Parking Assistance Using Reliable Learned Object Positions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The precision of relative position calculations in parking assistance systems deteriorates when the stored position of a movable target object changes.
Innovation Solution
A parking assistance method that stores positions of multiple target objects around a target parking position, counts the number of coincidences between stored and current target object positions, and assigns higher reliability to learned target object positions with more coincidences, allowing for more accurate relative position calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the position of target objects is stored for parking assistance calculation, then the relative position calculation can be performed, but the calculation precision deteriorates when the stored target object position changes
Solution Approach 1:
The system performs preliminary actions by detecting and storing target object positions multiple times before the actual parking assistance calculation. By accumulating position data in advance and determining reliability based on whether multiple detections coincide, the system prepares reliable reference data beforehand, preventing precision deterioration when movable objects are present
Solution Approach 2:
The system implements feedback by comparing newly detected target object positions with previously stored positions. When multiple detections coincide within a predetermined range, the system confirms the position as reliable. This feedback mechanism continuously validates and updates the reliability of stored positions, ensuring accurate relative position calculation even when movable objects change position
Data Source
AI summary
A parking assistance method includes: storing positions of target objects detected around the target parking position as learned target object positions; when the own vehicle travels in a vicinity of the target parking position after the learned target object positions are stored, counting the number of times that the learned target object position coincides with a surrounding target object position of a target object detected around the own vehicle and providing a higher degree of reliability to the learned target object position having a larger number of times of coincidence with the surrounding target object position; by comparing the learned target object position having a degree of reliability greater than or equal to a predetermined threshold degree of reliability with a position of a target object detected around the own vehicle, calculating a relative position of the own vehicle with respect to the target parking position.