Automatic Parking Control With Adaptive Feature Point Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic parking systems fail to account for changes in feature point information due to seasonal changes or the passage of time, leading to potential misidentification of parking spaces and inability to perform accurate automatic parking.
Innovation Solution
A moving body control device that includes an outside recognition unit to acquire data, a reception unit to receive parking position designation, and a control unit to perform automatic parking based on recognition data, registering feature points and updating them based on a match rate between new and registered points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If automatic parking control is performed based on stored feature point information, then parking automation is achieved, but the system fails when feature points change due to seasonal changes or passage of time
Solution Approach 1:
The system dynamically updates feature point information by comparing newly acquired feature points with stored feature points. When the number of unmatched feature points exceeds a threshold, the system acquires new feature points and updates the stored information, allowing the system to adapt to seasonal changes and temporal variations in the parking environment.
Solution Approach 2:
The system implements a feedback mechanism where the automatic parking control unit compares newly acquired feature points with stored feature points, evaluates the match quality using a threshold criterion, and decides whether to update the stored feature point information based on this comparison, ensuring continuous improvement of parking recognition accuracy.
2Measurement precision
If feature point information is stored for automatic parking, then parking position recognition is enabled, but the stored information becomes outdated over time
Solution Approach 1:
The system performs preliminary comparison of feature points before executing automatic parking control. By comparing newly acquired feature points with stored feature points and evaluating whether the number of unmatched points exceeds a threshold, the system prepares updated feature point information in advance, ensuring accurate parking position recognition even after long periods.
Solution Approach 2:
The system changes the parameter of feature point information by updating the stored feature points with newly acquired feature points when the match quality deteriorates below a threshold. This parameter change allows the system to maintain accurate parking position recognition despite temporal changes in the environment.
3Device complexity
If the system uses fixed feature point information, then processing simplicity is maintained, but adaptability to environmental changes is lost
Solution Approach 1:
The system provides self-service by automatically comparing newly acquired feature points with stored feature points and autonomously deciding whether to update the stored information based on a threshold criterion. This self-service mechanism enables the system to adapt to seasonal changes without requiring manual intervention, maintaining simple feature point management while achieving high adaptability.
Data Source
AI summary
A moving body control device includes an outside recognition unit, a reception unit, and a control unit. The control unit is configured to register feature points related to the designated parking position based on the recognition data of the outside as registered feature points, perform the automatic parking control for parking the moving body at the designated parking position based on the recognition data of the outside and the registered feature points, and update the registered feature points according to a match rate between the feature points obtained based on the recognition data of the outside when the automatic parking control is executed and the registered feature points.