US 10614344

US Patent 10,614,344, titled "System and method of predicting human interaction with vehicles," was issued to Perceptive Automata LLC.

Here's a concise summary of the patent:

• Title: System and method of predicting human interaction with vehicles
• Assignee: Perceptive Automata LLC (Current Assignee); Perceptive Automata Inc (Original Assignee)
• Inventors: Samuel English Anthony, Kshitij Misra, Avery Wagner Faller
• Filing Date: 2019-07-16
• Issue Date: 2020-04-07
• Abstract: A computing device receives an image and a video segment of a road scene from a participant's perspective, featuring entities like pedestrians, cyclists, or motor vehicles. It then generates stimulus data, which can be the original image/video or an altered version. This stimulus data is transmitted to a user interface, which collects response data from users, including actions or likelihoods of actions for the entities in the road scene. The computing device aggregates this response data into statistical data, creates a model based on it, and applies this model to a second image or video segment to predict user behavior in that segment.

Plain-Language Overview of Independent Claims:

• Independent Claim 1 (Method Claim):
  This claim describes a method for predicting human behavior in a road scene. It involves a computing device performing several steps:

  1. Receiving Input: The device takes in an image or video of a road scene, as seen from a vehicle's perspective, which includes pedestrians, cyclists, or other vehicles.
  2. Generating Stimulus: It then creates "stimulus data" from this input. This might be the original image/video or a modified version designed to highlight certain aspects.
  3. Transmitting to User Interface: This stimulus data is sent to a user interface (e.g., a screen for human observers).
  4. Collecting Responses: The device receives feedback from users through the interface. This feedback includes what actions users predict for the road scene participants (e.g., crossing the street, changing lanes) and how likely those actions are.
  5. Aggregating Data: The collected user responses are grouped and analyzed to form statistical data.
  6. Creating a Model: A predictive model is built using this statistical data.
  7. Applying the Model: This trained model is then used on new, live images or video segments of road scenes.
  8. Generating Prediction: Finally, the device generates a prediction of how road users in the live scene will behave, based on the model's analysis.

• Independent Claim 10 (System Claim):
  This claim describes a system that performs the same functions as the method of Claim 1. It specifies that a "computing device" is configured to carry out all the steps mentioned in Claim 1: receiving road scene data, generating stimulus data, transmitting it to a user interface, receiving user response data, aggregating that data into statistics, creating a predictive model, applying the model to new road scene data, and generating a prediction of user behavior.

Regarding litigation, the patent indicates that it is currently "Active" and has "Family has litigation". This includes a PTAB case IPR2025-01573 (Pending - Instituted) and US cases filed in the Texas Western District Court (case 7:25-cv-00594) and Texas Eastern District Court (case 2:25-cv-00742). However, as of April 26, 2026, there is no authoritative information available from the performed search regarding specific dockets for US Patent 10614344 within the CAFC's 2026 docket.