LiDAR Dirt Detection With Region-Specific Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor systems face challenges in quickly determining dirt attachment in regions where distance information changes frequently, leading to delayed detection due to the requirement of a predetermined number of abnormal points before dirt can be identified.
Innovation Solution
A sensor system with a LiDAR that divides the detection range into regions with different threshold values for determining dirt attachment, allowing for immediate detection by setting unique threshold values for each region based on the likelihood of distance information changes, enabling quicker identification of dirt in areas where it is easily determinable.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a predetermined number of abnormal points is required to determine dirt attachment, then determination accuracy is improved, but detection speed deteriorates in regions where distance information changes frequently
Solution Approach 1:
The patent applies local quality by setting different threshold values for different regions of the detection range. Regions where distance information is likely to change (e.g., regions with vehicles, pedestrians, buildings) are assigned higher thresholds, while regions where distance is stable (e.g., road surface, sky) are assigned lower thresholds. This allows the system to quickly detect dirt in stable regions while avoiding false positives in dynamic regions, thus resolving the contradiction between detection speed and accuracy.
2Reliability
If a high threshold is set for regions with frequent distance changes, then false dirt determination is reduced, but detection speed in those regions also decreases
Solution Approach 1:
The patent segments the detection range into multiple regions with different characteristics. By dividing the field of view into regions where distance information is likely to change and regions where it is stable, the system can apply appropriate threshold values to each segment. This segmentation allows the system to maintain high reliability in dynamic regions while achieving fast detection in stable regions, effectively managing the trade-off between reliability and detection time.
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
This approach enhances the speed and accuracy of dirt determination by setting region-specific threshold values, allowing for prompt identification of dirt in areas where it is easily detectable, while maintaining determination accuracy.
Implementation Method 1
a light emitting unit configured to emit light to a detection range via a transmissive part configured to transmit light, a light receiving unit configured to receive light emitted from the light emitting unit and reflected by hitting at an object
Implementation Method 2
a point group information output unit configured to output point group information including position information on the object and distance information to the object, based on the light received by the light receiving unit
Data Source
AI summary
A sensor system includes a LiDAR and a dirt determination unit. The LiDAR includes a light emitting unit configured to emit light to a detection range via a transmissive part configured to transmit light, a light receiving unit configured to receive light emitted from the light emitting unit and reflected by hitting at an object, and a point group information output unit configured to output point group information comprising position information on the object and distance information to the object, based on the light received by the light receiving unit. The dirt determination unit is configured to detect dirt attached to the transmissive part, based on the point group information. The detection range to which the light emitting unit emits light is divided into a plurality of regions including a first region and a second region different from the first region. The dirt determination unit is configured to determine, for each region, that dirt is attached to the region when a number of pieces of the point group information that continues to be in an abnormal state, where the distance information cannot be normally acquired at the same time, for a predetermined time or longer reaches a predetermined number or more. When the predetermined number for determining that dirt is attached in the first region is defined as a first threshold value and the predetermined number for determining that dirt is attached in the second region is defined as a second threshold value, the first threshold value is different from the second threshold value.