Flight management via model-based iterative optimization

{
  "publication_number": "US10832581B2",
  "country": "US",
  "kind": "B2",
  "title": "Flight management via model-based iterative optimization",
  "abstract": "A system, computer-readable medium, and a method including obtaining flight data for a specific aircraft for a prescribed flight; obtaining current sample measurements of at least one state or output of the specific aircraft; performing based on the obtained flight data, the current measurements or outputs, and a mathematical model accurately representing an actual operational performance of the specific aircraft and providing a predictive indication of a future performance of the specific aircraft, a control optimization to determine a cost-optimal control input for the prescribed flight; adjusting, in response to a consideration of actual operational characteristics of the specific aircraft, the optimized control input; and transmitting the adjusted optimized control input to the specific aircraft to operate the specific aircraft to minimize the direct operating cost for the prescribed flight.",
  "claims": [
    "1. A method implemented by a processor of a computing system to optimize aircraft guidance to minimize direct operating cost of a prescribed flight, the method comprising: obtaining flight data including a flight specification and other flight related data for a specific aircraft for a future prescribed flight, the flight specification including at least flight constraints, a starting location, a destination location, and transient performance limits of the specific aircraft; and the other flight related data including future atmospheric conditions the specific aircraft will be subjected to when employing control commands generated by the processor and executing the future prescribed flight, and air traffic control information relevant to the future prescribed flight; generating control histories based on historical performances and generating corresponding state trajectories for the specific aircraft based on the flight data and performing model predictive control to predict a future performance of the specific aircraft and adjusting current control input actions to further control the specific aircraft to perform in an optimized manner during the future prescribed flight; obtaining current sample measurements of at least one state or output of the specific aircraft during execution of the future prescribed flight; performing, a control optimization by a processor of a computational asset of the specific aircraft, the control optimization being based on the obtained flight data, the current measurements or outputs by a feedback loop, and a mathematical model representing an actual operational performance of the specific aircraft and further providing a predictive indication of a future performance of the specific aircraft in response to a current state or input to the specific aircraft; adjusting, in response to a consideration of actual operational characteristics of the specific aircraft, the optimized control input; transmitting the adjusted optimized control input to the specific aircraft for use thereby to operate the specific aircraft to execute the prescribed flight to minimize the direct operating cost for a first portion of the prescribed flight; and iteratively repeating, for successive sequential instances in time for a remaining portion of the prescribed flight, the operations of obtaining the flight data, obtaining the current sample measurements or outputs, performing the control optimization, adjusting the optimized control input according to the control optimization performed at each successive sequential instance in time, and transmitting the adjusted optimized control input to the specific aircraft.",
    "2. The method of claim 1, wherein the successive sequential instances in time include a time from the initial obtaining of the current sample measurements to an end of the prescribed flight.",
    "3. The method of claim 1, wherein the flight specification comprises a mathematical model including tail specific performance and operational characteristics for the particular aircraft.",
    "4. The method of claim 3, wherein the mathematical model models at least the specific aircraft, engines of the specific aircraft, and atmospheric conditions for the flight during a future period of time when the adjusted optimized control input will be used to guide the specific aircraft.",
    "5. The method of claim 1, wherein the actual operational characteristics include aspects of a flight control function of the specific aircraft.",
    "6. The method of claim 1, wherein at least some of the flight data is obtained from a source separate and distinct from an airborne system of the particular aircraft.",
    "7. The method of claim 1, wherein the at least one state of the specific aircraft, includes a plurality of states corresponding to a plurality of functions of the specific aircraft.",
    "8. The method of claim 7, wherein at least one of the plurality of states is unknown based on the current sample measurements and an estimate for the at least one unknown state is determined, at least in part, based on at least one of the plurality of states known from the current sample measurements.",
    "9. A system comprising: a memory storing processor-executable program instructions; and a processor to execute the processor-executable program instructions to: obtain flight data including a flight specification and other flight related data for a specific aircraft for a future prescribed flight, the flight specification including at least flight constraints, a starting location, a destination location, and transient performance limits of the specific aircraft; and the other flight related data including future atmospheric conditions the specific aircraft will be subjected to when employing control commands generated by the processor and executing the future prescribed flight, and air traffic control information relevant to the future prescribed flight; generating control histories based on historical performances and generating corresponding state trajectories for the specific aircraft based on the flight data and performing model predictive control to predict a future performance of the specific aircraft and adjusting current control input actions to further control the specific aircraft to perform in an optimized manner during the future prescribed flight; obtain current sample measurements of at least one state or output of the specific aircraft during execution of the future prescribed flight; perform a control optimization based on the obtained flight data, the current measurements or outputs by a feedback loop, and a mathematical model representing an actual operational performance of the specific aircraft and further providing a predictive indication of a future performance of the specific aircraft in response to a current state or input to the specific aircraft; determine, based on the control optimization, a control input optimized to minimize a direct operating cost of the specific aircraft for the prescribed flight; adjust, in response to a consideration of actual operational characteristics of the specific aircraft, the optimized control input; transmit the adjusted optimized control input to the specific aircraft for use thereby to operate the specific aircraft to execute the prescribed flight to minimize the direct operating cost for a first portion of the prescribed flight; and iteratively repeat, for successive sequential instances in time for a remaining portion of the prescribed flight, the operations of obtaining the flight data, obtaining the current sample measurements or outputs, performing the control optimization, adjusting the optimized control input according to the control optimization performed at each successive sequential instance in time, and transmitting the transmitting the adjusted optimized control input to the specific aircraft.",
    "10. The system of claim 9, wherein the successive sequential instances in time include a time from the initial obtaining of the current sample measurements to an end of the prescribed flight.",
    "11. The system of claim 9, wherein the flight specification comprises a mathematical model including tail specific performance and operational characteristics for the particular aircraft.",
    "12. The system of claim 11, wherein the mathematical model models at least the specific aircraft, engines of the specific aircraft, and atmospheric conditions for the flight during a future period of time when the adjusted optimized control input will be used to guide the specific aircraft.",
    "13. The system of claim 9, wherein the actual operational characteristics include aspects of a flight control function of the specific aircraft.",
    "14. The system of claim 9, wherein at least some of the flight data is obtained from a source separate and distinct from an airborne system of the particular aircraft.",
    "15. The system of claim 9, wherein the at least one state of the specific aircraft includes a plurality of states corresponding to a plurality of functions of the specific aircraft.",
    "16. The system of claim 15, wherein at least one of the plurality of states is unknown based on the current sample measurements and an estimate for the at least one unknown state is determined, at least in part, based on at least one of the plurality of states known from the current sample measurements.",
    "17. A non-transitory computer-readable medium having processor-executable program instructions stored thereon, the medium comprising: program instructions to obtain flight data including a flight, specification and other flight related data for a specific aircraft for a future prescribed flight, the flight specification including at least flight constraints, a starting location, a destination location, and transient performance limits of the specific aircraft: and the other flight related data including future atmospheric conditions the specific aircraft will be subjected to when employing control commands generated by the processor and executing the future prescribed flight, and air traffic control information relevant to the future prescribed flight: program instructions to generate control histories based on historical performances and to generate corresponding state trajectories for the specific aircraft based on the flight data and to perform model predictive control to predict a future performance of the specific aircraft and adjusting current control input actions to further control the specific aircraft to perform in an optimized manner during the future prescribed flight; program instructions to obtain current sample measurements of at least one state or output of the specific aircraft during execution of the future prescribed flight; program instructions to perform a control optimization based on the obtained flight data, the current measurements or outputs by a feedback loop, and a mathematical model representing an actual operational performance of the specific aircraft and further providing a predictive indication of a future performance of the specific aircraft in response to a current state or input to the specific aircraft; program instructions to determine, based on the control optimization, a control input optimized to minimize a direct operating cost of the specific aircraft for the prescribed flight; program instructions to adjust, in response to a consideration of actual operational characteristics of the specific aircraft, the optimized control input; program instructions to transmit the adjusted optimized control input to the specific aircraft for use thereby to operate the specific aircraft to execute the prescribed flight to minimize the direct operating cost for a first portion of the prescribed flight; and program instructions to iteratively repeat, for successive sequential instances in time for a remaining portion of the prescribed flight, the operations of obtaining the flight data, obtaining the current sample measurements or outputs, performing the control optimization, adjusting the optimized control input according to the control optimization performed at each successive sequential instance in time, and transmitting the adjusted optimized control input to the specific aircraft.",
    "18. The medium of claim 17, wherein the successive sequential instances in time include a time from the initial obtaining of the current sample measurements to an end of the prescribed flight.",
    "19. The medium of claim 17, wherein the flight specification comprises a mathematical model including tail specific performance and operational characteristics for the particular aircraft.",
    "20. The medium of claim 19, wherein the mathematical model models at least the specific aircraft, engines of the specific aircraft, and atmospheric conditions for the flight during a future period of time when the adjusted optimized control input will be used to guide the specific aircraft."
  ],
  "description_excerpt": "The field of the present disclosure relates generally to flight management, more particularly, to systems, devices and methods of operation for flight management and applications thereof.\n\nThe cost of fuel is typically a large share of the operating expense in commercial aviation. As a consequence, operating efficiency and fuel savings are driving research for improvements in aircraft design and aircraft operations. The focus is primarily on those technologies that save fuel: aircraft and engine design, control design, and flight path planning and execution (called flight guidance).\n\nFlight Management Systems (FMS) onboard aircraft typically determine climb, cruise, and descent speeds and constant cruise altitudes in an effort to reduce or minimize Direct Operating Cost (DOC) given takeoff weight and range and assuming a number of factors such as, for example, constant thrust for climb and idle thrust for descent. These simplifying assumptions have traditionally been applied to implement practical systems, even though such assumptions and simplifications yield suboptimal performance and compromised fuel savings. Additionally, conventional flight management systems are typically reactive to a current state or states of an aircraft. In some aspects, the control system of the aircraft might continually react to current or past states of the aircraft in an attempt control operations of the aircraft.\n\nTherefore, there exists a need for systems and methods that improve the optimization problem for flight that is not strictly reactive and without simplifying assumptions to achieve optimal guidance.",
  "cpc": [
    "G08G 5/34",
    "G05D 1/0005",
    "G05D 1/101",
    "G08G 5/0021",
    "G08G 5/0026",
    "G08G 5/0034",
    "G08G 5/0039",
    "G08G 5/21",
    "G08G 5/22",
    "G08G 5/32"
  ],
  "ipc": [
    "G08G 5/00",
    "G05D 1/00"
  ],
  "assignees": [
    "General Electric Co"
  ],
  "inventors": [
    "Eric Richard Westervelt",
    "Mark Lawrence Darnell",
    "Reza Ghaemi",
    "David LAX"
  ],
  "filing_date": "2017-03-31",
  "publication_date": "2020-11-10",
  "grant_date": "2020-11-10",
  "priority_date": "2017-03-31",
  "application_number": "US-201715476351-A",
  "family_id": "63594121",
  "cited_by_count": 32,
  "citations": [
    "US4760530A",
    "WO2002099769A1",
    "US7437225B1",
    "US8888046B2",
    "US8255148B2",
    "US9031720B2",
    "US8463535B2",
    "US20140018980A1",
    "US9460629B2",
    "US9625261B2",
    "US20160246277A1",
    "US10049586B2",
    "US10086948B2",
    "US9536435B1",
    "US20180239364A1",
    "US10460610B2"
  ]
}