US 12,111,492 B2
Adaptable optical neural network system
Sean P. Rodrigues, Ann Arbor, MI (US); Paul Donald Schmalenberg, Ann Arbor, MI (US); Hideo Iizuka, Ann Arbor, MI (US); Jae Seung Lee, Ann Arbor, MI (US); and Ercan Mehmet Dede, Ann Arbor, MI (US)
Assigned to Toyota Motor Engineering & Manufacturing North America, Inc., Plano, TX (US)
Filed by Toyota Motor Engineering & Manufacturing North America, Inc., Plano, TX (US)
Filed on Oct. 1, 2019, as Appl. No. 16/589,603.
Prior Publication US 2021/0097378 A1, Apr. 1, 2021
Int. Cl. G06N 3/044 (2023.01); G02B 6/12 (2006.01); G06F 18/24 (2023.01); G06N 3/04 (2023.01); G06N 3/048 (2023.01); G06N 3/067 (2006.01); G06V 10/82 (2022.01)
CPC G02B 6/12004 (2013.01) [G06F 18/24 (2023.01); G06N 3/04 (2013.01); G06N 3/067 (2013.01); G06V 10/82 (2022.01)] 16 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
an optical input that provides an optical signal that is an analog light wave;
a chassis component including at least one modular mounting location for receiving a modular network component; and
an optical neural network (ONN) operably connected with the optical input, the ONN configured to perform optical processing on the optical signal according to a deep learning algorithm, wherein the ONN includes optical components arranged into layers to form a physical architecture of the ONN, wherein the optical input provides the optical signal into the ONN via an optical relay to maintain the optical signal as the analog light wave from acquisition into the ONN, and
wherein the modular network component being an additional optical processing component that is configured to function in cooperation with the ONN to adapt the deep learning algorithm by altering the physical architecture.