US 12,347,169 B2
Autonomous vehicle perception multimodal sensor data management
Julio Fernando Jarquin Arroyo, Baden-Wuerttemberg (DE); Ignacio J. Alvarez, Portland, OR (US); Cornelius Buerkle, Karlsruhe (DE); and Fabian Oboril, Karlsruhe (DE)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Dec. 22, 2021, as Appl. No. 17/559,422.
Prior Publication US 2022/0114805 A1, Apr. 14, 2022
Int. Cl. G06V 10/774 (2022.01); G06N 3/08 (2023.01); G06V 10/32 (2022.01); G06V 10/80 (2022.01); B60W 60/00 (2020.01)
CPC G06V 10/774 (2022.01) [G06N 3/08 (2013.01); G06V 10/32 (2022.01); G06V 10/803 (2022.01); B60W 60/001 (2020.02); B60W 2420/403 (2013.01); B60W 2420/408 (2024.01); B60W 2555/20 (2020.02)] 25 Claims
OG exemplary drawing
 
1. A system for autonomous vehicle perception development and training, the system comprising:
processing circuitry; and
a memory that includes instructions, the instructions, when executed by the processing circuitry, cause the processor circuitry to:
receive a multimodal perception dataset, the multimodal perception dataset including an image dataset captured by an image capture device and a ranging dataset captured by a ranging sensor device;
generate a perception model dataset based on the multimodal perception dataset, the perception model dataset matching a plurality of target perception model parameters of a target perception model;
generate a multimodal obfuscated machine learning dataset based on the perception model dataset, the multimodal obfuscated machine learning dataset including a sensor noise injection for both the image dataset and the ranging dataset; and
train a perception model based on the multimodal obfuscated machine learning dataset.