초록 ▼

Control systems, apparatus, and methods can apply quantum algorithms to control a control object in the presence of uncertainty and/or information risk. A self-organizing controller can include a quantum inference unit that can generate a set of robust control gains for a controller that can meet th

Control systems, apparatus, and methods can apply quantum algorithms to control a control object in the presence of uncertainty and/or information risk. A self-organizing controller can include a quantum inference unit that can generate a set of robust control gains for a controller that can meet the control objectives for the particular realization of the control object. In one embodiment, the quantum inference unit can include a quantum correlator configured to generate a plurality of quantum states based on a plurality of controller parameters and a correlation type. In this embodiment, the quantum inference unit can also include a quantum optimizer configured to select the correlation type of the quantum correlator and to select a quantum state from the plurality of the quantum states. The self-organizing controller can control the control object with one or more controller gains that are based on the selected quantum state.

대표청구항 ▼

1. A method comprising: selecting a correlation type based on a result of a quantum genetic search algorithm;generating quantum states based on a plurality of controller parameters and the selected correlation type; andtuning a controller configured to control a control object based on at least one

1. A method comprising: selecting a correlation type based on a result of a quantum genetic search algorithm;generating quantum states based on a plurality of controller parameters and the selected correlation type; andtuning a controller configured to control a control object based on at least one of the generated quantum states. 2. The method of claim 1, wherein selecting comprises: simulating a closed-loop system to evaluate a candidate correlation type;evaluating a different candidate correlation type when the simulating indicates that fitness requirements are not satisfied; andreturning the candidate correlation type as the selected correlation type when the fitness requirements are satisfied. 3. The method of claim 1, wherein selecting is based on evaluating fitness functions associated with different correlation types. 4. The method of claim 1, wherein the correlation type comprises one of a spatial correlation, a temporal correlation, or a spatial-temporal correlation. 5. The method of claim 1, wherein the correlation type comprises an external correlation. 6. The method of claim 1, wherein the plurality of controller parameters comprise a set of controller gains. 7. The method of claim 6, wherein generating quantum states comprises: generating a set of candidate quantum states based on the set of controller gains; andperforming a correlation of the selected correlation type on the set of candidate quantum states from which a selected quantum state of the candidate quantum states is identified. 8. The method of claim 7, further comprising identifying a quantum state having the highest probabilities of the candidate quantum states as the selected quantum state. 9. The method of claim 7, further comprising identifying the selected quantum state by performing Grover's search algorithm. 10. The method of claim 1, further comprising identifying one of the generated quantum states as a selected quantum state, wherein tuning is based on the selected quantum state. 11. The method of claim 1, wherein the quantum genetic search algorithm comprises Grover's search algorithm. 12. The method of claim 1, wherein the generated quantum states each comprise at least one real state and at least one virtual state. 13. The method of claim 1, wherein generating the quantum states provides at least one quantum state associated with a higher probability than any linear combination of the controller parameters. 14. The method of claim 1, wherein the controller parameters are based on at least one information risk production rule. 15. The method of claim 1, further comprising generating two sets of controller gains with one or more proportional-integral-derivative (PID) controllers, wherein the plurality of controller parameters include the two sets of controller gains, and wherein the controller comprises a quantum PID controller. 16. A control system comprising: a quantum inference unit configured to provide one or more controller gains to a controller configured to control a control object, the quantum inference unit comprising: a quantum correlator configured to generate a plurality of quantum states based on a plurality of controller parameters and a correlation type; anda quantum optimizer configured to select the correlation type of the quantum correlator and to select a quantum state from the plurality of the quantum states,wherein the one or more controller gains are based on the selected quantum state. 17. The control system of claim 16, further comprising a knowledge base storing at least one of the plurality of candidate controller parameters, the knowledge base configured to provide one or more of the plurality of candidate controller parameters to the quantum inference unit. 18. The control system of claim 17, wherein the knowledge base stores one or more information risk production rules. 19. The control system of claim 16, wherein the quantum inference unit further comprises a quantum encoder configured to generate a plurality of candidate quantum states based on the at least one of the plurality of controller parameters and provide the quantum correlator with the candidate quantum states, wherein the quantum correlator is configured to generate the plurality of quantum states based on performing a correlation of the selected correlation type on the candidate quantum states. 20. The control system of claim 16, wherein the quantum inference unit further comprises a decoder configured to output the one or more controller gains based on decoding the selected quantum state. 21. The control system of claim 16, wherein the quantum optimizer comprises a quantum search unit configured to perform a superposition operation, an entanglement operation, and an interference operation on the at least one of the plurality of quantum states generated by the quantum correlator. 22. The control system of claim 16, wherein the quantum optimizer comprises a genetic algorithm unit configured to simulate a closed-loop system and select the selected quantum state based on evaluating a fitness function associated with the selected correlation type. 23. The control system of claim 16, wherein the correlation type comprises one of a spatial correlation, a temporal correlation, or a spatial-temporal correlation. 24. The control system of claim 16, further comprising the controller. 25. The control system of claim 24, wherein the controller comprises a proportional-integral-derivative (PID) controller. 26. The control system of claim 25, wherein the controller comprises a fractional PID controller. 27. The control system of claim 25, wherein the controller comprises a quantum PID controller configured to tune the control object based, and wherein the plurality of controller parameters are based on two sets of gains generated by two PID controllers. 28. The control system of claim 24, wherein the controller comprises a sliding mode controller. 29. A method comprising: generating quantum states based on a plurality of controller parameters;executing a quantum search algorithm on the generated quantum states to select a quantum state based on information risk; andcontrolling a control object based on an indicator of the selected quantum state. 30. The method of claim 29, wherein executing is performed online. 31. The method of claim 29, further comprising adjusting at least one of the controller parameters based on an information risk production rule. 32. The method of claim 29, wherein the information risk is associated with a current state of a control system that includes the control object. 33. The method of claim 32, wherein the controller parameters are not designed to control the control object in the current state of the control system. 34. The method of claim 29, wherein executing comprises: identifying the selected state when a risk termination condition is satisfied; andgenerating new quantum states when the risk termination condition is not satisfied. 35. The method of claim 29, further comprising selecting a correlation type based on a result of performing another quantum search algorithm, wherein generating is based on the selected correlation type. 36. The method of claim 35, wherein selecting is based on evaluating fitness functions associated with different correlation types, wherein the fitness functions incorporate information risk. 37. The method of claim 29, wherein the quantum search algorithm is a quantum genetic search algorithm. 38. A control system comprising: a quantum inference unit configured to provide one or more controller gains to a controller configured to control a control object, the quantum inference unit comprising: a quantum correlator configured to generate a plurality of quantum states based on a plurality of controller parameters; anda quantum optimizer configured to select a quantum state from the plurality of the quantum states based on information risk,wherein the one or more controller gains are indicative of the selected quantum state. 39. The control system of claim 38, further comprising a knowledge base storing the controller parameters and one or more information risk production rules, the knowledge base configured to provide the one or more controller parameters to the quantum inference unit, wherein the information risk comprises the one or more information production rules. 40. The control system of claim 39, wherein the controller parameters comprise controller gains, and wherein the knowledge base is configured to compute the controller gains provided to the quantum inference unit based on a risk estimation signal indicative of information risk. 41. The control system of claim 38, wherein the quantum genetic optimizer comprises a genetic algorithm unit configured to simulate a closed-loop system and select the selected quantum state based on evaluating a fitness function that incorporates information risk. 42. The control system of claim 38, wherein the quantum inference unit further comprises a quantum encoder configured to generate a plurality of candidate quantum states based on the controller parameters and provide the quantum correlator with the candidate quantum states, wherein the quantum correlator is configured to generate the plurality of quantum states based on performing a correlation on the candidate quantum states. 43. The control system of claim 38, wherein the quantum inference unit further comprises a decoder configured to decode the one or more controller gains from the selected quantum state. 44. The control system of claim 38, further comprising the controller. 45. The control system of claim 44, wherein the controller comprises one of a proportional-integral-derivative (PID) controller, a fractional PID controller, and a sliding mode controller. 46. The control system of claim 38, wherein the quantum inference unit is a quantum fuzzy inference unit configured for providing one or more controller gains to a fuzzy controller.