IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0393886
(2003-03-21)
|
발명자
/ 주소 |
- Ruokangas,Corinne C.
- Mengshoel,Ole J.
- Uckun,Serdar
- Rand,Timothy W.
- Donohue,Patrick
- Tuvi,Selim
|
출원인 / 주소 |
- Rockwell Scientific Licensing LLC
|
인용정보 |
피인용 횟수 :
25 인용 특허 :
6 |
초록
▼
An Aviation Weather Awareness and Reporting Enhancements (AWARE) system provides situational awareness by effectively filtering, analyzing and visualizing aviation weather data and specific hazard alerts in preflight, in-cockpit and controller applications. The AWARE system includes a temporal-spati
An Aviation Weather Awareness and Reporting Enhancements (AWARE) system provides situational awareness by effectively filtering, analyzing and visualizing aviation weather data and specific hazard alerts in preflight, in-cockpit and controller applications. The AWARE system includes a temporal-spatial databases that filters weather data and a Bayesian network that assesses specific hazards in the filtered weather data in the context of pilot preferences, aircraft properties and airport properties. The filtered weather data and hazard alerts are then displayed on a client.
대표청구항
▼
We claim: 1. An Aviation Weather Awareness and Reporting Enhancements (AWARE) system, comprising: a database configured to store text and graphic weather data and to store context information including at least one of pilot preferences, aircraft properties and airport properties; a server that extr
We claim: 1. An Aviation Weather Awareness and Reporting Enhancements (AWARE) system, comprising: a database configured to store text and graphic weather data and to store context information including at least one of pilot preferences, aircraft properties and airport properties; a server that extracts temporally-spatially filtered weather data and context information from the database and uses a Bayesian network (BN) to assess the weather data and context information and issue hazard alerts; and a client that receives user requests, issues client requests to the server to extract the weather data and context information and displays the filtered weather data and hazard alerts. 2. The AWARE system of claim 1, wherein the Bayesian network is defined as a tuple (V,E,P), where V is a set of nodes, E is a set of edges and P is a set of conditional probability distributions, said set of nodes V comprising at least one hazard node that is used to compute a hazard alert, said hazard node having at least one parent evidence node that represents weather data or context information. 3. The AWARE system of claim 2, wherein the server further comprises a BN wrapper that instantiates, in the BN, at least one evidence node with the filtered weather data and/or the relevant context data and executes an algorithm to compute marginal distributions over the hazard nodes, which are compared to at least one threshold and, if applicable, declared hazard alerts. 4. The AWARE system of claim 3, wherein the evidence node comprises a weather data source node and a context node. 5. The AWARE system of claim 4, wherein the wrapper extracts actual context information from the database to instantiate the context node. 6. The AWARE system of claim 4, wherein the context node has a default value. 7. The AWARE system of claim 6, wherein the default value is probabilistic. 8. The AWARE system of claim 4, wherein the wrapper extracts actual weather data from the database to instantiate the weather data source node. 9. The AWARE system of claim 4, wherein the weather data source node has probabilistic value. 10. The AWARE system of claim 3, wherein the database stores pilot preferences including at least one of qualifications, ratings and weather preferences, aircraft properties including at least one of make, fuel capacity, range, cruise speed and takeoff and landing distances and airport properties including at least one of identification, runway length, surface conditions, lighting and runway open/closed. 11. The AWARE system of claim 3, wherein the set of nodes V includes a plurality of first tier hazard nodes whose marginal distributions and hazard alert status are determined by the conditional probability distributions P and a plurality of said parent evidence nodes, said plurality of first tier hazard nodes and their marginal distributions being the parent nodes for at least one second tier hazard node, said wrapper executing the algorithm to compute a marginal distribution over the second tier hazard node, which is compared to at least one threshold and, if applicable, declared the hazard alert. 12. The AWARE system of claim 11, wherein the first tier hazard nodes are assigned respective utilities based on a severity measure that compares weather data to context information, said algorithm computing a utility-based measure of hazard by weighting the marginal distributions with said utilities. 13. The AWARE system of claim 11, wherein said first and second tier hazard nodes are assigned respective utilities based on a criticality measure of the hazard node, said algorithm computing a utility-based measure of hazard by weighting the marginal distributions with said utilities. 14. The AWARE system of claim 3, wherein said set of nodes V comprises a Mission Hazard node. 15. The AWARE system of claim 14, wherein said Mission Hazard node has parents hazard nodes Depart/Climb Hazard, Cruise Hazard and Descend/Land Hazard. 16. The AWARE system of claim 15, wherein said Depart/Climb hazard node has at least one parent hazard node selected from Visibility/Ceiling Hazard Takeoff (TO), Wind Hazard TO, Runway Hazard TO, Precipitation Hazard TO and Other Overall Weather Hazard TO, which in turn have at least one parent evidence node. 17. The AWARE system of claim 15, wherein said Cruise hazard node has at least one parent hazard node selected from Visibility/Ceiling Hazard Cruise (C), Winds Aloft Hazard C, Endurance Hazard C and Other Overall Weather Hazard C, which in turn have at least one parent evidence node. 18. The AWARE system of claim 15, wherein said Descend/Land hazard node has at least one parent hazard node selected from Visibility/Ceiling Hazard Land (L), Wind Hazard L, Precipitation Hazard L and Other Overall Weather Hazard L, which in turn have at least one parent evidence node. 19. The AWARE system of claim 3, wherein the server comprises a text and graphics postprocessor (TGP) that integrates the filtered weather data for display by the client. 20. The AWARE system of claim 19, wherein the TGP contextually filters the T/S filtered weather data, said contextual filter being either hard-coded or accessed from the context information in the database. 21. The AWARE system of claim 19, wherein the T/S filtered weather data extracted from the database and returned to the Bayesian network and the TGP cover different spatial regions. 22. The AWARE system of claim 21, wherein the T/S filtered weather data returned to the postprocessor is buffered to encompass a wider region than the filtered weather data returned to the Bayesian network. 23. The AWARE system of claim 19, wherein the TGP identifies events in the weather text data and specifies icon identifiers to those events that are registered to the weather graphic data. 24. The AWARE system of claim 3, wherein the client displays the filtered graphic weather data, a portion of the filtered text weather data, hazard alerts and hazard assessment information in different frames. 25. The AWARE system of claim 24, wherein the portion of the filtered text weather data and the hazard assessment information are keyed to either the selection of a hazard alert or proximity of the aircraft to the hazard. 26. The AWARE system of claim 25, wherein the display portion includes the raw text weather data from the database that contributed to the declaration of the alert. 27. The AWARE system of claim 26, wherein the hazard assessment information comprises a summary of the parent evidence nodes and the marginal distribution of the hazard node. 28. The AWARE system of claim 27, wherein the client allows a user to drill-down into the hazard assessment information or text weather data to access additional detail. 29. The AWARE system of claim 24, wherein the client overlays graphic icons assigned to events in the weather text data on the graphic weather data. 30. The AWARE system of claim 24, wherein the client is configured to receive user requests to enter or update pilot, aircraft or airport context information into the database or to override stored context information. 31. The AWARE system of claim 24, wherein the marginal distribution is compared to a plurality of thresholds associated with different alert states and the hazard alert associated with the highest exceeded threshold is declared, said client displaying the hazard alerts associated with the different alert states in different colors. 32. The AWARE system of claim 24, wherein the set of nodes V includes a plurality of hazard nodes organized in a graph structure in which the evidence nodes are the parents to the lowest tier hazard nodes in the tree, which are in turn the parents to the next tier of hazard nodes, said client displaying alerts for all the hazard nodes that declare alerts in the graph. 33. The AWARE system of claim 24, wherein the set of nodes V includes a plurality of hazard nodes organized in a graph structure in which the evidence nodes are the parents to the lowest tier hazard nodes in the tree, which are in turn the parents to the next tier of hazard nodes, said client displaying alerts for the lowest hazard nodes in a branch of declared alerts in the graph. 34. The AWARE system of claim 3, wherein essentially the same Bayesian network and contextual information is used for preflight, in-cockpit and controller applications. 35. The AWARE system of claim 34, wherein the preflight application uses both past and current weather data sources and displays weather data and hazard alerts for the depart/climb, cruise and descend/land legs of a flight. 36. The AWARE system of claim 34, wherein the in-cockpit application uses only current weather data sources and displays weather data and hazard alerts only for the current leg of a flight. 37. The AWARE system of claim 34, wherein the controller application displays weather data and hazard alerts for multiple flights, each flight having its own Bayesian network. 38. The AWARE system of claim 37, wherein the client selectively groups hazard alerts from one or more flights. 39. An Aviation Weather Awareness and Reporting Enhancements (AWARE) system, comprising: a temporal-spatial (T/S) database that stores the text and graphic weather data, and a contextual-information database comprising data selected from at least one of pilot preferences, aircraft properties and airport properties; and a Bayesian network (BN) defined as a tuple (V,E,P), where V is a set of nodes, E is a set of edges and P is a set of conditional probability distributions, said set of nodes V comprising at least one hazard node whose state determines a hazard alert, said hazard node having at least one parent data node that is instantiated with weather data from the T/S database and at least one parent context node that is instantiated with preferences or properties from the contextual-information databases. 40. The AWARE system of claim 39, wherein the set of nodes V includes a plurality of first tier hazard nodes whose marginal distributions and hazard alert status are determined by the conditional probability distributions P and a plurality of said parent data and context nodes, said plurality of first tier hazard nodes and their marginal distributions being the parent nodes for at least one second tier hazard node, said wrapper executing the algorithm to compute a marginal distribution over the second tier hazard node, which is compared to at least one threshold and, if applicable, declared the hazard alert. 41. The AWARE system of claim 40, wherein the first tier hazard nodes are assigned respective utilities based on a severity measure that compares weather data to preferences or properties, said algorithm computing a utility-based measure of hazard by weighting the marginal distributions with said utilities. 42. The AWARE system of claim 40, wherein said first and second tier hazard nodes are assigned respective utilities based on a criticality measure of the hazard node, said algorithm computing a utility-based measure of hazard by weighting the marginal distributions with said utilities. 43. The AWARE system of claim 39, wherein essentially the same Bayesian network and contextual information database is used for preflight, in-cockpit and controller applications. 44. An Aviation Weather Awareness and Reporting Enhancements (AWARE) system, comprising: a temporal-spatial (T/S) database that stores the text and graphic weather data, a contextual-information database comprising data selected from at least one of pilot preferences, aircraft properties and airport properties a text and graphics postprocessor (TGP) that integrates filtered text and graphic weather data extracted from the T/S database, said TGP identifying events in the weather text data and assigning icons to those events that are registered to the weather graphic data, a Bayesian network (BN) defined as a tuple (V,E,P), where V is a set of nodes, E is a set of edges and P is a set of conditional probability distributions, that assesses the filtered text and graphic weather data in the context of preferences and properties from the contextual-information database to declare hazard alerts and provide hazard assessment information, and a client that displays the filtered graphic weather data, the overlaid icons and hazard alerts and, for a selected hazard alert, displays the relevant text weather data and hazard assessment information. 45. The AWARE system of claim 44, wherein the relevant text weather data includes the text weather data from the database that contributed to the declaration of the alert. 46. The AWARE system of claim 44, wherein the hazard assessment information comprises a summary of the parent evidence nodes and the marginal distribution of the hazard node. 47. The AWARE system of claim 44, wherein the client allows a user to drill-down into the hazard assessment information or text weather data to access lower nodes in the Bayesian network. 48. The AWARE system of claim 44, wherein a marginal distribution is calculated for each hazard node and compared to a plurality of thresholds associated with different alert states and the hazard alert associated with the highest exceeded threshold is declared, said client displaying the hazard alerts associated with the different alert states in different colors. 49. The AWARE system of claim 44, wherein the set of nodes V includes a plurality of hazard nodes organized in a graph structure in which the evidence nodes are the parents to the lowest tier hazard nodes in the tree, which are in turn the parents to the next tier of hazard nodes, said client displaying alerts for all the hazard nodes that declare alerts in the graph. 50. The AWARE system of claim 44, wherein the set of nodes V includes a plurality of hazard nodes organized in a graph structure in which the evidence nodes are the parents to the lowest tier hazard nodes in the tree, which are in turn the parents to the next tier of hazard nodes, said client displaying alerts for the lowest hazard nodes in a branch of declared alerts in the graph. 51. An Aviation Weather Awareness and Reporting Enhancements (AWARE) system, comprising: An AWARE client configured to receiver user requests, issue client requests and display weather data and hazard alerts; an AWARE database comprising; a temporal-spatial (T/S) database that stores the text and graphic weather data, and a contextual-information database comprising data selected from at least one of pilot preferences, aircraft properties and airport properties; an AWARE server comprising; a query formulator that issues queries to extract filtered weather data from the T/S database in response to client requests, a text and graphics postprocessor (TGP) that integrates the filtered text and graphic weather data and forwards it to the AWARE client, and a decision support system (DSS) including a Bayesian network (BN) defined as a tuple (V,E,P), where V is a set of nodes, E is a set of edges and P is a set of conditional probability distributions, said set of nodes V being organized in a graph structure in which evidence nodes are the parents to first tier hazard nodes, which are in turn the parents to second tier hazard nodes, said evidence nodes being instantiated with the filtered weather data and relevant preferences and properties from the contextual-information database to determine the state of said first tier hazard nodes, which in turn determine the state of said second tier hazard nodes, the state of said hazard nodes determining the hazard alerts displayed by the client. 52. The AWARE system of claim 51, wherein the DSS executes an algorithm to compute marginal distributions over the hazard nodes, compares the marginal distributions to at least one threshold to determine the state and declares, if applicable, hazard alerts. 53. The AWARE system of claim 51, wherein the evidence node comprises a weather data source node and a context node. 54. The AWARE system of claim 51, wherein the first tier hazard nodes are assigned respective utilities based on a severity measure that compares weather data to context information, said algorithm computing a utility-based measure of hazard by weighting the marginal distributions with said utilities. 55. The AWARE system of claim 51, wherein said first and second tier hazard nodes are assigned respective utilities based on a criticality measure of the hazard node, said algorithm computing a utility-based measure of hazard by weighting the marginal distributions with said utilities. 56. The AWARE system of claim 51, wherein said set of nodes V comprises a Mission Hazard node, which has parents hazard nodes Depart/Climb Hazard, Cruise Hazard and Descend/Land Hazard. 57. The AWARE system of claim 51, wherein the client displays a portion of the filtered text weather data and hazard assessment information that are keyed to user selection of a particular hazard alert. 58. A method of providing Aviation Weather Awareness and Reporting Enhancements (AWARE), comprising: Storing text and graphic weather data; Storing contextual-information including at least one of pilot preferences, aircraft properties and airport properties; Temporally and spatially filtering the text and graphic weather data; Displaying the filtered text and graphic weather data; Retrieving relevant preferences and properties to the filtered weather data; Using a Bayesian network to assess the filtered text and graphic weather data in the context of the relevant preferences and properties and issue hazard alerts; and Displaying the hazard alerts. 59. The method of claim 58, the text and graphic weather data is temporally and spatially filtered in response to a user generated request. 60. The method of claim 59, wherein the user is a pilot or a controller using AWARE in one of a preflight, in-cockpit or controller application. 61. The method of claim 58, wherein the Bayesian network defined as a tuple (V,E,P), where V is a set of nodes organized in a graph structure in which data source nodes and context nodes are parents to first tier hazard nodes, which are in turn parents to second tier hazard nodes, E is a set of edges and P is a set of conditional probability distributions, said Bayesian network assessing the weather data by; a) Instantiating the data source nodes with filtered text and graphic weather data; b) Instantiating the context nodes with relevant preferences and properties; c) Computing marginal distributions over the first tier hazard nodes; d) Comparing the marginal distributions to at least one threshold; e) Declaring a hazard alert for those hazard nodes whose marginal distributions exceed the threshold; and f) Repeating steps c through e for the second tier hazard nodes. 62. The method of claim 61, further comprising the steps of; assigning a utility to each first tier hazard node based on a severity measure that compares the weather data to the preferences or properties; computing a utility-based measure of hazard by weighting the marginal distributions with the utilities; and comparing the utility-based measure to at least one threshold to declare the hazard alerts. 63. The method of claim 61, further comprising the steps of; assigning a utility to each first tier hazard node based on a criticality measure of the hazard node; computing a utility-based measure of hazard by weighting the marginal distributions with the utilities; and comparing the utility-based measure to at least one threshold to declare the hazard alerts. 64. The method of claim 58, wherein temporally-spatially filtered text and weather data that is displayed is buffered to encompass a wider region than the temporally-spatially filtered text and weather that is provided to the Bayesian network. 65. The method of claim 58, further comprising the step of identifying events in the weather text data, assigning icons to those events that are registered to the graphic weather data, and displaying the icons on the graphic weather data. 66. The method of claim 58, wherein hazard assessment information from the Bayesian network and a portion of the filtered text weather data are displayed and keyed to either a selection of a particular hazard alert or proximity of the aircraft to the hazard. 67. The method of claim 66, further comprising a capability to drill-down into the hazard assessment information or text weather data to access additional detail.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.