Robot Camera Edge Matching for 3D Model Localization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robot localization techniques face challenges in accuracy, computational costs, and the expense of beacons or sensors, making it difficult for robots to determine their pose in an environment effectively.

Innovation Solution

A method using a robot's camera to capture images, detect edges, and compare them to a 3D model of the environment to determine the robot's pose, employing edge detection algorithms and rendering techniques to refine the robot's location and orientation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If dead reckoning or beacon-based localization techniques are used, then the robot can determine its pose in the environment, but the localization accuracy is insufficient or the computational costs are high

Engineering Contradiction:
Improvelocalization accuracyVSAvoidcomputational costs
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and compares only the essential edge features from the camera image with the 3D model, rather than processing all image data. This selective extraction of key geometric features reduces computational complexity while maintaining localization accuracy through precise edge matching.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs preliminary rendering of the 3D model to generate expected edge positions before comparing with actual camera image edges. This pre-computation of reference edges enables faster and more accurate pose determination by eliminating the need for complex real-time optimization algorithms.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If active or passive beacons are deployed throughout the environment, then the robot can triangulate its pose, but the costs associated with placing beacons or landmarks are high

Engineering Contradiction:
Improvepose determination accuracyVSAvoidbeacon placement costs
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent creates a virtual copy of the environment through a 3D model that contains all necessary geometric information for localization. This digital replica replaces the need for physical beacons or landmarks, allowing the robot to determine its pose by comparing camera image edges with rendered edges from the 3D model, thereby eliminating beacon deployment costs entirely.

Inventive Principle:
Principle #26Copying

3Measurement precision

If specialized sensors are used for beacon detection, then the robot can perform localization, but the costs associated with these sensors are high

Engineering Contradiction:
Improvelocalization capabilityVSAvoidsensor costs
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent makes the standard camera sensor perform multiple functions: it captures images for navigation, obstacle detection, and localization by edge comparison with the 3D model. This multi-functional use of the camera eliminates the need for specialized beacon detection sensors, reducing hardware costs while maintaining precise localization capability through computational geometry matching.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3347875B1Localization of a robot in an environment using detected edges of a camera image from a camera of the robot and detected edges derived from a three-dimensional model of the environment
Publication Date: 2023.11.01 INTRINSIC INNOVATION LLC
  • EP3347875B1 patent drawingFigure 1
  • EP3347875B1 patent drawingFigure 2
  • EP3347875B1 patent drawingFigure 3

AI summary

Methods, apparatus, systems, and computer-readable media are provided for using a camera of a robot to capture an image of the robot's environment, detecting edges in the image, and localizing the robot based on comparing the detected edges in the image to edges derived from a three-dimensional ("3D") model of the robot's environment from the point of view of an estimated pose of the robot in the environment. In some implementations, the edges are derived based on rendering, from the 3D model of the environment, a model image of the environment from the point of view of the estimated pose - and applying an edge detector to the rendered model image to detect model image edges from the model image.