Industrial Vehicle Localization Using Ceiling Light and Skylight Features

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing industrial vehicle navigation systems struggle with illumination variations, particularly in environments with mixed lighting conditions, such as skylights and artificial lights, leading to inaccurate feature detection and localization.

Innovation Solution

The system utilizes a Gaussian scale space pyramid and Hessian response analysis to distinguish between ceiling lights and skylights by calculating determinant and trace of Hessian responses, applying filters and machine learning techniques to identify valid features, and navigating the vehicle based on these identifications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional feature detection methods are used in mixed lighting environments, then the system is simple to implement, but the localization accuracy deteriorates due to illumination variations

Engineering Contradiction:
Improvelocalization accuracyVSAvoidfeature detection complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the feature detection process into distinct stages: Gaussian scale space pyramid construction, Hessian matrix calculation, determinant and trace computation, and classification filtering. This segmentation allows each stage to address specific aspects of illumination invariance, thereby improving localization accuracy while maintaining manageable system complexity through modular processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transforms the 2D image space into a multi-scale Gaussian pyramid structure, adding the scale dimension to the detection process. By computing Hessian responses across multiple scales and analyzing determinant-trace relationships in this extended dimensionality, the system achieves illumination invariance and improved localization accuracy without proportionally increasing computational complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If advanced Hessian response analysis is applied to distinguish light sources, then the feature detection accuracy is improved, but the computational complexity increases

Engineering Contradiction:
Improvefeature detection accuracyVSAvoidcomputational power consumption
Core Design Contradiction:
Measurement precisionVSPower

Solution Approach 1:

The patent performs preliminary Gaussian smoothing and scale space pyramid construction before Hessian matrix calculation. This preliminary action reduces the complexity of subsequent differential operations by pre-processing the image data, thereby achieving accurate feature detection while optimizing computational power consumption through staged processing.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent extracts and analyzes only the determinant and trace components of the Hessian matrix, rather than computing the full matrix spectrum. This extraction of critical features enables accurate distinction between ceiling lights and skylights while reducing computational power requirements by focusing only on the most informative aspects of the Hessian response.

Inventive Principle:
Principle #2Taking out (Extraction)

3Reliability

If multiple filters and machine learning techniques are applied to classify features, then the reliability of light source identification is improved, but the processing time increases

Engineering Contradiction:
Improveidentification reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies different filtering and classification techniques to different types of feature candidates based on their local characteristics. Ceiling light candidates undergo specific determinant-based filtering, while skylight candidates receive trace-based analysis. This localized approach to feature classification improves identification reliability for each light type while reducing overall processing time by avoiding uniform application of all filters to every candidate.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system implements feedback mechanisms where classification results from initial filtering stages inform subsequent processing decisions. Valid ceiling lights and skylights identified through determinant-trace analysis provide feedback that refines the feature set for navigation, improving identification reliability while optimizing processing time through iterative refinement rather than exhaustive filtering.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3344952B1Industrial vehicle with feature-based localization and navigation
Publication Date: 2025.12.10 CROWN EQUIP CORP
  • EP3344952B1 patent drawingFigure 1
  • EP3344952B1 patent drawingFigure 2
  • EP3344952B1 patent drawingFigure 3

AI summary

An industrial vehicle is provided comprising a drive mechanism, a steering mechanism, a vehicle controller, a camera, and a navigation module. The camera is communicatively coupled to the navigation module, the vehicle controller is responsive to commands from the navigation module, and the drive mechanism and the steering mechanism are responsive to commands from the vehicle controller. The camera is configured to capture an input image of a warehouse ceiling comprising elongated skylights characterized by different rates of image intensity change along longitudinal and transverse axial directions, and ceiling lights characterized by a circularly symmetric rate of image intensity change. The navigation module is configured to distinguish between the ceiling lights and the skylights and send commands to the vehicle controller for localization, or to navigate the industrial vehicle through the warehouse based upon valid ceiling light identification, valid skylight identification, or both.