3D Point Cloud Obstacle Sensing for Dusty Work Machines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional obstacle sensing technologies in autonomously travelling dump trucks fail to accurately distinguish between road surface reflections and obstacles like rocks due to similar reflection intensities, leading to unwanted removal of valid measurement data and potential sensing failures.
Innovation Solution
A work vehicle equipped with a measurement sensor that captures three-dimensional point cloud information, an object sensor that identifies and deletes microparticle data based on distance and variance thresholds, and an object sensing section to detect obstacles accurately by processing the cleaned point cloud data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If measurement data is filtered based on reflection intensity thresholds to remove microparticles, then sensing precision is improved, but valid obstacle data (e.g., rocks) is incorrectly removed
Solution Approach 1:
The patent segments the measurement data processing into multiple stages: initial reflection intensity filtering, followed by spatial distribution analysis, and finally temporal consistency verification. This multi-stage segmentation allows progressive refinement of data quality while preserving valid obstacles through subsequent verification steps that check spatial patterns and temporal stability beyond simple intensity thresholds.
Solution Approach 2:
The patent introduces an intermediary verification mechanism that acts as a mediator between the reflection intensity filter and the final obstacle detection. This intermediary layer analyzes spatial distribution patterns and temporal consistency of measurement data, serving as a buffer that prevents premature rejection of valid obstacle data while maintaining removal of microparticles.
2Loss of information
If reflection intensity thresholds are lowered to preserve more measurement data, then data loss is reduced, but microparticle interference increases
Solution Approach 1:
The patent transitions from single-dimension reflection intensity filtering to multi-dimensional analysis by incorporating spatial distribution patterns and temporal consistency as additional dimensions. This dimensional expansion allows the system to distinguish microparticles from valid obstacles using combined criteria across multiple dimensions, reducing data loss while maintaining microparticle rejection.
Solution Approach 2:
The patent dynamically adjusts filtering parameters based on environmental conditions and measurement context. Rather than using fixed reflection intensity thresholds, the system modifies filtering criteria according to spatial patterns, temporal variations, and environmental context, enabling adaptive balance between data preservation and microparticle removal.
3Measurement precision
If strict filtering is applied to remove microparticles, then sensing accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary spatial distribution analysis and temporal consistency checks on measurement data before final obstacle determination. By conducting these verification actions in advance, the system prepares refined data sets that reduce the computational burden of final obstacle detection, improving overall processing efficiency while maintaining high sensing accuracy.
Solution Approach 2:
The patent implements dynamic processing that adapts filtering strictness based on environmental conditions and operational context. The system adjusts the intensity of multi-stage filtering operations according to real-time conditions, applying more rigorous processing only when necessary, thereby balancing sensing accuracy with processing time efficiency.
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
Enhances the precision of obstacle sensing by effectively distinguishing microparticles from obstacles, reducing sensing errors and ensuring safe navigation in off-road environments.
Implementation Method 1
a TOF distance image sensor that measures a distance on the basis of a temporal difference between emitted light and reflected light
Implementation Method 2
by setting a threshold on the basis of reflection intensities that are obtained by measurement of the ground, measurement data about a road surface, and measurement data about microparticles such as dust and water vapors are removed
Data Source
AI summary
A measurement sensor that measures, as three-dimensional point cloud information having a plurality of vertically-adjacent layers, a surface position of an object around the work machine; and an object sensor that senses an object around the work machine on a basis of information from the measurement sensor are included, and the object sensor acquires three-dimensional point cloud information by measurement by the measurement sensor; senses, as point data, a point where microparticles are measured, based on a relation between distances, from the measurement sensor, of point data of vertically-adjacent layers and variations of distance differences, regarding a plurality of pieces of point data included in the three-dimensional point cloud information; deletes the point data of the point sensed as the point where the microparticles are measured, from the three-dimensional point cloud information; and senses an object around the work machine on a basis of the three-dimensional point cloud information.