LIDAR system with fog detection and adaptive response
A LIDAR system can automatically determine, based on analysis of data collected from the sensor channels during operation, whether fog is present. If fog is present, the LIDAR system can operate in a fog mode, and if fog is not present, the LIDAR system can operate in a “clear-air” mode. The two modes can differ from each other with respect to the emitter signals and/or the signal processing applied to the sensor data.
1. A method of operating a LIDAR system, the method comprising:
determining, based on photosensor data from a plurality of sensor channels of the LIDAR system, whether atmospheric fog is present in a field of view of the LIDAR system;
in response to determining that atmospheric fog is present, operating the LIDAR system in a fog mode; and
in response to determining that atmospheric fog is not present, operating the LIDAR system in a clear-air mode,
wherein operating the LIDAR system in the fog mode includes operating a plurality of emitter channels to generate a pulse train having a first pulse pattern and operating the LIDAR system in the clear-air mode includes operating the plurality of emitter channels to generate a pulse train having second pulse pattern different from the first pulse pattern.
2. The method of claim 1 wherein operating the LIDAR system in the fog mode includes applying a Barker coding procedure and operating the LIDAR system in the clear-air mode includes not applying the Barker coding procedure.
3. The method of claim 2 wherein applying the Barker coding procedure includes:
operating an emitter channel of the LIDAR system to generate a pulse train for each of a plurality of shots, wherein the pulse trains for different shots in the plurality of shots are of at least two different pulse-train types; and
generating an accumulated histogram of detected light intensity across the plurality of shots wherein shots having a first one of the pulse-train types are accumulated with a positive weight and shots having a second one of the pulse-train types are accumulated with a negative weight.
4. The method of claim 3 wherein applying the Barker coding procedure further includes:
applying a matched filter to the accumulated histogram, wherein the matched filter is defined based on combining the pulse-train types.
5. The method of claim 3 wherein the number of shots in the plurality of shots is two.
6. The method of claim 1 wherein:
operating the LIDAR system in the fog mode includes operating an emitter of the LIDAR system to generate a pulse train having a single pulse; and
operating the LIDAR system in the clear-air mode includes operating the emitter to generate a pulse train having a plurality of pulses.
7. The method of claim 1 wherein determining whether atmospheric fog is present includes:
analyzing raw histogram data from the plurality of sensor channels of the LIDAR system; and
determining whether the raw histogram data from each of at least a minimum number of the sensor channels has a signal shape characteristic of fog.
8. The method of claim 7 wherein the minimum number of sensor channels is equal to a total number of sensor channels in the LIDAR system.
9. The method of claim 7 wherein the minimum number of sensor channels is more than half of a total number of sensor channels in the LIDAR system and less than the total number of sensor channels in the LIDAR system.
10. The method of claim 1 wherein determining whether atmospheric fog is present includes:
determining a total background flux across a plurality of sensor channels;
determining that atmospheric fog is present in the event that the total background flux exceeds a threshold; and
determining that atmospheric fog is not present in the event that the total background flux does not exceed the threshold.
11. A LIDAR system comprising:
an array of emitter channels;
an array of sensor channels;
a signal processing circuit coupled to the array of sensor channels and configured to determine, based on photosensor data from the array of sensor channels, whether atmospheric fog is present in a field of view; and
a control circuit coupled to the signal processing circuit, the array of emitter channels and the array of sensor channels, the control circuit being configured to:
operate the array of emitter channels and the array of sensor channels in a fog mode in response to determining that atmospheric fog is present; and
operate the array of emitter channels and the array of sensor channels in a clear-air mode in response to determining that atmospheric fog is not present,
wherein operating the LIDAR system in the fog mode includes operating a plurality of emitter channels to generate a pulse train having a first pulse pattern and operating the LIDAR system in the clear-air mode includes operating the plurality of emitter channels to generate a pulse train having second pulse pattern different from the first pulse pattern.
12. The LIDAR system of claim 11 wherein the control circuit is further configured such that operating the array of emitter channels and the array of sensor channels in the fog mode includes applying a Barker coding procedure and operating the array of emitter channels and the array of sensor channels in the clear-air mode includes not applying the Barker coding procedure.
13. The LIDAR system of claim 12 wherein the control circuit is further configured such that applying the Barker coding procedure includes:
operating an emitter channel of the array of emitter channels to generate a pulse train for each of a plurality of shots, wherein the pulse trains for different shots in the plurality of shots are of at least two different pulse-train types; and
generating an accumulated histogram of detected light intensity across the plurality of shots wherein shots having a first one of the pulse-train types are accumulated with a positive weight and shots having a second one of the pulse-train types are accumulated with a negative weight.
14. The LIDAR system of claim 13 wherein the control circuit is further configured such that applying the Barker coding procedure further includes:
applying a matched filter to the accumulated histogram, wherein the matched filter is defined based on combining the pulse-train types.
15. The LIDAR system of claim 13 wherein the number of shots in the plurality of shots is two.
16. The LIDAR system of claim 11 wherein the control circuit is further configured such that:
operating the array of emitter channels and the array of sensor channels in the fog mode includes operating an emitter of the LIDAR system to generate a pulse train having a single pulse; and
operating the array of emitter channels and the array of sensor channels in the clear-air mode includes operating the emitter to generate a pulse train having a plurality of pulses.
17. The LIDAR system of claim 11 wherein the control circuit is further configured such that determining whether atmospheric fog is present includes:
analyzing raw histogram data from the array of sensor channels; and
determining whether the raw histogram data from each of at least a minimum number of the sensor channels has a signal shape characteristic of fog.
18. The LIDAR system of claim 17 wherein the minimum number of sensor channels is equal to a total number of sensor channels in the LIDAR system.
19. The LIDAR system of claim 17 wherein the minimum number of sensor channels is more than half of a total number of sensor channels in the LIDAR system and less than the total number of sensor channels in the LIDAR system.
20. The LIDAR system of claim 11 wherein the control circuit is further configured such that determining whether atmospheric fog is present includes:
determining a total background flux across a plurality of sensor channels in the array of sensor channels;
determining that atmospheric fog is present in the event that the total background flux exceeds a threshold; and
determining that atmospheric fog is not present in the event that the total background flux does not exceed the threshold.