Lidar Data Correctness Verification via ROI Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately localizing themselves using Lidar sensor data due to uncertainties such as adverse weather conditions and potential damage or alignment issues, which existing mechanisms fail to correct in real-time.
Innovation Solution
A method and system that identify Regions of Interest (ROIs) in the Lidar sensor's Field of View, form clusters of reflection points, and determine the correctness of Lidar sensor data by comparing distance values between these clusters and navigation map obstacle points, allowing for real-time correction and navigation adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing mechanisms adjust model characteristics based on weather information, then the system can account for weather impact on sensors, but the mechanism fails to correct Lidar point data in real-time caused by natural obstacles, damage, or alignment issues
Solution Approach 1:
The patent divides the Lidar Field of View into multiple Regions of Interest (ROIs) and processes reflection points within each ROI separately. This segmentation allows the system to focus computational resources on specific areas where obstacles may be present, enabling real-time correction without processing the entire dataset, thus resolving the contradiction between reliability and complexity.
Solution Approach 2:
The patent applies different processing qualities to different regions by forming clusters of reflection points within each ROI and comparing cluster centers against navigation map obstacle points. This local quality approach allows targeted correction in affected regions while maintaining efficiency, addressing the real-time correction requirement without overwhelming system complexity.
2Measurement precision
If the system processes all Lidar reflection points comprehensively, then measurement precision improves, but processing time increases which hinders real-time navigation
Solution Approach 1:
By segmenting the Lidar data into multiple ROIs and processing clusters independently within each region, the system maintains measurement precision through thorough local analysis while reducing overall processing time through parallelizable operations. This resolves the contradiction between comprehensive processing and real-time requirements.
Solution Approach 2:
The patent applies partial action by focusing processing efforts only on regions where obstacles are detected or suspected, rather than uniformly processing all Lidar data. This allows the system to achieve sufficient measurement precision in critical areas while minimizing unnecessary processing time in clear regions, enabling real-time navigation.
3Reliability
If the system continuously monitors Lidar data for correctness, then navigation safety improves, but computational load increases
Solution Approach 1:
The continuous monitoring is achieved through segmentation into multiple ROIs with independent cluster formation and validation processes. This allows the system to maintain high navigation safety through continuous oversight while distributing computational energy requirements across multiple smaller, manageable tasks that can be executed efficiently and potentially in parallel.
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 accurate and real-time determination of Lidar sensor data correctness, ensuring safe navigation by discontinuing navigation if data is deemed incorrect, thereby improving the reliability of autonomous vehicle localization.
Implementation Method 1
The autonomous vehicles may be equipped with various types of sensors such as, Lidar, sonar, radar, cameras, and other sensors to detect objects in its environment
Data Source
AI summary
Disclosed herein is method and system for determining correctness of Lidar sensor data used for localizing autonomous vehicle. The system identifies one or more Region of Interests (ROIs) in Field of View (FOV) of Lidar sensors of autonomous vehicle along a navigation path. Each ROI includes one or more objects. Further, for each ROI, system obtains Lidar sensor data comprising one or more reflection points corresponding to the one or more objects. The system forms one or more clusters in each ROI. The system identifies a distance value between, one or more clusters projected on 2D map of environment and corresponding navigation map obstacle points, for each ROI. The system compares distance value between one or more clusters and obstacle points based on which correctness of Lidar sensor data is determined. In this manner, present disclosure provides a mechanism to detect correctness of Lidar sensor data for navigation in real-time.