Resolving region-of-interest (ROI) overlaps for distributed simultaneous localization and mapping (SLAM) in edge cloud architectures
An illustrative, non-limiting method for resolving Region-On-Interest (ROI) overlaps includes receiving Simultaneous Localization and Mapping (SLAM) data obtained by a plurality of co-located Head-Mounted Devices (HMD), where the SLAM data comprises a plurality of landmarks and each HMD has an Information Handling System (IHS) in communication therewith. Each IHS may produce a subset of the plurality of landmarks based upon a respective HMD's SLAM data, and each IHS is: local with respect to the plurality of HMDs, at an edge of a network serving an HMD, or on a cloud. The method may include identifying an overlap between a first landmark produced by a first IHS and a second landmark produced by a second IHS, selecting between the first and second landmarks, at least in part, based upon the locations of the first and second IHSs, and rendering a map for display by a given HMD using the selected landmark.
1. A first Information Handling System (IHS), comprising:
a processor; and
a memory coupled to the processor, the memory having program instructions stored thereon that, upon execution by the processor, cause the first IHS to:
receive first Simultaneous Localization and Mapping (SLAM) data obtained by a first Head-Mounted Device (HMD), wherein the first SLAM data comprises a first plurality of landmarks;
in response to a determination that a first landmark of the first plurality of landmarks has a first confidence value below a threshold, receive second SLAM data obtained by a second HMD co-located with the first HMD, wherein the second SLAM data comprises a second plurality of landmarks, and wherein the second HMD is selected among a plurality of co-located HMDs, at least in part, based upon a location of a second IHS processing the second SLAM data;
identify an overlap between the first landmark and a second landmark of the second plurality of landmarks; and
in response to a determination that the second landmark has a second confidence value greater than the threshold, and in response to a determination that the second landmark has been generated within a selected time interval of the first landmark, apply a transform matrix to the second landmark to produce a transformed second landmark in a common coordinate system with respect to the first SLAM data, and render at least a partial map for display by the first HMD using the second transformed landmark instead of the first landmark.
2. The first IHS of claim 1 , wherein the first SLAM data comprises landmark data found in a Region of Interest (ROI) within an infrared (IR) or near-IR (NIR) frame captured via a first camera mounted on the first HMD, and wherein the second SLAM data comprises landmark data found in another ROI within another IR or NIR frame captured via a second camera mounted on the second HMD.
3. The first IHS of claim 1 , wherein the first confidence value comprises a first Kalman Gain, and wherein the second confidence value comprises a second Kalman Gain.
4. The first IHS of claim 1 , wherein to identify the overlap, the program instructions, upon execution, further cause the first IHS to determine that the first landmark and the second landmark are separated by a threshold distance.
5. The first IHS of claim 1 , wherein to determine the distance between the first landmark and the second landmark, the program instructions, upon execution, further cause the first IHS to receive at least one of: a first absolute pose of the first HMD or a second absolute pose of the second HMD.
6. The first IHS of claim 5 , wherein the program instructions, upon execution, further cause the first IHS to transform the coordinates of at least one of: the first landmark or the second landmark, into a common coordinate system using at least one of: the first or second absolute poses.
7. The first IHS of claim 1 , wherein the second IHS comprises one of: a co-located server, an edge server, or a cloud server.
8. The first IHS of claim 6 , wherein the second IHS is an edge server in response to the first HMD having a first latency requirement, or wherein the second IHS is a cloud server in response to the first HMD having a second latency requirement.
9. The first IHS of claim 6 , wherein the second IHS is an edge server in response to the first HMD having a first compute requirement, or wherein the second IHS is a cloud server in response to the first HMD having a second compute requirement.
10. The first IHS of claim 6 , wherein the second IHS is an edge server in response to the first HMD moving with a first speed, or wherein the second IHS is a cloud server in response to the first HMD moving with a second speed.
11. The first IHS of claim 6 , wherein the second IHS is an edge server in response to the first landmark having moved with a first speed in previous SLAM data, or wherein the second IHS is a cloud server in response to the first landmark having moved with a second speed in the previous SLAM data.
12. A hardware memory of a Head-Mounted Device (HMD), the hardware memory having program instructions stored thereon that, upon execution by a processor of the HMD, cause the HMD to:
receive first Simultaneous Localization and Mapping (SLAM) data obtained by a first Head-Mounted Device (HMD), wherein the first SLAM data comprises a first plurality of landmarks;
receive second SLAM data obtained by a second HMD co-located with the first HMD, wherein the second SLAM data comprises a second plurality of landmarks, wherein the second HMD is selected among a plurality of co-located HMDs based upon location of a second IHS processing the second SLAM data, and wherein the location is one of: co-located with respect to a first IHS, at a network edge, or on a cloud;
identify an overlap between a first landmark of the first plurality of landmarks and a second landmark of the second plurality of landmarks; and
in response to a determination that the second landmark has a second confidence value greater than a first confidence value of the first landmark, produce a transformed second landmark, and render a map for display by the first HMD using the second transformed landmark.
13. The hardware memory of claim 12 , wherein the second IHS is a co-located server in response to the first HMD having a first latency requirement, wherein the second IHS is an edge server in response to the first HMD having a second latency requirement, or wherein the second IHS is a cloud server in response to the first HMD having a third latency requirement.
14. The hardware memory of claim 12 , wherein the second IHS is a co-located server in response to the first HMD having a first compute requirement, wherein the second IHS is an edge server in response to the first HMD having a second compute requirement, or wherein the second IHS is a cloud server in response to the first HMD having a third compute requirement.
15. The hardware memory of claim 12 , wherein the second IHS is a co-located server in response to the first HMD moving with a first speed, wherein the second IHS is an edge server in response to the first HMD moving with a second speed, or wherein the second IHS is a cloud server in response to the first HMD moving with a third speed.
16. The hardware memory of claim 12 , wherein the second IHS is a co-located server in response to the first landmark having moved with a first speed in previous SLAM data, wherein the second IHS is an edge server in response to the first landmark having moved with a second speed in previous SLAM data, or wherein the second IHS is a cloud server in response to the first landmark having moved with a third speed in the previous SLAM data.
17. A method, comprising:
receiving first Simultaneous Localization and Mapping (SLAM) data obtained by a first Head-Mounted Device (HMD), wherein the first SLAM data comprises a first plurality of landmarks;
receiving second SLAM data obtained by a second HMD co-located with the first HMD, wherein the second SLAM data comprises a second plurality of landmarks, wherein the second HMD is selected among a plurality of co-located HMDs based upon location of a second IHS processing the second SLAM data, and wherein the location is one of: co-located with respect to a first Information Handling System (IHS), at a network edge, or on a cloud;
identifying an overlap between a first landmark of the first plurality of landmarks and a second landmark of the second plurality of landmarks; and
in response to a determination that the second landmark has a second confidence value greater than a first confidence value of the first landmark, producing a transformed second landmark, and rendering a map for display by the first HMD using the second transformed landmark.
18. The method of claim 17 , wherein the second IHS is a co-located server in response to the first HMD having a first latency requirement, wherein the second IHS is an edge server in response to the first HMD having a second latency requirement, or wherein the second IHS is a cloud server in response to the first HMD having a third latency requirement.
19. The method of claim 17 , wherein the second IHS is a co-located server in response to the first HMD having a first compute requirement, wherein the second IHS is an edge server in response to the first HMD having a second compute requirement, or wherein the second IHS is a cloud server in response to the first HMD having a third compute requirement.
20. The method of claim 17 , wherein the second IHS is a co-located server in response to the first HMD moving with a first speed, wherein the second IHS is an edge server in response to the first HMD moving with a second speed, wherein the second IHS is a cloud server in response to the first HMD moving with a third speed, wherein the second IHS is a co-located server in response to the first landmark having moved with a first speed in previous SLAM data, wherein the second IHS is an edge server in response to the first landmark having moved with a second speed in previous SLAM data, or wherein the second IHS is a cloud server in response to the first landmark having moved with a third speed in the previous SLAM data.