Airport management system with gate performance UI and related methods
An airport management system is for an airport with gates. The airport management system includes a server, and image sensors respectively at the gates. Each image sensor is configured to generate image data for an apron of a respective gate. The server is configured to receive the image data from each of the image sensors, detect objects and apron events within the apron of the respective gate based upon a machine learning model, and generate a gate performance UI for viewing the detected objects and the detected apron events. The gate performance UI includes a timeline menu of the detected apron events for each respective gate.
1. An airport management system for an airport with a plurality of gates, the airport management system comprising:
a server; and
a plurality of image sensors respectively at the plurality of gates, each image sensor configured to generate image data for an apron of a respective gate;
the server configured to
receive the image data from each of the plurality of image sensors,
detect a plurality of objects and a plurality of apron events associated therewith and within the apron of the respective gate based upon a machine learning model,
for each object, generate an object detection level of confidence value, an object identification, date/time values for acquisition, x-y coordinates of the object, and a size value of the object,
generate a gate performance user interface (UI) for viewing the detected plurality of objects and the detected plurality of apron events, the gate performance UI comprising
a timeline menu of the detected plurality of apron events for each respective gate,
a video menu, and
a context menu,
when a given detected apron event is selected in the timeline menu, generate and display the video menu to comprise
a video associated with the given detected apron event,
a plurality of boundary boxes respectively displayed over the detected plurality of objects in the video associated with the given detected apron event, and
respective object detection level of confidence values for each of the plurality of boundary boxes displayed with the detected plurality of objects, and
when the given detected apron event is selected in the timeline menu, generate and display the context menu for the given detected apron event comprising a time duration for the given detected apron event, and start/end times for the given detected apron event.
2. The airport management system of claim 1 wherein the machine learning model is uniquely trained for each gate.
3. The airport management system of claim 1 wherein the timeline menu comprises a duration indicator for each of the detected plurality of apron events.
4. The airport management system of claim 1 wherein the server is configured to generate an airport performance UI for viewing the detected plurality of objects and the detected plurality of apron events for the plurality of gates, the airport performance UI comprising a gate performance menu comprising, for each gate, a number and a length of air traffic holdouts, a number of early arrivals, a passenger loading bridge extension delay, and a gate utilization metric.
5. The airport management system of claim 4 wherein the airport performance UI comprises a dashboard menu comprising a number of flights for the airport, an average turn time for the airport, a gate utilization metric, and an on-time arrival metric.
6. The airport management system of claim 4 wherein the airport performance UI comprises an event performance menu comprising average time durations for each type of detected apron events.
7. The airport management system of claim 4 wherein the airport performance UI comprises a compliance menu comprising compliance metrics for a plurality of user-defined safety rules for the plurality of gates.
8. The airport management system of claim 1 wherein the detected plurality of apron events comprises a Foreign Object Debris (FOD) walk, a ground power unit connection, a fuel hose connection, a belt loader connection, a passenger loading bridge connection, and a cargo event.
9. The airport management system of claim 1 wherein each gate has a single image sensor configured to generate the image data.
10. The airport management system of claim 1 wherein each gate has a plurality of image sensors configured to generate the image data.
11. An airport management system for an airport with a plurality of gates, the airport management system comprising:
a server; and
a plurality of image sensors respectively at the plurality of gates, each image sensor configured to generate image data for an apron of a respective gate;
the server configured to
receive the image data from each of the plurality of image sensors,
detect a plurality of objects and a plurality of apron events associated therewith and within the apron of the respective gate based upon a machine learning model, the machine learning model being uniquely trained for each gate,
for each object, generate an object detection level of confidence value, an object identification, date/time values for acquisition, x-y coordinates of the object, and a size value of the object,
generate a gate performance user interface (UI) for viewing the detected plurality of objects and the detected plurality of apron events, the gate performance UI comprising
a timeline menu of the detected plurality of apron events for each respective gate, the timeline menu comprising a duration indicator for each of the detected plurality of apron events,
a video menu, and
a context menu,
when a given detected apron event is selected in the timeline menu, generate and display the video menu to comprise
a video associated with the given detected apron event,
a plurality of boundary boxes respectively displayed over the detected plurality of objects in the video associated with the given detected apron event, and
respective object detection level of confidence values for each of the plurality of boundary boxes displayed with the detected plurality of objects, and
when the given detected apron event is selected in the timeline menu, generate and display the context menu for the given detected apron event comprising a time duration for the given detected apron event, and start/end times for the given detected apron event.
12. The airport management system of claim 11 wherein the server is configured to generate an airport performance UI for viewing the detected plurality of objects and the detected plurality of apron events for the plurality of gates, the airport performance UI comprising a gate performance menu comprising, for each gate, a number and a length of air traffic holdouts, a number of early arrivals, a passenger loading bridge extension delay, and a gate utilization metric.
13. The airport management system of claim 12 wherein the airport performance UI comprises a dashboard menu comprising a number of flights for the airport, an average turn time for the airport, a gate utilization metric, and an on-time arrival metric.
14. The airport management system of claim 12 wherein the airport performance UI comprises an event performance menu comprising average time durations for each type of detected apron events.
15. The airport management system of claim 12 wherein the airport performance UI comprises a compliance menu comprising compliance metrics for a plurality of user-defined safety rules for the plurality of gates.
16. The airport management system of claim 11 wherein the detected plurality of apron events comprises a Foreign Object Debris (FOD) walk, a ground power unit connection, a fuel hose connection, a belt loader connection, a passenger loading bridge connection, and a cargo event.
17. The airport management system of claim 11 wherein each gate has a single image sensor configured to generate the image data.
18. A method for management of an airport with a plurality of gates, the method comprising:
operating a plurality of image sensors respectively at the plurality of gates, each image sensor configured to generate image data for an apron of a respective gate; and
operating a server to
receive the image data from each of the plurality of image sensors,
detect a plurality of objects and a plurality of apron events associated therewith and within the apron of the respective gate based upon a machine learning model,
for each object, generate an object detection level of confidence value, an object identification, date/time values for acquisition, x-y coordinates of the object, and a size value of the object,
generate a gate performance user interface (UI) for viewing the detected plurality of objects and the detected plurality of apron events, the gate performance UI comprising
a timeline menu of the detected plurality of apron events for each respective gate,
a video menu, and
a context menu, and
when a given detected apron event is selected in the timeline menu, generate and display the video menu to comprise
a video associated with the given detected apron event,
a plurality of boundary boxes respectively displayed over the detected plurality of objects in the video associated with the given detected apron event, and
respective object detection level of confidence values for each of the plurality of boundary boxes displayed with the detected plurality of objects, and
when the given detected apron event is selected in the timeline menu, generate and display the context menu for the given detected apron event comprising a time duration for the given detected apron event, and start/end times for the given detected apron event.
19. The method of claim 18 wherein the machine learning model is uniquely trained for each gate.
20. The method of claim 18 wherein the timeline menu comprises a duration indicator for each of the detected plurality of apron events.