초록 ▼

The invention comprises apparatuses and methods for providing the capability to stabilize and control a non-minimum phase, nonlinear plant with unmodeled dynamics and/or parametric uncertainty through the use of adaptive output feedback. A disclosed apparatus can comprise a reference model unit for

The invention comprises apparatuses and methods for providing the capability to stabilize and control a non-minimum phase, nonlinear plant with unmodeled dynamics and/or parametric uncertainty through the use of adaptive output feedback. A disclosed apparatus can comprise a reference model unit for generating a reference model output signal ym. The apparatus can comprise a combining unit that combines and differences a plant output signal y of a non-minimum phase plant for which not all of the states can be sensed, and a plant output signal y, to generate an output error signal {tilde over (y)}. The apparatus can further comprise an adaptive control unit for generating an adaptive control signal uad used to control the plant.

대표청구항 ▼

The invention claimed is: 1. An apparatus receiving a commanded output signal yc to control a plant, the apparatus having a nominal controller implementing a plant model to control a plant to generate plant output signal y that is a function of all states of the plant, at least one of the plant sta

The invention claimed is: 1. An apparatus receiving a commanded output signal yc to control a plant, the apparatus having a nominal controller implementing a plant model to control a plant to generate plant output signal y that is a function of all states of the plant, at least one of the plant states being unavailable, the nominal controller generating a nominal controller signal uec(y), the apparatus comprising: a reference model unit generating a plant model output signal ym based on the commanded output signal yc, using a reference controller identical to the nominal controller, and a plant model identical to that upon which the nominal controller was based; a first combining unit coupled to receive and difference the plant output signal y and the plant model output signal ym to generate the plant output error signal {tilde over (y)}; an adaptive control unit coupled to receive the plant output error signal {tilde over (y)}, and comprising an error observer to generate an observed error signal 횎 that estimates errors in all states of the plant, including those plate states that are unavailable, the adaptive control unit further comprising a neural network having connection weights M, N updated based on the observed error signal 횎, the neural network receiving a plant control signal u and the plant output signal y as inputs, and generating a neural network control signal uNN based on the plant control signal u, the plant output signal y, and the connection weights M, N, the adaptive control unit outputting an adaptive control signal uad based on the neural network control signal uNN; and a second combining unit coupled to receive the nominal controller signal uec(y) and the adaptive control signal uad, and generating the plant control signal u to control the plant based on the nominal controller signal uec(y) and the adaptive control signal uad. 2. An apparatus as claimed in claim 1 wherein the adaptive control unit comprises a time delay unit coupled to receive and delay the plant control signal u and the plant output signal y to generate a delayed plant control signal ud and a delayed plant output signal yd, respectively, the time delay unit coupled to provide the delayed plant control signal ud and the delayed plant output signal yd as inputs to the neural network for generation of the neural network control signal uNN. 3. An apparatus as claimed in claim 1 wherein the nominal controller generates a nominal controller state signal xc, and the reference controller generates a reference controller state signal xcm, and the adaptive control unit comprises a combining unit coupled to receive and difference the nominal controller state signal xc and the reference controller state signal xcm, to generate a controller state error signal xcm-xc for use by the error observer in generating the observed error signal 횎. 4. An apparatus as claimed in claim 3 wherein the adaptive control unit comprises an auxiliary controller coupled to receive the plant output signal y and the controller state error signal xcm-xc, and generating an auxiliary controller signal udc based on the plant output signal y and the controller state error signal xcm-xc, the adaptive control unit further comprising a third combining unit coupled to receive the auxiliary controller signal udc and the neural network control signal uNN to generate the adaptive control signal uad, the auxiliary controller further coupled to provide the auxiliary controller signal udc to the error observer for use in generating the observed error signal 횎. 5. An apparatus as claimed in claim 4 wherein the auxiliary controller is a linear controller. 6. An apparatus as claimed in claim 1 wherein the plant comprises at least one of an aircraft, spacecraft, missile, munition, satellite, robot, ground vehicle, underwater vehicle, surface ship, chemical process, industrial process, weather system, economic system, and any subsystem thereof. 7. An apparatus as claimed in claim 1 wherein the apparatus is used to control a non-minimum phase system. 8. An apparatus receiving a commanded output signal yc to control a plant, the apparatus for augmenting a nominal controller, the nominal controller generating a nominal controller signal uec(y) and a controller state signal xc based on the commanded output signal yc and the plant output signal y, the apparatus comprising: a reference model unit generating a reference model output signal ym and reference controller state signal xc based on the commanded output signal yc and a reference controller that is a replication of the nominal controller and a plant model upon which the nominal controller is based; and an adaptive control unit coupled to the nominal controller and the reference model unit, and generating an adaptive control signal uad based on the output error signal {tilde over (y)}, the controller state signal xc, and the reference controller state signal xcm, the adaptive control unit comprising a combining unit coupled to receive and difference the controller state signal xc and the reference controller state signal xcm to generate the controller state error signal xcm-xc; and an error observer coupled to receive the output error signal {tilde over (y)} and the controller state error signal xcm-xc and generating an observed error signal 횎 based the output error signal {tilde over (y)} and the controller state error signal xcm-xc; a time delay unit coupled to receive and delay the plant control signal u to generate a delayed plant control signal ud, and coupled to receive and delay the plant output signal y to generate a delayed plant output signal yd; a neural network coupled to receive the observed error signal 횎, the current plant control signal u, the delayed plant control signal ud, the current plant output signal y, and the delayed plant output signal yd, and generating a neural network output signal uNN for generation of the adaptive control signal uad, the plant controlled based on the nominal controller signal uec(y) generated by the nominal controller and the adaptive control signal uad generated by the adaptive control unit. 9. An apparatus as claimed in claim 8 wherein the reference model unit comprises a combining unit, a reference controller, and a plant model, the combining unit coupled to receive and difference the commanded output signal yc and the plant model output signal ym to generate reference controller input signal yc-ym supplied to the reference controller for use in generating the reference controller signal uec(ym) supplied to the plant model to generate the reference model output signal ym. 10. An apparatus as claimed in claim 9 wherein the nominal controller and the reference controller are identical and designed for controlling the plant according to the plant model. 11. An apparatus as claimed in claim 8 further comprising: a combining unit coupled to receive and difference the nominal controller output signal uec(y) and the adaptive control signal uad to generate an nominal controller signal u to control the plant. 12. An apparatus as claimed in claim 8 further comprising: a combining unit coupled to receive and difference the plant output signal y and the reference model output signal ym to generate an output error signal {tilde over (y)} supplied to the adaptive control unit for use in generating the adaptive control signal. 13. An apparatus as claimed in claim 8 wherein the adaptive control unit further comprises: an auxiliary controller coupled to receive the output error signal {tilde over (y)} and the controller state error signal xcm-xc, and generating a auxiliary control signal udc, the auxiliary controller coupled to provide the auxiliary controller signal udc to the error observer; and a fifth combining unit coupled to receive and difference the neural network output signal uNN and the auxiliary controller signal udc to generate the adaptive control signal uad. 14. An apparatus as claimed in claim 13 wherein the auxiliary controller is linear. 15. An apparatus as claimed in claim 8 further comprising: at least one actuator coupled to receive the plant control signal u and coupled to the plant, and controlling the plant based on the plant control signal u. 16. An apparatus as claimed in claim 15 wherein the actuator comprises at least one of an aerodynamic control surface, synthetic jet, vortex generator, reaction control jet, thrust vectoring nozzle, and reaction wheel. 17. An apparatus as claimed in claim 8 further comprising: at least one sensor coupled to the plant and generating the plant output signal y based on a sensed output of the plant. 18. An apparatus as claimed in claim 17 wherein the plant comprises an aircraft, and the sensed output comprises at least one of pitch, roll, and yaw attitudes, thrust, body angular rates, air speed, angle of attack, sideslip, angle of sideslip, position, velocity, and acceleration of the aircraft, and one or more components thereof. 19. An apparatus as claimed in claim 8 wherein the plant comprises at least one of an aircraft, spacecraft, missile, munition, satellite, robot, ground vehicle, underwater vehicle, surface ship, chemical process, industrial process, weather system, economic system, and any subsystem thereof. 20. An apparatus receiving a commanded output signal yc to control a plant, the apparatus comprising: a nominal controller generating a nominal controller output signal uec(y) and a controller state signal xc based on the commanded output signal yc and the plant output signal y; a reference model unit generating a reference model output signal ym and reference controller state signal xcm based on the commanded output signal yc and a replication of the nominal controller and a plant model upon which a control law of the nominal controller is based; and an adaptive control unit coupled to the nominal controller and the reference model unit, and generating an adaptive control signal uad based on the output error signal {tilde over (y)}, the controller state signal xc, and the reference controller state signal xcm, the adaptive control unit comprising a combining unit coupled to receive and difference the controller state signal xc and the reference controller state signal xcm to generate the controller state error signal xcm-xc; and an error observer coupled to receive the output error signal {tilde over (y)} and the controller state error signal xcm-xc and generating an observed error signal 횎 based the output error signal {tilde over (y)} and the controller state error signal xcm-xc; a time delay unit coupled to receive and delay the plant control signal u to generate a delayed plant control signal ud, and coupled to receive and delay the plant output signal y to generate a delayed plant output signal yd; a neural network coupled to receive the observed error signal 횎, the current plant control signal u, the delayed plant control signal ud, the current plant output signal y, and the delayed plant output signal yd, and generating a neural network output signal uNN for generation of the adaptive control signal uad, the plant controlled based on the nominal controller output signal uec(y) generated by the nominal controller and the adaptive control signal uad generated by the adaptive control unit. 21. An apparatus as claimed in claim 20 wherein the reference model unit comprises a combining unit, a reference controller, and a plant model, the combining unit coupled to receive and difference the commanded output signal yc and the plant model output signal ym to generate commanded output error signal yc-y supplied to the reference controller for use in generating the reference controller signal uec(ym) supplied to the plant model to generate the reference model output signal ym. 22. An apparatus as claimed in claim 21 wherein the nominal controller and the reference controller are identical and designed for controlling the plant according to the plant model. 23. An apparatus as claimed in claim 20 further comprising: a combining unit coupled to receive and difference the nominal controller output signal uec(y) and the adaptive control signal uad to generate an plant control signal u to control the plant. 24. An apparatus as claimed in claim 20 further comprising: a combining unit coupled to receive and difference the plant output signal y and the reference model output signal ym to generate an output error signal {tilde over (y)} supplied to the adaptive control unit for use in generating the adaptive control signal uad. 25. An apparatus as claimed in claim 20 wherein the adaptive control unit further comprises: an auxiliary controller coupled to receive the output error signal {tilde over (y)} and the controller state error signal xcm-xc, and generating an auxiliary control signal udc, the auxiliary controller coupled to provide the auxiliary control signal udc to the error observer; and a fifth combining unit coupled to receive and difference the neural network output signal NN and the auxiliary control signal udc to generate the adaptive control signal uad. 26. An apparatus as claimed in claim 25 wherein the auxiliary controller comprises a linear controller. 27. An apparatus as claimed in claim 20 further comprising: at least one actuator coupled to receive the plant control signal u and coupled to the plant, and controlling the plant based on the plant control signal u. 28. An apparatus as claimed in claim 27 wherein the actuator comprises at least one of an aerodynamic control surface, synthetic jet, vortex generator, reaction control jet, thrust vectoring nozzle, and reaction wheel. 29. An apparatus as claimed in claim 20 further comprising: at least one sensor coupled to the plant and generating the plant output signal y based on a sensed output of the plant. 30. An apparatus as claimed in claim 29 wherein the plant is an aircraft, and the sensed output comprises at least one of pitch, roll, and yaw attitudes, thrust, body angular rates, air speed, angle of attack, sideslip, angle of sideslip, position, velocity, and acceleration of the aircraft, and one or more components thereof. 31. An apparatus as claimed in claim 20 wherein the plant comprises at least one of an aircraft, spacecraft, missile, munition, satellite, robot, ground vehicle, underwater vehicle, surface ship, chemical process, industrial process, weather system, economic system, and any subsystem thereof. 32. An apparatus receiving a commanded output signal yc to control an plant, the apparatus comprising: a first combining unit coupled to receive and difference the commanded output signal yc and a plant output signal y to generate a commanded output error signal yc-y; an nominal controller coupled to receive a commanded output error signal yc-y, the nominal controller generating the nominal controller output signal uec(y) and a controller state signal xc based on the commanded output error signal yc-y; a reference model unit coupled to receive a commanded output signal yc, and having a reference controller and a plant model, the reference model unit generating a reference controller signal uec(ym) and a reference controller state signal xcm based on the commanded output signal yc and a reference model output signal ym of the plant model, the reference controller coupled to provide the reference controller signal uec(ym) to the plant model, the plant model generating the reference model output signal ym based on the reference controller signal uec(ym); a second combining unit coupled to receive and difference the reference controller signal uec(y) and an adaptive control signal uad to generate an plant control signal u, the second combining unit coupled to provide the plant control signal u to the plant, the plant generating the plant output signal y based on the plant control signal u; a third combining unit coupled to receive and difference the plant output signal y and the plant model output signal ym to generate an output error signal {tilde over (y)}; an adaptive control unit coupled to receive the output error signal {tilde over (y)}, the controller state signal xc, and the reference controller state signal xcm, and generating the adaptive control signal uad based on the output error signal {tilde over (y)}, the controller state signal xc, and the reference controller state signal xcm. 33. An apparatus as claimed in claim 32 wherein the nominal controller and the reference controller are identical. 34. An apparatus as claimed in claim 32 wherein the adaptive control unit comprises: a fourth combining unit coupled to receive and difference the controller state signal xc, and the reference controller state signal xcm to generate the controller state error signal xcm-xc; and an error observer coupled to receive the output error signal {tilde over (y)} and the controller state error signal xcm-xc and generating the observed error signal 횎 based the output error signal {tilde over (y)} and the controller state error signal xcm-xc; a time delay unit coupled to receive and delay the plant control signal u to generate a delayed plant control signal ud, and coupled to receive and delay the plant output signal y to generate a delayed plant output signal yd; a neural network coupled to receive the observed error signal 횎, the current plant control signal u, the delayed plant control signal ud, the current plant output signal y, and the delayed plant output signal yd, and generating a neural network output signal uNN for generation of the adaptive control signal uad. 35. An apparatus as claimed in claim 32 wherein the adaptive control unit further comprises: an auxiliary controller coupled to receive the output error signal {tilde over (y)} and the controller state error signal xcm-xc, and generating a auxiliary control signal udc, the auxiliary controller coupled to provide the auxiliary control signal udc to the error observer; and a fifth combining unit coupled to receive and difference the neural network output signal uNN and the auxiliary control signal udc to generate the adaptive control signal uad. 36. An apparatus as claimed in claim 35 wherein the auxiliary controller is linear. 37. An apparatus as claimed in claim 32 further comprising: at least one actuator coupled to receive the plant control signal u and coupled to the plant, the actuator controlling the plant based on the plant control signal u. 38. An apparatus as claimed in claim 37 wherein the actuator comprises at least one of an aerodynamic control surface, synthetic jet, vortex generator, reaction control jet, thrust vectoring nozzle, and reaction wheel. 39. An apparatus as claimed in claim 32 further comprising: at least one sensor coupled to the plant and generating the plant output signal y based on a sensed output of the plant. 40. An apparatus as claimed in claim 39 wherein the plant is an aircraft, and the sensed output comprises at least one of pitch, roll, and yaw attitudes, thrust, body angular rates, air speed, angle of attack, sideslip, angle of sideslip, position, velocity, and acceleration of the aircraft, and one or more components thereof. 41. An apparatus comprising: an auxiliary controller coupled to receive an output error signal {tilde over (y)} generated based on a combination of a plant output signal y and a reference model output signal ym, and a combination of the nominal controller state signal xc and the nominal controller state signal xcm, and generating an auxiliary controller signal udc; a first combining unit coupled to receive and difference the nominal controller state signal xc, and the reference controller state signal xcm to generate the controller state error signal xcm-xc; an error observer coupled to receive the output error signal {tilde over (y)} and a state error signal xcm-xc, and generating an observed error signal 횎 based on the output error signal {tilde over (y)} and a controller state error signal xcm-xc; a time delay unit receiving and delaying the plant control signal u and the plant output signal y to generate the delayed plant control signal ud and the delayed plant output signal yd; and a neural network coupled to receive the observed error signal 횎, the current plant control signal u, delayed plant control signal ud, current plant output signal y, and the delayed plant output signal yd, and generating a neural network control signal uNN based on the observed error signal 횎, current plant control signal u, the delayed plant control signal ud, current plant output signal y, and the delayed plant output signal yd, the neural network providing the neural network control signal uNN as the adaptive control signal uad used to control a plant. 42. An apparatus as claimed in claim 41 wherein the current plant control signal u, delayed plant control signal ud, the current plant output signal y, and the delayed plant output signal yd are input to neurons of the neural network and the observed error signal 횎 is used to adapt the connection weights M, N of the functions σ( ) of the neurons. 43. An apparatus as claimed in claim 41 wherein the auxiliary controller comprises a linear controller. 44. A method comprising the steps of: receiving a commanded output signal yc; receiving an plant output signal y; generating a nominal controller signal uec(y) and a controller state signal xc based on the commanded output signal yc and the plant output signal y; generating a reference controller signal uec(ym) and a reference controller state signal xcm based on the commanded output signal yc and the plant model output signal ym; generating the reference model output signal ym based on the reference controller signal uec(ym) and a plant model; generating an output error signal {tilde over (y)} based on the plant output signal y and the reference model output signal ym; generating an adaptive control signal uad based on the output error signal {tilde over (y)}, a plant control signal u, the nominal controller state signal xc, and the reference controller state signal xcm; generating a plant control signal u based on the nominal controller signal uec(y) and the adaptive control signal uad; controlling the plant based on the plant control signal u; generating a controller state error signal xcm-xc based on the controller state signal xc and the reference controller state signal xcm; generating the observed error signal 횎 based the output error signal {tilde over (y)} and the controller state error signal xcm-xc; delaying the plant control signal u to generate a delayed plant control signal ud; delaying the plant output signal y to generate a delayed plant output signal yd; and generating a neural network output signal uNN for generation of the adaptive control signal uad based on the observed error signal 횎, the current plant control signal u, the delayed plant control signal ud, the current plan output signal y, and the delayed plant output signal yd. 45. A method as claimed in claim 44 wherein the nominal controller signal uec(y) and the controller state signal xc are generated by a nominal controller. 46. A method as claimed in claim 45 wherein the reference controller signal uec(ym) and a reference controller state signal xcm are generated by a reference controller identical to the nominal controller, and the plant model upon which the design of the nominal controller was based. 47. A method as claimed in claim 44 further comprising the steps of: generating an auxiliary controller signal udc based on the output error signal {tilde over (y)} and the controller state error signal xcm-xc; providing the auxiliary controller signal udc to an error observer for use in generating the observed error signal 횎; and generating the adaptive control signal uad based on the neural network output signal uNN and the auxiliary controller signal udc.