Obviousness

An obviousness analysis under 35 U.S.C. § 103 for US patent 10614344, based on the provided information, identifies a specific combination of prior art references that the Patent Trial and Appeal Board (PTAB) found to render several claims likely unpatentable.

Combinations of Prior Art References and Motivation for Combination

Claims 1, 2, 4-13, 16-20, and 22-26 of US patent 10614344 were challenged in Inter Partes Review (IPR) IPR2025-01573 as unpatentable under 35 U.S.C. § 103. The PTAB instituted trial on these claims, finding a reasonable likelihood of obviousness over the combination of:

• U.S. Patent Publication No. 2017/0286820 (Krumm)
• U.S. Patent Publication No. 2017/0053154 (Cho)
• U.S. Patent No. 8,674,809 (Tiwari)

The motivation for a person having ordinary skill in the art to combine these references stems from their collective teachings regarding the use of human response data to train machine learning models for predicting human behavior in road scenes. The PTAB was persuaded by arguments that these references, when combined, teach the collection of human response data for training such models to predict human interaction with vehicles. This indicates that the references individually disclose elements pertinent to the patent's claims, and their combination would have been an obvious step to achieve the claimed functionality of predicting human interaction, such as pedestrian or cyclist behavior, in traffic environments.