초록
In one example, a method includes receiving, over an aircraft data communications bus, a plurality of non-pneumatic inputs corresponding to aircraft operational parameters. The method further includes processing the plurality of non-pneumatic inputs through an artificial intelligence network to generate an air data output value, and outputting the air data output value to a consuming system for use when a pneumatic-based air data output value is determined to be unreliable.
In one example, a method includes receiving, over an aircraft data communications bus, a plurality of non-pneumatic inputs corresponding to aircraft operational parameters. The method further includes processing the plurality of non-pneumatic inputs through an artificial intelligence network to generate an air data output value, and outputting the air data output value to a consuming system for use when a pneumatic-based air data output value is determined to be unreliable.
대표
청구항
▼
1. A method comprising: receiving, over an aircraft data communications bus, a plurality of non-pneumatic inputs corresponding to aircraft operational parameters;processing the plurality of non-pneumatic inputs through an artificial intelligence network to generate an air data output value; andoutputting the air data output value to a consuming system for use when a pneumatic-based air data output value is determined to be unreliable;wherein the artificial intelligence network comprises an artificial neural network having at least one internal layer of n...
1. A method comprising: receiving, over an aircraft data communications bus, a plurality of non-pneumatic inputs corresponding to aircraft operational parameters;processing the plurality of non-pneumatic inputs through an artificial intelligence network to generate an air data output value; andoutputting the air data output value to a consuming system for use when a pneumatic-based air data output value is determined to be unreliable;wherein the artificial intelligence network comprises an artificial neural network having at least one internal layer of neurons that apply one or more weights, biases, or transfer functions to each of the plurality of non-pneumatic inputs to generate the air data output value;wherein the artificial neural network is pre-trained to determine the one or more weights, biases, or transfer functions; andwherein processing the plurality of non-pneumatic inputs through the artificial intelligence network to generate the air data output value comprises processing the plurality of non-pneumatic inputs through the artificial neural network without changing the one or more weights, biases, or transfer functions. 2. The method of claim 1, wherein the plurality of non-pneumatic inputs comprise one or more of an aircraft engine thrust parameter, an aircraft engine throttle setting, a flight control surface position, a flight control surface loading, an aircraft fuel usage rate, an aircraft weight, a landing gear position, an aircraft mass balance, an aircraft acceleration, and an aircraft angular rate. 3. The method of claim 1, wherein the generated air data output value is selected from a group comprising an aircraft calibrated airspeed, an aircraft true airspeed, an aircraft Mach number, an aircraft pressure altitude, an aircraft angle of attack, an aircraft vertical speed, and an aircraft angle of sideslip. 4. The method of claim 1, wherein the artificial neural network is a feed-forward neural network. 5. The method of claim 1, further comprising: receiving the pneumatic-based air data output value from a pneumatic-based air data system; andidentifying whether the received pneumatic-based air data output value is determined to be reliable or whether the received pneumatic-based air data output value is determined to be unreliable;wherein processing the plurality of non-pneumatic inputs through the artificial intelligence network to generate the air data output value further comprises: processing the non-pneumatic inputs and the received pneumatic-based air data output value through the artificial intelligence network to generate the air data output value when the received pneumatic-based air data output value is determined to be reliable; andprocessing the non-pneumatic inputs without the received pneumatic-based air data output value through the artificial intelligence network to generate the air data output value when the received pneumatic-based air data output value is determined to be unreliable. 6. The method of claim 1, further comprising: outputting the air data output value to a consuming system that determines whether the pneumatic-based air data output value is unreliable based at least in part on the generated air data value. 7. The method of claim 1, further comprising: determining whether the pneumatic-based air data output value is unreliable. 8. A synthetic air data system comprising: at least one processor; andnon-transitory computer-readable memory encoded with instructions that, when executed by the at least one processor, cause the synthetic air data system to: receive, over an aircraft data communications bus, a plurality of non-pneumatic inputs corresponding to aircraft operational parameters;process the plurality of non-pneumatic inputs through an artificial intelligence network to generate an air data output value, wherein the artificial intelligence network is a pre-trained artificial neural network having at least one internal layer of neurons that apply one or more weights, biases, or transfer functions to each of the plurality of non-pneumatic inputs to generate the air data output value without changing the one or more weights, biases, or transfer functions; andoutput the air data output value to a consuming system for use when a pneumatic-based air data output value is determined to be unreliable. 9. The synthetic air data system of claim 8, wherein the plurality of non-pneumatic inputs comprise one or more of an aircraft engine thrust parameter, an aircraft engine throttle setting, a flight control surface position, a flight control surface loading, an aircraft fuel usage rate, an aircraft weight, a landing gear position, an aircraft mass balance, an aircraft acceleration, and an aircraft angular rate. 10. The synthetic air data system of claim 8, wherein the computer-readable memory is encoded with instructions that, when executed by the at least one processor, cause the synthetic air data system to process the plurality of non-pneumatic inputs through the artificial intelligence network to generate the air data output value that is selected from a group comprising an aircraft calibrated airspeed, an aircraft true airspeed, an aircraft Mach number, an aircraft pressure altitude, an aircraft angle of attack, an aircraft vertical speed, and an aircraft angle of sideslip. 11. The synthetic air data system of claim 8, wherein the computer-readable memory is further encoded with instructions that, when executed by the at least one processor, cause the synthetic air data system to: receive the pneumatic-based air data output value from a pneumatic-based air data system; andidentify whether the received pneumatic-based air data output value is determined to be reliable or whether the received pneumatic-based air data output value is determined to be unreliable; andwherein the computer-readable memory is encoded with instructions that, when executed by the at least one processor, cause the synthetic air data system to process the plurality of non-pneumatic inputs through the artificial intelligence network to generate the air data output value by at least causing the synthetic air data system to: process the non-pneumatic inputs and the received pneumatic-based air data output value through the artificial intelligence network to generate the air data output value when the received pneumatic-based air data output value is determined to be reliable; andprocess the non-pneumatic inputs without the received pneumatic-based air data output value through the artificial intelligence network to generate the air data output value when the received pneumatic-based air data output value is determined to be unreliable. 12. The synthetic air data system of claim 8, wherein the computer-readable memory is further encoded with instructions that, when executed by the at least one processor, cause the synthetic air data system to output the air data output value to a consuming system that determines whether the pneumatic-based air data output value is unreliable based at least in part on the generated air data value. 13. The synthetic air data system of claim 8, wherein the computer-readable memory is further encoded with instructions that, when executed by the at least one processor, cause the synthetic air data system to determine whether the pneumatic-based air data output value is unreliable. 14. A method comprising: receiving, over an aircraft data communications bus, a plurality of non-pneumatic inputs corresponding to aircraft operational parameters;processing the plurality of non-pneumatic inputs through an artificial intelligence network to generate an air data output value; andoutputting the air data output value to a consuming system for use when a pneumatic-based air data output value is determined to be unreliable;wherein the artificial intelligence network comprises an artificial neural network having at least one internal layer of neurons that apply one or more weights, biases, or transfer functions to each of the plurality of non-pneumatic inputs to generate the air data output value;wherein the artificial neural network is pre-trained to determine the one or more weights, biases, or transfer functions; andwherein the pre-trained artificial neural network modifies the one or more weights, biases, or transfer functions based on the plurality of non-pneumatic inputs corresponding to the aircraft operational parameters. 15. A method comprising: receiving, over an aircraft data communications bus, a plurality of non-pneumatic inputs corresponding to aircraft operational parameters;receiving a pneumatic-based air data output value corresponding to the air data output value from a pneumatic-based air data system;identifying whether the received pneumatic-based air data output value is determined to be reliable or whether the received pneumatic-based air data output value is determined to be unreliable;processing the non-pneumatic inputs and the received pneumatic-based air data output value through an artificial intelligence network to generate an air data output value when the received pneumatic-based air data output value is determined to be reliable;processing the non-pneumatic inputs without the received pneumatic-based air data output value through the artificial intelligence network to generate the air data output value when the received pneumatic-based air data output value is determined to be unreliable; andoutputting the air data output value to a consuming system for use when the pneumatic-based air data output value is determined to be unreliable. 16. A synthetic air data system comprising: at least one processor; andnon-transitory computer-readable memory encoded with instructions that, when executed by the at least one processor, cause the synthetic air data system to: receive, over an aircraft data communications bus, a plurality of non-pneumatic inputs corresponding to aircraft operational parameters;process the plurality of non-pneumatic inputs through an artificial intelligence network to generate an air data output value, wherein the artificial intelligence network is a pre-trained artificial neural network having at least one internal layer of neurons that apply one or more weights, biases, or transfer functions to each of the plurality of non-pneumatic inputs to generate the air data output value by modifying the one or more weights, biases, or transfer functions based on the plurality of non-pneumatic inputs corresponding to the aircraft operational parameters; andoutput the air data output value to a consuming system for use when a pneumatic-based air data output value is determined to be unreliable. 17. A synthetic air data system comprising: at least one processor; andnon-transitory computer-readable memory encoded with instructions that, when executed by the at least one processor, cause the synthetic air data system to: receive, over an aircraft data communications bus, a plurality of non-pneumatic inputs corresponding to aircraft operational parameters;receive a pneumatic-based air data output value from a pneumatic-based air data system;identify whether the received pneumatic-based air data output value is determined to be reliable or whether the received pneumatic-based air data output value is determined to be unreliable;process the non-pneumatic inputs and the received pneumatic-based air data output value through the artificial intelligence network to generate an air data output value when the received pneumatic-based air data output value is determined to be reliable;process the non-pneumatic inputs without the received pneumatic-based air data output value through the artificial intelligence network to generate the air data output value when the received pneumatic-based air data output value is determined to be unreliable; andoutput the air data output value to a consuming system for use when the pneumatic-based air data output value is determined to be unreliable.
