and later extended the scope to more general issues [23][24][25][26][27] including nonlinearities, asymmetric failures, recoverability, diversified 3 redundancy, etc. In our proposed approach, we considered the nominal and all foreseeable actuator jam scenarios and their associated plants, 0 G , 1 G , … G l .…”

Section: Introductionmentioning

confidence: 99%

Aircraft Loss-of-Control Accident Prevention: Switching Control of the GTM Aircraft with Elevator Jam Failures

Chang

Kwatny

Belcastro

2008

AIAA Guidance, Navigation and Control Conference and Exhibit

View full text Add to dashboard Cite

Switching control, servomechanism, and H2 control theory are used to provide a practical and easy-to-implement solution for the actuator jam problem. A jammed actuator not only causes a reduction of control authority, but also creates a persistent disturbance with uncertain amplitude. The longitudinal dynamics model of the NASA GTM UAV is employed to demonstrate that a single fixed reconfigured controller design based on the proposed approach is capable of accommodating an elevator jam failure with arbitrary jam position as long as the thrust control has enough control authority. This paper is a first step towards solving a more comprehensive in-flight loss-of-control accident prevention problem that involves multiple actuator failures, structure damages, unanticipated faults, and nonlinear upset regime recovery, etc.

show abstract

Controller reconfiguration for nonlinear systems using composite observer

Cited by 7 publications

References 13 publications

Fault tolerant control of feedback linearizable systems with stuck actuators

Fault tolerant control of feedback linearizable systems with stuck actuators

Online Sensor/Actuator Failure Isolation and Reconfigurable Control Using the Generalized Likelihood Ratio Method

Aircraft Loss-of-Control Accident Prevention: Switching Control of the GTM Aircraft with Elevator Jam Failures

Contact Info

Product

Resources

About