Intersection Environmental Model Using Radar Topology Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intelligent intersection systems face challenges in accurately sensing and classifying objects such as vehicles and pedestrians due to the complexity of intersection environments, which affects the performance of traffic control and accident detection functions.
Innovation Solution
The system employs a central processing unit that receives and fuses raw radar data from multiple sensors, combined with lane, crosswalk, and sidewalk topology data to create an object list, using algorithms like Kalman filters or neural networks to classify objects as vehicles or pedestrians, forming an environmental model for intelligent intersection functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple radar sensors are used to sense objects in the intersection, then the coverage and detection capability are improved, but the data processing complexity and computational load increase
Solution Approach 1:
The patent segments the object detection and classification process into distinct modules: radar data reception, object creation from fused radar data, and object classification using topology data. This segmentation allows each module to handle specific tasks independently, reducing overall system complexity while maintaining detection accuracy.
Solution Approach 2:
The patent introduces an intermediary environmental model that integrates radar sensor data with pre-stored topology data (lane, crosswalk, sidewalk information). This environmental model serves as a mediator that simplifies the classification process by providing contextual information about object locations relative to intersection features, reducing the computational burden on the classification algorithm.
2Measurement precision
If object classification algorithms are made more sophisticated to improve classification accuracy, then the classification precision is improved, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary actions by pre-storing topology data (lane definitions, crosswalk locations, sidewalk boundaries) before the object classification process. This pre-prepared contextual information allows the classification algorithm to quickly determine object types by comparing sensor data against known intersection layouts, significantly reducing processing time while maintaining high classification accuracy.
Solution Approach 2:
The patent changes the approach from complex feature-based classification to a parameter-based method that uses object position coordinates and compares them against pre-defined topology parameters. By transforming the classification problem into a geometric parameter comparison task, the system achieves high accuracy with reduced computational complexity and faster processing.
3Reliability
If the system creates detailed environmental models with multiple object attributes, then the reliability of intelligent intersection functions is improved, but the data storage and processing requirements increase
Solution Approach 1:
The patent extracts only the essential attributes needed for intelligent intersection functions from the raw sensor data, such as object position, velocity, and classification (vehicle, pedestrian, cyclist). By extracting and storing only these critical parameters in the environmental model rather than all possible sensor measurements, the system maintains high reliability for traffic control and accident detection while minimizing data storage requirements.
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 enables accurate and efficient classification of objects, enhancing the system's ability to perform intelligent intersection functions, such as traffic control and accident detection, by creating a reliable environmental model without relying on maps.
Implementation Method 1
sensors for sensing and classifying objects in and around the intersection
Data Source
AI summary
An intelligent intersection method includes receiving raw sensor data from a sensors mounted relative to a street intersection. The received raw sensor data is fused to create at least one object sensed by the sensors. An object list is created or updated with information pertaining to created object, the object list serving as an environmental model. One or more intelligent intersection functions is subsequently performed based in part upon the environmental model. The method may further include determining whether the created object is associated with first data defining a topology of at least one of a plurality of lanes, a crosswalk or a sidewalk corresponding to the intersection. Upon an affirmative determination that the created object is associated with the first data, the method classifies the created object as a vehicle or a pedestrian for use as an attribute of the created object in the object list.