AV Lidar Point-Cloud Segmentation for Particulate Matter Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately identifying and distinguishing particulate matter from solid objects using existing sensing technologies, leading to potential safety issues due to misrecognition, reduced visibility, and sub-optimal driving decisions.
Innovation Solution
The implementation of lidar-assisted identification and classification systems that utilize time-of-flight, coherent Doppler-assisted lidar data, and intensity mapping to differentiate particulate matter from rigid objects by analyzing velocity distributions and spatial dispersion, enabling precise segmentation of point clouds and determination of wind velocity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing sensing technologies are used to detect objects in the environment, then the autonomous vehicle can obtain basic sensing data, but the vehicle cannot accurately distinguish particulate matter from solid objects, leading to reduced navigation safety and precision
Solution Approach 1:
The patent segments the detection task by separating particulate matter detection from solid object detection through velocity-based classification. The system divides return points into different categories based on velocity thresholds, allowing distinct processing paths for particulate matter (with random velocity patterns) versus solid objects (with consistent velocity patterns), thereby improving both detection accuracy and navigation safety
Solution Approach 2:
The patent introduces velocity information as an intermediary parameter to mediate between the sensing system and object classification. By adding this intermediate layer of analysis, the system can differentiate between particulate matter and solid objects that appear similar in spatial distribution, resolving the contradiction between basic detection capability and accurate classification
2Reliability
If the autonomous vehicle uses basic sensing data for path determination, then the system can operate with simpler processing, but the precision and safety of the driving path are compromised due to inability to identify particulate matter
Solution Approach 1:
The patent applies dynamics by making the detection system adaptive through velocity-based classification. The system dynamically adjusts its interpretation of sensing data based on velocity patterns, switching between particulate matter detection mode and solid object detection mode. This dynamic approach improves driving path safety without requiring a completely different sensing system, thus managing complexity
Solution Approach 2:
The patent changes the parameter used for object classification from purely spatial (distance, position) to include temporal velocity information. By incorporating velocity as an additional parameter, the system can identify particulate matter and adjust driving paths accordingly, improving safety while using the same basic sensing hardware
3Ease of operation
If the system processes all sensing data as potential obstacles, then no particulate matter identification is performed, but this leads to unsafe driving maneuvers due to misclassification of particulate matter as solid objects
Solution Approach 1:
The patent segments the processing workflow by creating distinct branches for particulate matter and solid object processing. Return points are divided based on velocity characteristics, with particulate matter points routed to one processing path and solid object points to another. This segmentation maintains operational simplicity while preventing misclassification errors that would compromise driving maneuver safety
Solution Approach 2:
The patent inverts the conventional approach by not assuming all detected objects are solid obstacles. Instead, it starts with the assumption that moving targets may be particulate matter and applies velocity-based filtering to distinguish them. This inverted logic prevents unsafe driving maneuvers caused by misclassification while keeping processing straightforward
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 accuracy of object recognition, improves driving path determination, and optimizes vehicle control by correctly identifying particulate matter, reducing the likelihood of unsafe maneuvers and improving visibility assessment.
Implementation Method 1
sensing system of an autonomous vehicle (AV) obtains a plurality of return points, each return point comprising one or more velocity values and one or more coordinates of a reflecting region that reflects a signal emitted by the sensing system
Implementation Method 2
utilizing Time-of-Flight (ToF) and coherent Doppler-assisted lidar data to differentiate between rigid objects and particulate matter
Implementation Method 3
utilizing Time-of-Flight (ToF) and coherent Doppler-assisted lidar data to differentiate between rigid objects and particulate matter
Data Source
AI summary
Aspects and implementations of the present disclosure address shortcomings of the existing technology by enabling lidar-assisted segmentation and identification of particulate matter in autonomous vehicle (AV) applications, by: obtaining, by a sensing system of the AV, a plurality of return points, each return point having one or more velocity values and one or more coordinates of a reflecting region that reflects a signal emitted by the sensing system, identifying, in view of the one or more velocity values of each of a first set of the return points of the plurality of return points, that the first set of the return points is associated with a particulate matter in an environment of the AV, and causing a driving path of the AV to be determined in view of the particulate matter.