LIDAR False Return Filtering for Dust, Smoke, and Steam
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
LIDAR systems often produce false positive indications of surfaces due to reflections from fine particulate matter such as fog, smoke, and steam, leading to inaccurate distance measurements and potential safety issues in applications like autonomous vehicles.
Innovation Solution
The techniques involve identifying and suppressing false returns by analyzing output signals from LIDAR channels, using methods such as calculating reflectivity, comparing signal characteristics at different power and frequency settings, and employing machine learning models to differentiate between solid objects and particulate matter based on temporal and spectral characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LIDAR system detects all light reflections, then detection sensitivity is improved, but false positive rate increases due to particulate matter reflections
Solution Approach 1:
The system varies operating parameters including light source power levels and detection thresholds to differentiate between true surface reflections and particulate matter reflections. By changing these parameters dynamically, the system can identify patterns characteristic of false positives while maintaining sensitivity to true targets.
Solution Approach 2:
The system employs feedback mechanisms where detection results are continuously analyzed and used to adjust subsequent detection parameters. Machine learning models process detection patterns and provide feedback for refining the distinction between true returns and false positives from particulate matter.
2Reliability
If LIDAR system increases detection threshold to reduce false positives, then false positive rate decreases, but detection sensitivity deteriorates
Solution Approach 1:
The detection threshold is made dynamic rather than static, adjusting in real-time based on environmental conditions, signal characteristics, and machine learning model predictions. This allows the system to maintain high sensitivity when true targets are present while suppressing false positives from particulate matter through adaptive threshold modification.
Solution Approach 2:
The system changes detection parameters including threshold levels, integration times, and power settings based on analyzed signal characteristics. By dynamically adjusting these parameters, the system optimizes the balance between detecting true targets and rejecting false positives from fog, smoke, and other particulate matter.
3Measurement precision
If LIDAR system uses multiple detection methods to differentiate true returns from false returns, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system replaces complex mechanical or manual analysis methods with machine learning models and automated signal processing algorithms. These computational approaches efficiently analyze multiple signal characteristics simultaneously to differentiate true returns from false returns, reducing the need for complex hardware modifications while improving detection accuracy.
Solution Approach 2:
The signal processing system performs multiple functions including false positive detection, target identification, and parameter optimization using integrated machine learning models. This multi-functional approach consolidates what would otherwise require separate systems into a unified processing framework, managing complexity while enhancing measurement precision.
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
These methods significantly reduce false positives in LIDAR detections, enhancing the accuracy and safety of systems relying on LIDAR data by correctly distinguishing between true returns from solid objects and false returns from particulate matter.
Implementation Method 1
When the emitted light is incident on a surface, a portion of the light is reflected and received by the light sensor
Implementation Method 2
fine particulate matter may also reflect light. Problematically, fog, smoke, fog, exhaust, steam, and other such vapors may reflect light emitted by a LIDAR system
Implementation Method 3
the system may measure the propagation time of a light signal as it travels from the laser emitter, to the surface, and back to the light sensor. A distance is then calculated based on the flight time and the known speed of light
Data Source
AI summary
A machine-learned (ML) model for detecting that depth data (e.g., lidar data, radar data) comprises a false positive attributable to particulate matter, such as dust, steam, smoke, rain, etc. The ML model may be trained based at least in part on simulated depth data generated by a fluid dynamics model and/or by collecting depth data during operation of a device (e.g., an autonomous vehicle. In some examples, an autonomous vehicle may identify depth data that may be associated with particulate matter based at least in part on an outlier region in a thermal image. For example, the outlier region may be associated with steam.