/Patent Tracker
NvidiaGranted patentCars & transportInterest 6/10

Nvidia patents method to check if self-driving cars know where they are

Nvidia describes a system that checks how accurately a vehicle knows its own position by comparing two different estimates. One estimate comes from tracking the vehicle's movement, while the other comes from matching what the sensors see to a pre-made map. The system calculates the difference between these two to figure out how much error exists in the vehicle's location data.

Why it mattersThis helps autonomous vehicles trust their own position data more, which is critical for safe navigation in complex environments.

In the filing’s own words

In various examples, accuracy determinations for localization in autonomous and semi-autonomous systems and applications are described herein. Systems and methods are disclosed that determine one or more errors associated with vehicle localization using various types of sensor data generated using a vehicle. For instance, a first component of the vehicle may use a map and first sensor data to determine an estimated pose of the vehicle. A second component of the vehicle may then determine the error(s) associated with the estimated pose based on both actual motion of the vehicle within the environment, as determined using second sensor data, and comparing features represented by the first sensor data to features represented by the map. In some examples, the second component may further determine information associated with the error(s), such as one or more uncertainties associated with the error(s).

What exactly they claim (claim 1)

1. A method comprising: determining, based at least on motion data obtained using one or more motion sensors of a machine, a first pose associated with the machine within an environment; determining, based at least on one or more correspondences between one or more first features represented by sensor data obtained using one or more sensors of the machine and one or more second features represented by map data, a second pose of the machine within the environment; after the determining the second pose: determining a first error corresponding to the second pose based at least on one or more differences between the first pose and the second pose associated with the machine; and determining a second error corresponding to the second pose based at least on the one or more correspondences between the one or more first features represented by the sensor data and the one or more second features represented by the map data; determining, based at least on the first error and the second error, a third error corresponding to the second pose; and causing, based at least on the third error corresponding to the second pose, the machine to navigate from a first location within an environment to a second location within the environment.

Facts

Company
Nvidia (NVIDIA Corporation)
Official title
Machine localization accuracy
Patent number
US12728887B2
Application number
18446649
Filed
Aug 9, 2023
Granted
Sep 8, 2026
Inventors
3
Claims
20

Classified under

B60W 60/001 Operations & transportG01C 21/3867 Physics & computingB60W 2520/14 Operations & transport
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Nvidia in Cars & transport

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.