Nvidia patents using cheap radar data with AI trained on expensive LiDAR
Nvidia describes a system where a neural network is trained on high-quality LiDAR data to learn how to interpret lower-quality radar data. Once trained, the model can process the cheaper radar inputs to detect obstacles and guide a machine's movements without needing the more expensive LiDAR sensors during operation.
In the filing’s own words
Systems and methods are disclosed relating to performing machine control operations using processing of lower-fidelity sensor data, such as RADAR data, by a machine learning model (e.g., neural network) that was trained using higher-fidelity ground truth sensor data, such as LiDAR data. During training, labels derived from the higher-fidelity ground truth data are propagated to lower-fidelity training data, and model parameters are updated accordingly, enabling the trained model to perform high-quality inference on lower-fidelity inputs. The machine learning model processes lower-fidelity sensor data to generate an output, such as freespace or occupancy probability detection within an operational environment, which is then used to cause a machine to perform control operations.
What exactly they claim (claim 1)
1. A system comprising: one or more central processing units (CPUs), one or more graphics processing units (GPUs), one or more hardware accelerators, and one or more sensors, wherein the system performs operations comprising: obtaining output data based at least on processing, using a neural network, sensor data having a first fidelity, the neural network having been previously trained to generate the output data using ground truth sensor data having a second fidelity higher than the first fidelity; and causing a machine to perform one or more control operations based at least on the output data.
Facts
- Company
- Nvidia (NVIDIA Corporation)
- Official title
- SENSOR DATA PROCESSING
- Publication number
- US20260321020A1
- Application number
- 19643792
- Filed
- Apr 9, 2026
- Published
- Oct 15, 2026
- Inventors
- 11
- Claims
- 19
Classified under
G06V 20/56 Physics & computingG01S 13/89 Physics & computingG01S 17/89 Physics & computingG06V 10/774 Physics & computingMore from Nvidia in AI & software
- Nvidia patents AI coach that analyzes your gameplay to suggest strategies Oct 13, 2026
- Nvidia patents AI system that writes and voices character conversations Sep 10, 2026
- Nvidia patents system that uses AI to find bugs in game videos Sep 17, 2026
- Nvidia patents AI method to make digital characters move realistically Sep 8, 2026
- Nvidia patents AI system that builds image generators from text prompts Sep 15, 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.