Mine Vehicle Path Classification Using Mesh-Based Route Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mine vehicle travel path classification methods are prone to errors due to GNSS inaccuracies and complex path deviations, leading to incorrect productivity evaluations and labor-intensive path setting updates, especially in dynamic mining environments.
Innovation Solution
A mine management system that uses mesh sets defined by IDs of equally divided ground surface areas to classify travel paths, employing a processing device to compute union and difference sets for accurate path classification and productivity indexing, with a display terminal for real-time data visualization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of comparison points is increased to improve path classification accuracy, then classification accuracy is improved, but setting of distance threshold values becomes more complicated and setting comparison points while avoiding interference with other specified paths becomes more difficult
Solution Approach 1:
The ground surface is divided into multiple mesh areas, each with a unique ID. The travel path is represented as a sequence of mesh IDs rather than continuous coordinate data. This segmentation simplifies the comparison process by breaking down the complex path matching problem into discrete mesh unit comparisons, allowing accurate classification without requiring complex threshold settings at multiple comparison points.
Solution Approach 2:
The patent changes the parameter representation from continuous GNSS coordinates to discrete mesh IDs. By transforming the path data into a sequence of mesh identifiers, the system simplifies the classification process and eliminates the need for complex distance threshold calculations at multiple comparison points, while maintaining high classification accuracy.
2Productivity
If GNSS position data is used for path classification, then real-time path tracking is enabled, but classification errors occur when the vehicle deviates from the specified path to avoid obstacles or rough roads
Solution Approach 1:
The patent creates a buffer zone by defining mesh areas that are slightly larger than the actual travel path. This allows the vehicle to temporarily deviate into adjacent mesh areas when avoiding obstacles or rough roads without being classified as taking a completely different path. The mesh-based approach provides a cushion that absorbs normal operational deviations while still detecting significant path changes.
Solution Approach 2:
Instead of directly comparing GNSS coordinates with the specified path, the patent creates a mesh model as an intermediate representation. The travel path is copied into the mesh framework, allowing for more flexible and tolerant path matching that accounts for real-world operational variations while maintaining accurate classification.
3Productivity
If traditional path classification methods are used, then path evaluation can be performed, but labor is required to update settings when dumping places or loading places are moved
Solution Approach 1:
The mesh-based path classification system automatically adapts when dumping or loading places are moved. The system self-updates by recalculating the mesh sequence based on the new positions, eliminating the need for manual intervention to adjust path settings. The automated mesh generation and comparison process handles dynamic changes in the mining operation without requiring operator input.
4Measurement precision
If multiple comparison points are set to improve classification accuracy, then path matching precision is improved, but the processing complexity for path classification increases
Solution Approach 1:
The patent segments both the ground surface and the travel path into discrete mesh units. This segmentation transforms the complex continuous path matching problem into a simpler discrete sequence comparison problem. The processing complexity is reduced because the system only needs to compare mesh ID sequences rather than calculate distances and angles at multiple continuous comparison points.
Solution Approach 2:
The patent replaces the mechanical geometric calculations (distance measurements, angle computations, threshold comparisons) with a simpler information processing approach using mesh ID sequences. This substitution of mechanical computation with information-based comparison significantly reduces processing complexity while maintaining or improving matching precision.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
For a mesh set A of an unclassified travel path being a travel path that has not yet been classified and a mesh set B of a classified travel path being a travel path that has been classified, a union set of the mesh set A and the mesh set B, a first difference set of the union set and the mesh set A, and a second difference set of the union set and the mesh set B are computed, the unclassified travel path is classified into a path ID of the classified travel path in a case in which the size of the first difference set and the size of the second difference set are both equal to or smaller than a threshold value, and the unclassified travel path is classified into a new path ID in a case in which at least one of the size the first difference set and the size of the second difference set is larger than the threshold value. As a result, the travel path of a mine vehicle is accurately and easily classified.