LiDAR-Fused Hyperspectral Mapping for Real-Time Ore-Grade Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing commercially available hyperspectral sensors are not Fit-For-Purpose (FFP) for routine mine face mapping, requiring significant system customization to convert raw data streams to geo-spatially accurate geological face maps.
Innovation Solution
An integrated geological mapping platform using a scanning system with a hyperspectral imager, LiDAR, and computational resources to fuse rock face position and hyperspectral image data, mitigating sensor errors and accounting for lighting conditions, and utilizing machine learning for real-time ore grade detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If commercially available off-the-shelf hyperspectral sensors are used, then device complexity is reduced, but measurement precision and geo-spatial accuracy are insufficient for routine mine face mapping
Solution Approach 1:
The system divides the hyperspectral sensor into functional modules: a standard COTS hyperspectral sensor for spectral data acquisition, a separate LiDAR unit for spatial data, and a processing system that integrates both. This segmentation allows each component to be optimized independently while achieving overall geo-spatial accuracy through coordinated operation and data fusion.
Solution Approach 2:
The patent introduces an intermediary processing system that acts as a mediator between the COTS hyperspectral sensor and the final geo-spatially accurate map. This intermediary layer includes calibration modules, coordinate transformation algorithms, and data fusion techniques that convert raw sensor data into accurate geological face maps without requiring the sensor itself to be customized.
2Measurement precision
If significant system customization is performed to convert raw data streams to geo-spatially accurate maps, then measurement precision improves, but device complexity increases
Solution Approach 1:
The system creates a universal processing platform that can handle multiple functions: spectral data acquisition, spatial data acquisition, data fusion, calibration, and geo-spatial mapping. By designing the system to perform all these functions within a unified architecture, the patent reduces the need for separate customized systems while maintaining high measurement precision.
Solution Approach 2:
The patent employs parameter changes through calibration processes, where the system adjusts geometric parameters, radiometric parameters, and coordinate system parameters to transform raw sensor data into geo-spatially accurate maps. These parameter transformations are implemented through standardized algorithms that maintain precision without requiring complex hardware customization.
3Productivity
If real-time processing is implemented for ore grade detection, then productivity improves, but computational resources and processing complexity increase
Solution Approach 1:
The system performs preliminary actions by pre-calibrating the hyperspectral sensor and LiDAR unit, pre-establishing the relationship between sensor data and geo-spatial coordinates, and pre-processing algorithms for ore grade detection. These preliminary preparations enable real-time processing during actual mining operations without requiring excessive computational resources at the time of detection.
Solution Approach 2:
The patent replaces complex mechanical/customized sensor systems with a standardized COTS hyperspectral sensor combined with software-based processing algorithms. By substituting hardware customization with software-based solutions for data fusion and ore grade detection, the system achieves real-time processing capabilities without proportionally increasing device complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables real-time, accurate orebody knowledge for precision ore mining, enhancing mine yield, optimizing material flow, and supporting autonomous mining operations.
Implementation Method 1
scanning the rock face with the range determination unit to determine rock face position information
Implementation Method 2
scanning the rock face with the hyperspectral imager to produce a corresponding rock face hyperspectral image
Data Source
AI summary
The present invention relates to a method for detecting changes in the ore grade of a rock face in near real time. The method includes the step of providing a scanning system having at least a hyperspectral imager, a position system, a LiDAR or range determination unit and computational resources. Further, the method involves determining a precise location of the scanning system utilising the position system. The rock face is scanned with the range determination unit to determine rock face position information. The method involves scanning the rock face with the hyperspectral imager to produce a corresponding rock face hyperspectral image. Further the method involves utilising the computational resources to fuse together the rock face position information and the corresponding rock face hyperspectral image to produce a rock face position and content information map of the rock face.