Neuromorphic cameras for aircraft
An onboard aircraft landing system includes one or more event-based cameras disposed at known locations to capture the runway and visible surrounding features such as lights and runway markings. The event-based cameras produce a continuous stream of event data that may be quickly processed to identify both light and dark features contemporaneously, and calculate an aircraft pose relative to the runway based on the identified features and the known locations of the event-based cameras. Composite features are identified via the relative location of individual features corresponding to pixel events.
1. A computer apparatus comprising:
one or more event-based cameras; and
at least one processor in data communication with the one or more event-based cameras and a memory storing processor executable code for configuring the at least one processor to:
receive a continuous stream of pixel change events from each of the one or more event-based cameras;
identify one or more light sources via the continuous stream of pixel change events;
identify one or more low-light features via the continuous stream of pixel change events; and
determine a pose estimate for a platform including the one or more event-based cameras based on the identified one or more light sources and the identified one or more low-light features, wherein the one or more light sources and identified one or more low-light features are related by known, standard relative locations.
2. The computer apparatus of claim 1 , wherein the at least one processor is further configured to identify at least one composite feature based on a relative disposition of two or more of the identified light sources and identified low-light features.
3. The computer apparatus of claim 1 , wherein the at least one processor is further configured to identify a flashing LED light source.
4. The computer apparatus of claim 1 , wherein identifying the one or more light sources comprises executing a random sample consensus algorithm.
5. The computer apparatus of claim 4 , wherein the at least one processor is further configured to organize pixel change events in an x/y-time space.
6. The computer apparatus of claim 1 , wherein the at least one processor is further configured to identify one or more edge features via the continuous stream of pixel change events.
7. The computer apparatus of claim 1 , wherein the at least one processor is further configured to correlate locations of the identified light sources and the identified low-light features in time based on time stamps of the pixel change events.
8. An aircraft landing system comprising:
one or more event-based cameras disposed at known locations on an aircraft to capture images of a runway during a landing procedure; and
at least one processor in data communication with the one or more event-based cameras and a memory storing processor executable code for configuring the at least one processor to:
receive a continuous stream of pixel change events from each of the one or more event-based cameras;
identify one or more light sources via the continuous stream of pixel change events;
identify one or more low-light features via the continuous stream of pixel change events; and
determine a pose estimate for the aircraft based on the identified one or more light sources and the identified one or more low-light features, wherein the one or more light sources and identified one or more low-light features are related by known, standard relative locations.
9. The aircraft landing system of claim 8 , wherein the at least one processor is further configured to identify at least one composite feature based on a relative disposition of two or more of the identified light sources and identified low-light features.
10. The aircraft landing system of claim 8 , wherein the at least one processor is further configured to identify a flashing LED light source.
11. The aircraft landing system of claim 8 , wherein identifying the one or more light sources comprises executing a random sample consensus algorithm.
12. The aircraft landing system of claim 11 , wherein the at least one processor is further configured to organize pixel change events in an x/y-time space.
13. The aircraft landing system of claim 8 , wherein the at least one processor is further configured to identify one or more edge features via the continuous stream of pixel change events.
14. The aircraft landing system of claim 8 , wherein the at least one processor is further configured to correlate locations of the identified light sources and the identified low-light features in time based on time stamps of the pixel change events.
15. A method for determining an aircraft pose estimate comprising:
receiving a continuous stream of pixel change events from one or more event-based cameras;
identifying one or more light sources via the continuous stream of pixel change events;
identifying one or more low-light features via the continuous stream of pixel change events; and
determining a pose estimate for the aircraft based on the identified one or more light sources and the identified one or more low-light features, wherein the one or more light sources and identified one or more low-light features are related by known, standard relative locations.
16. The method of claim 15 , further comprising identifying at least one composite feature based on a relative disposition of two or more of the identified light sources and identified low-light features.
17. The method of claim 15 , further comprising identifying a flashing LED light source.
18. The method of claim 15 , wherein identifying the one or more light sources comprises executing a random sample consensus algorithm.
19. The method of claim 18 , further comprising organizing pixel change events in an x/y-time space.
20. The method of claim 15 , further comprising correlate locations of the identified light sources and the identified low-light features in time based on time stamps of the pixel change events.