Underground Tunnel Floor Model Extraction for Route Planning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In complex underground environments like mines, using full 3D models for route calculation and location tracking is resource-intensive and inefficient, as it includes unnecessary data from walls and ceilings, which are not crucial for mobile object movement.

Innovation Solution

A method to generate a simplified floor model by identifying and extracting floor points from a 3D model based on surface normal directions and probability indicators, reducing the dataset to a 2D representation of the tunnel floor for use in route planning and positioning systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If full 3D models are used for route calculation and location tracking, then comprehensive environmental data is available, but computational resources are excessively consumed and processing efficiency decreases

Engineering Contradiction:
Improvecomprehensive environmental dataVSAvoidroute calculation efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent extracts only the floor surface points from the complete 3D point cloud model by analyzing surface normals and identifying points with vertical orientations. This extraction removes unnecessary wall and ceiling data while retaining the essential floor information needed for mobile object positioning and route planning, thereby reducing computational load while maintaining necessary environmental awareness.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the 3D point cloud data by categorizing points based on their surface normal directions. Floor points are separated from wall and ceiling points through geometric analysis, creating a segmented representation that focuses computational resources only on the relevant floor surface for navigation purposes.

Inventive Principle:
Principle #1Segmentation

2Loss of information

If full 3D models are used for route calculation and location tracking, then complete environmental representation is achieved, but data processing time and computational complexity increase

Engineering Contradiction:
Improveenvironmental representation completenessVSAvoiddata processing time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The patent extracts only the floor surface points from the complete 3D point cloud model by analyzing surface normals and identifying points with vertical orientations. This extraction removes unnecessary wall and ceiling data while retaining the essential floor information needed for mobile object positioning and route planning, thereby reducing computational load while maintaining necessary environmental awareness.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by processing only the necessary subset of point cloud data (floor points) rather than the complete 3D model. This selective processing approach reduces data processing time while maintaining sufficient environmental representation for navigation tasks, avoiding the excessive computation of irrelevant surface data.

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If simplified floor models are generated by extracting floor points, then computational efficiency is improved, but accuracy of surface representation may be reduced

Engineering Contradiction:
Improveroute calculation efficiencyVSAvoidsurface model accuracy
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent applies local quality by differentiating between floor points and non-floor points based on surface normal analysis. Each point in the point cloud is evaluated individually, and floor points are identified with high precision using geometric criteria (vertical surface normals). This ensures that the simplified floor model maintains accurate representation of the actual floor surface geometry while excluding irrelevant structures.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the parameter used for point selection from arbitrary or uniform sampling to surface normal direction analysis. By using the vertical orientation of floor surfaces as the selection criterion, the method accurately distinguishes floor points from wall and ceiling points, preserving the geometric fidelity of the floor model while simplifying the overall data structure.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3660266B1Model generation for route planning or positioning of mobile object in underground worksite
Publication Date: 2023.10.11 SANDVIK MINING & CONSTR OY
  • EP3660266B1 patent drawingFigure 1~2
  • EP3660266B1 patent drawingFigure 3~4
  • EP3660266B1 patent drawingFigure 5a~6b

AI summary

According to an example aspect of the present invention, there is provided a method, comprising: receiving a three-dimensional model of an underground tunnel, identifying floor points among points of the three-dimensional model; extracting the floor points, and applying at least a part of the extracted floor points as a floor model of the tunnel for positioning or route planning of a mobile object in the underground tunnel.