Nvidia patents map system that helps self-driving cars switch locations smoothly
Nvidia describes a high-definition map structure broken into small geographic tiles and road segments. This design allows a vehicle to download only the specific map data it needs for its current location, rather than loading the entire map. When the car moves from one road segment to another, the system uses the connection between them to instantly adjust the car's position data so it knows exactly where it is.
In the filing’s own words
In various examples, a high definition (HD) map is provided that includes a segmented data structure that allows for selective access to desired road segments and corresponding layers of map data. For example, the HD map may be segmented into a series of tiles that may correspond to a geographic region, and each of the tiles may include any number of road segments corresponding to portions of the geographic region. Each road segment may include a corresponding set of layers—which may include driving layers for use by the ego-machine and/or training layers for generating ground truth data—from the HD map that are associated with the road segment alone. As such, when traversing the environment, an ego-machine may determine one or more road segments within a tile corresponding to a current location, and may selectively download one or more layers for each of the one or more road segments.
What exactly they claim (claim 1)
1. At least one processor comprising: one or more circuits to: access, based at least on a current location of an ego-machine, first localization data from a first road segment of road segments representing respective map regions aligned to one or more roads of a high definition (HD) map, the HD map including a graph having nodes representing the road segments, wherein the first road segment stores the first localization data corresponding to a first region of the respective map regions and a second road segment of the road segments stores second localization data corresponding to a second region of the respective map regions; perform one or more localization operations using the first localization data from the first road segment to determine first localization results of the first road segment, the first localization results including a location of the ego-machine within a first local coordinate system of the first road segment; identify an edge in the graph based at least on the edge representing a road-level connection between the first road segment and the second road segment; identify, using the edge, a spatial transformation between the first local coordinate system of the first road segment and a second local coordinate system of the second road segment; translate, when the ego-machine traverses into the second road segment from the first road segment and using the spatial transformation, the first localization results from the first local coordinate system to second localization results of the second road segment in the second local coordinate system, the second localization results including a translated location of the ego-machine; and perform one or more of autonomously controlling at least one of steering or speed of the ego-machine using the second localization data and the translated location in the second local coordinate system.
Facts
- Company
- Nvidia (NVIDIA Corporation)
- Official title
- High definition mapping
- Patent number
- US12736368B2
- Application number
- 17726429
- Filed
- Apr 21, 2022
- Granted
- Sep 15, 2026
- Inventors
- 3
- Claims
- 30
Classified under
G01C 21/3878 Physics & computingB60W 60/001 Operations & transportG01C 21/3881 Physics & computingB60W 2420/403 Operations & transportB60W 2420/408 Operations & transportB60W 2552/53 Operations & transportMore from Nvidia in Cars & transport
- Nvidia patents robotaxi system with AI dispatcher and virtual rails Sep 15, 2026
- Nvidia patents car drowsiness check that uses your body clock Oct 13, 2026
- Nvidia patents 3D camera system to measure passengers in vehicles Oct 13, 2026
- Nvidia patents cruise control that predicts driver's next move Oct 20, 2026
- Nvidia patents dual-brain system for self-driving cars Sep 17, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.