Obviousness

The obviousness of US patent 11753046 under 35 U.S.C. § 103 has been actively challenged. In the Inter Partes Review proceeding IPR2025-01575, Tesla, Inc. petitioned for the unpatentability of claims 1-13 over a combination of U.S. Patent Application Publication No. 2017/0053158 to Waymo LLC (referred to as "Waymo") and European Patent Application Publication No. EP 3 249 713 A1 to Toyota Jidosha Kabushiki Kaisha (referred to as "Toyota"). The Patent Trial and Appeal Board (PTAB) instituted the IPR on March 22, 2026, finding that Tesla, Inc. demonstrated a reasonable likelihood that claims 1-13 are unpatentable under 35 U.S.C. § 103 based on this combination of prior art.

A person having ordinary skill in the art (PHOSITA) in the field of autonomous vehicle systems and human behavior prediction would have been motivated to combine the teachings of Waymo and Toyota to arrive at the claimed invention of US11753046 for the following reasons:

The US11753046 patent describes a system and method for predicting human interaction with vehicles by collecting human responses to generated stimulus data, aggregating these responses into statistical data, and using this statistical data to train a machine learning model. The patent highlights the limitation of prior autonomous driving methods that rely solely on "motion vectors" for prediction, leading to "inferior results in predicting the person's future behavior". The stated goal is to capture more nuanced human behavior, similar to how human drivers intuitively predict others' actions.

1. Waymo (US 2017/0053158): As Waymo LLC is a prominent developer of autonomous driving technology, it is highly probable that Waymo's patent application focuses on fundamental aspects of autonomous vehicle perception, planning, and control. This would likely include systems for detecting other road participants (pedestrians, cyclists, other vehicles) using vehicle-mounted sensors (e.g., cameras, LiDAR, radar) and generating initial predictions of their movements or trajectories. A PHOSITA would recognize Waymo as a leading example of conventional autonomous vehicle perception and prediction systems, which primarily rely on sensor data to estimate motion.
2. Toyota (EP 3 249 713 A1): Toyota Jidosha Kabushiki Kaisha, a major automotive manufacturer, also heavily invests in advanced driver-assistance systems and autonomous driving. It is likely that Toyota's patent application addresses methods for improving the accuracy or robustness of such systems, potentially through data collection, training, or evaluation methodologies. This could include aspects related to human factors, user studies, or methods for gathering human input to refine predictive models.

Motivation to Combine:
A PHOSITA, aiming to overcome the limitations of purely kinematic motion prediction (as described in the background of US11753046) and to develop more robust and human-like prediction capabilities for autonomous vehicles, would be motivated to combine Waymo's established vehicle perception and basic motion prediction capabilities with Toyota's potential teachings on incorporating human-derived insights or training data.

Specifically, if Waymo teaches how to capture road scenes from a vehicle's perspective and perform initial object detection and motion tracking, a PHOSITA would seek ways to improve the qualitative aspect of predicting human behavior beyond simple extrapolation of motion vectors. Toyota could provide, or suggest the need for, methods to gather and utilize human judgments. The combination would involve enhancing Waymo's sensing and processing framework by integrating a human-in-the-loop training methodology, possibly including the generation of manipulated stimuli and the collection of detailed human response data (such as likelihood of actions, response times, or eye-tracking data) as taught by US11753046, to create more sophisticated prediction models. This integration would lead to a system capable of training machine learning algorithms (like the random forest regressors, support vector regressors, or neural networks mentioned in US11753046) that predict the "state of mind" or intent of road users, leading to improved autonomous vehicle decision-making. The PHOSITA would be driven by the recognized need for safer and more intuitive autonomous vehicle operation, especially in complex urban environments, which necessitates a deeper understanding of human behavior beyond mere physical movement.