Cross Traffic Alert Parking Angle Trajectory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle cross traffic alert systems are limited in accuracy due to variations in parking space configurations and infrastructure, leading to false alerts and missed targets, necessitating a tradeoff between alert reliability and accuracy.
Innovation Solution
A method that determines the parking angle of a host vehicle relative to a target vehicle's trajectory, allowing the system to focus only on relevant areas of the sensor's field of view, thereby reducing false alerts and improving target detection by activating cross traffic alerts only when objects are in the designated 'portion of interest'.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the sensor field of view is expanded to cover more parking configurations, then the system can detect more potential targets, but false alerts increase due to extraneous objects
Solution Approach 1:
The patent applies local quality by dynamically adjusting the sensor field of view based on the detected parking angle. Instead of using a fixed wide field of view that captures all extraneous objects, the system calculates the specific parking angle and restricts the sensor monitoring to only the relevant angular sector. This localized approach ensures comprehensive coverage of actual cross-traffic paths while excluding irrelevant areas, thereby reducing false alerts while maintaining high target detection accuracy
Solution Approach 2:
The patent implements dynamics by making the sensor field of view adaptive rather than static. The system continuously determines the parking angle based on the relationship between the host vehicle trajectory and target vehicle trajectory, then dynamically adjusts the monitored angular sector accordingly. This dynamic adjustment allows the system to optimize its detection coverage for each specific parking configuration, resolving the contradiction between comprehensive detection and false alert reduction
2Reliability
If the system monitors all areas in the sensor field of view, then no targets are missed, but false alerts are generated from non-threatening objects
Solution Approach 1:
The patent applies local quality by precisely defining the 'portion of interest' within the sensor field of view based on the calculated parking angle. The system identifies which angular sectors correspond to actual cross-traffic paths and restricts alert generation to objects detected in those specific sectors. This localized monitoring approach ensures that no genuine threats are missed while automatically excluding non-threatening objects from areas outside the relevant angular sector
Solution Approach 2:
The patent applies segmentation by dividing the sensor field of view into distinct portions: a 'portion of interest' where objects trigger alerts and a 'portion not of interest' where objects are ignored. This segmentation is based on the calculated parking angle and creates clear boundaries for alert generation. The system processes objects in different segments differently, ensuring comprehensive monitoring of critical areas while filtering out irrelevant detections
Data Source
AI summary
The subject of the present invention is a host vehicle that determines a parking angle while backing out of a parking space and employs this parking angle when using a cross traffic alert system. The parking angle is determined based on the trajectories of the host vehicle and a target vehicle. The parking angle is used to determine areas of interest and areas not of interest within the fields of view of vehicle mounted sensors. The areas of interest are those used to determine if cross traffic alerts need to be issued.