Parking Area Layout Modeling for Earlier Slot Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing parking systems have high latency due to the need to confirm a detected gap is unoccupied after passing it, significantly increasing the overall execution time for parking maneuvers.
Innovation Solution
A method for determining a parking area layout by extracting parameters from sensing data, assigning confidence scores to layout models, and selecting the highest scoring model to control vehicle movements, allowing earlier detection of unoccupied parking slots and planning smoother trajectories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the existing parking system waits for confirmation that a detected gap is unoccupied after passing it, then the reliability of parking slot detection is improved, but the execution time for parking maneuvers is significantly increased
Solution Approach 1:
The system performs preliminary classification of detected gaps as parking slots using extracted parameters (gap length, position, orientation) before the vehicle actually passes them. By pre-processing and pre-classifying potential parking slots during the approach phase rather than waiting for post-passing confirmation, the system reduces waiting time while maintaining detection reliability through the use of confidence scores and multiple parameter validation.
2Measurement precision
If the parking system uses multiple sensors and perception stack to collect environmental information, then the measurement precision of parking slot detection is improved, but the device complexity is significantly increased
Solution Approach 1:
The system extracts only the essential parameters (gap length, position, orientation) needed for parking slot classification from the environmental information collected by multiple sensors. Rather than processing all sensor data in full detail, the system selectively extracts and processes only the critical parameters required for parking slot detection, thereby reducing computational complexity while maintaining detection precision.
3Reliability
If the parking system tracks slot occupancy status and plans path only after detecting unoccupied slots, then the reliability of parking maneuver planning is improved, but the productivity of the parking system is decreased
Solution Approach 1:
The system performs preliminary classification and occupancy assessment of detected gaps as potential parking slots before the vehicle reaches them. By pre-planning potential parking maneuvers and pre-assessing slot availability using extracted parameters and confidence scores, the system reduces the time required for real-time decision-making while maintaining planning reliability through validated parameter matching.
Data Source
AI summary
Disclosed herein are methods of determining a layout for a parking area. In an aspect, a method includes determining environmental information about a surrounding environment of a vehicle moving in a parking area from sensing data provided by a sensing system of the vehicle; extracting parameters associated with parking slots identified from the environmental information; determining layout models for the parking area based at least on the extracted parameters, each layout model including an underlaying parking slot template; assigning to each of the determined layout models a confidence score; and selecting the layout model of the parking area having the highest confidence score as a current layout model used to determine the layout, the selected layout model is updated over time based on the confidence scores assigned to the layout models generated from sensing data collected while the vehicle is moving.