US 12,488,241 B2
Behavior-guided path planning in autonomous machine applications
Urs Muller, Keyport, NJ (US); Mariusz Bojarski, Brooklyn, NY (US); Chenyi Chen, Fremont, CA (US); and Bernhard Firner, Highland Park, NJ (US)
Assigned to NVIDIA Corporation, Santa Clara, CA (US)
Filed by NVIDIA Corporation, Santa Clara, CA (US)
Filed on Dec. 8, 2023, as Appl. No. 18/533,860.
Application 18/533,860 is a continuation of application No. 16/409,056, filed on May 10, 2019, granted, now 11,966,838.
Claims priority of provisional application 62/687,200, filed on Jun. 19, 2018.
Prior Publication US 2024/0127062 A1, Apr. 18, 2024
Int. Cl. G06N 3/08 (2023.01); G05D 1/00 (2024.01); G06N 20/00 (2019.01); G06V 10/774 (2022.01); G06V 20/56 (2022.01)
CPC G06N 3/08 (2013.01) [G05D 1/0221 (2013.01); G05D 1/0246 (2013.01); G05D 1/0257 (2013.01); G05D 1/027 (2013.01); G06N 20/00 (2019.01); G06V 10/774 (2022.01); G06V 20/56 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving first image data obtained using a camera associated with a machine in an environment, the camera having a first perspective view with respect to the machine;
determining, based at least on one or more first parameters associated with the camera and one or more second parameters associated with a reference camera associated with the machine and having a second perspective view with respect to the machine, at least one of a rotation or a shift between the first perspective view with respect to the machine and the second perspective view with respect to the machine;
generating second image data by transforming, using the at least one of the rotation or the shift, an image represented by the first image data from the first perspective view with respect to the machine to the second perspective view with respect to the machine;
applying the second image data to one or more machine learning models, the one or more machine learning models being trained at least using training image data associated with the second perspective view of the reference camera;
determining, using the one or more machine learning models and based at least on the second image data, a trajectory for the machine through the environment; and
causing the machine to perform one or more operations based at least on the trajectory.