Mobile Robot Localization Using Direction Descriptors and Mode Switching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current mobile robot navigation and localization methods face challenges with significant illumination changes and image variations, leading to errors in feature point matching and adaptive driving control, especially when inertial sensor data is unreliable.

Innovation Solution

A control method for mobile robots that employs local direction descriptors, using image processing units to extract feature points, generate descriptors, and switch between driving modes based on image quality and matching degrees, allowing for adaptive control through image and sensing information integration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional feature point descriptors are used for mobile robot navigation, then the system can process images, but the performance deteriorates when illumination changes significantly or image variations occur

Engineering Contradiction:
Improvefeature point matching accuracyVSAvoidillumination change sensitivity
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent transforms image patches into frequency domain representations using Fourier transforms, changing the parameter space from spatial domain to frequency domain. This transformation makes the descriptors invariant to illumination changes and rotational variations, as frequency domain representations capture essential structural information while filtering out intensity variations.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces traditional mechanical/optical feature detection methods with frequency domain analysis. Instead of relying on intensity-based feature detection that is sensitive to illumination, the system uses spectral analysis to extract rotationally invariant and illumination-resistant descriptors.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If inertial sensor data is used for driving control, then the robot can navigate using sensor fusion, but wrong recognition occurs when sensor errors accumulate

Engineering Contradiction:
Improveposition estimation accuracyVSAvoiddriving control reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where image-based feature point matching results are used to correct and verify inertial sensor-derived position estimates. When feature point matching succeeds, it provides feedback to reset accumulated drift errors from inertial sensors, creating a hybrid navigation system that leverages the strengths of both sensing modalities.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system prepares multiple candidate feature point matches and pre-validates them using geometric constraints and consistency checks before accepting position corrections. This beforehand validation cushions against wrong recognition errors by filtering out incorrect matches before they can corrupt the navigation state.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

3Adaptability or versatility

If adaptive driving control is implemented, then the robot can respond to different conditions, but the system complexity increases when multiple driving modes are required

Engineering Contradiction:
Improvedriving mode adaptabilityVSAvoidcontrol system complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic switching between different driving modes based on real-time assessment of image quality, feature point matchability, and sensor reliability. The system transitions from inertial-sensor-dependent navigation to image-based navigation when illumination changes or sensor drift is detected, providing adaptability without requiring permanently complex control structures.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2460629B1Control method for localization and navigation of mobile robot and mobile robot using same
Publication Date: 2022.06.29 YUJIN ROBOT
  • EP2460629B1 patent drawingFigure 1
  • EP2460629B1 patent drawingFigure 2(a)~2(b)
  • EP2460629B1 patent drawingFigure 3

AI summary

The present invention relates to a control method for the localization and navigation of a mobile robot and a mobile robot using the same. More specifically, the localization and navigation of a mobile robot are controlled using inertial sensors and images, wherein local direction descriptors are employed, the mobile robot is changed in the driving mode thereof according to the conditions of the mobile robot, and errors in localization may be minimized.