PTAB

IPR2025-01604

Avidbots USA Corp v. Brain Corp

Log In
Key Events
Petition

1. Case Identification

2. Patent Overview

  • Title: Object Detection Sensor Apparatus and Associated Methods
  • Brief Description: The ’436 patent discloses methods and systems for a robot to detect objects by analyzing images. The core technique involves calculating a "contrast parameter" for an image, corresponding to the maximum absolute deviation of pixels from a reference value, and comparing this parameter to a dynamically configured threshold to determine an object's presence.

3. Grounds for Unpatentability

Ground 1: Obviousness over Rosenstein, Canny, and Xu - Claims 1-18 are obvious over Rosenstein in view of Canny and Xu.

  • Prior Art Relied Upon: Rosenstein (WO 2011/146259 A2), Canny (a 1986 IEEE Transactions paper on edge detection), and Xu (a 2011 IEEE workshop paper on a distributed Canny edge detector).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner asserted that Rosenstein taught a mobile robot with a controller, drive system, and imaging sensors for navigation and obstacle avoidance, expressly contemplating an "edge detection routine." However, Rosenstein did not detail the specific algorithm. Petitioner argued that the claimed method of determining object presence—by comparing a "contrast parameter" to a "dynamically configured threshold"—was disclosed by the combination of Canny and Xu, which describe a widely-used edge detection algorithm. Specifically, Xu's implementation of the Canny algorithm calculated a "gradient magnitude" for each pixel and applied non-maximum suppression (NMS) to find the local maximum gradient, which Petitioner mapped to the claimed "contrast parameter corresponding to a maximum absolute deviation of pixels...from a reference value." Xu further taught computing hysteresis high and low thresholds based on the histogram of gradient magnitudes for the entire image or based on local image block characteristics, satisfying the "dynamically configured threshold" limitation.
    • Motivation to Combine: A person of ordinary skill in the art (POSITA), seeking to implement Rosenstein's disclosed "edge detection routine," would have been motivated to use the well-known and high-performing Canny algorithm. A POSITA would have further looked to Xu's improvements on the Canny method to achieve an efficient implementation with reduced memory and latency, which are desirable characteristics for an autonomous mobile robot.
    • Expectation of Success: Given that the Canny algorithm was a standard, widely used method for edge detection in image processing, a POSITA would have had a high expectation of successfully integrating the Canny/Xu edge detection method into Rosenstein's robotic system to perform its intended object detection function.

Ground 2: Obviousness over Nourbakhsh and Tsutsumi - Claims 1-2, 6-7, 11, 13-14, and 16-17 are obvious over Nourbakhsh in view of Tsutsumi.

  • Prior Art Relied Upon: Nourbakhsh (Patent 5,793,900) and Tsutsumi (Patent 9,076,214).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued this combination taught the claimed invention by using focus-based object detection. Nourbakhsh disclosed a mobile robot with multiple cameras focused at different distances to generate a "depth map" for object detection and collision avoidance by identifying which image regions were in focus. Tsutsumi disclosed a similar system but provided a more detailed method for calculating a "contrast value" to determine focus. Tsutsumi calculated its contrast value using the maximum and minimum pixel values in an image area, which Petitioner mapped to the claimed "contrast parameter." This contrast value was then compared to a threshold that was dynamically adjusted based on the number of in-focus images found, satisfying the "dynamically configured threshold" limitation.
    • Motivation to Combine: Petitioner contended that Nourbakhsh and Tsutsumi disclosed very similar methods for generating depth information. A POSITA would have been motivated to replace Nourbakhsh's general sharpness calculation with Tsutsumi's specific and robust method for calculating a contrast value and dynamically adjusting the threshold. This substitution would predictably improve the accuracy of the depth map and the precision of the robot's navigation without requiring any undue experimentation.
    • Expectation of Success: The fundamental similarity between the two systems—both using multiple images at different focal lengths to determine depth based on contrast/sharpness—would have provided a POSITA with a strong and reasonable expectation that combining them would be successful and yield the predictable result of a more accurate object detection system.

4. Relief Requested

  • Petitioner requests institution of an inter partes review (IPR) and cancellation of claims 1-18 of the ’436 patent as unpatentable.