Parking Object Alerts Using Prior Detection Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vehicle parking systems fail to adequately alert users to undetected objects around parking spaces, particularly during automatic parking, which can lead to safety risks.
Innovation Solution
An information processing apparatus that stores detection object information from previous parking and compares it with current data to identify undetected objects, adjusting the notification alert level based on the presence of humans or animals, enhancing user awareness through varying alarm volumes, types, or display patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the notification system uses a fixed alert level for all detected objects, then the system is simple to operate, but it cannot adequately distinguish between different types of objects (e.g., humans, animals, inanimate objects) leading to insufficient safety warnings
Solution Approach 1:
The notification system applies different alert levels and notification methods based on the type of detected object. Humans trigger high-priority notifications with multiple alert methods, animals trigger medium-priority notifications, and inanimate objects trigger low-priority or no notifications. This local differentiation of notification quality based on object type resolves the contradiction by making the system reliable for important objects while keeping the overall structure manageable.
Solution Approach 2:
The system changes notification parameters (alert level, notification method, urgency) based on the detected object's characteristics. By dynamically adjusting these parameters according to object type classification, the system achieves high reliability for safety-critical objects without requiring a completely complex system architecture, as the same notification infrastructure is used with varying parameters.
2Reliability
If the system sends notifications for every detected object, then safety monitoring is comprehensive, but it causes user inconvenience and alarm fatigue when no actual hazards are present
Solution Approach 1:
The system applies different notification behaviors to different object types: high-priority notifications for humans, medium-priority for animals, and low-priority or suppressed notifications for inanimate objects. This selective notification approach ensures comprehensive safety monitoring for actual hazards while preventing user inconvenience from false or low-priority alerts.
Solution Approach 2:
The notification system combines multiple filtering and classification layers to process detected objects. By composite-ing object detection, classification, and selective notification in a multi-stage process, the system achieves both comprehensive safety monitoring and user convenience by filtering out non-critical alerts while maintaining alertness to real hazards.
3Reliability
If the notification aspect remains unchanged regardless of object type, then the system is easy to implement, but it fails to provide differentiated warning levels for different hazard levels
Solution Approach 1:
The system changes notification parameters (volume, type, urgency, repetition) based on the classified object type. This parameter-based differentiation allows the same notification infrastructure to provide multiple warning levels, achieving reliable differentiated warnings without proportionally increasing system complexity.
Solution Approach 2:
The notification system is designed to handle multiple object types and warning levels using a single unified notification mechanism. By making the notification system multi-functional to handle different alert levels, object types, and notification methods through parameter adjustment rather than separate hardware systems, the patent achieves differentiated warning capability while controlling overall system complexity.
4Measurement precision
If the system only detects objects in the current parking attempt, then the detection process is simple and fast, but it cannot identify undetected objects by comparing with previous parking data
Solution Approach 1:
The system performs preliminary detection and stores object information from previous parking attempts. By having detection data ready in advance from prior parkings, the system can quickly compare current detection results with historical data to identify undetected objects, achieving high detection accuracy without excessive processing time during the actual parking operation.
Solution Approach 2:
The system creates a copy of detection data from previous parking attempts and stores it for comparison. By using copied historical detection data rather than requiring complete re-detection, the system achieves improved object detection accuracy through comparison while minimizing additional processing time, as the comparison operation is computationally efficient compared to full re-detection.
Data Source
AI summary
An information processing apparatus for performing notification to a user of a vehicle when an object is detected around a parking space is provided. The information processing apparatus includes a storage that stores a traveling route to the parking space in a first parking, and a processor that acquires first detection object information around the parking space in the first parking and second detection object information around the parking space in a second parking performed after the first parking. The first parking is performed by manual driving, and the second parking is performed by automatic parking based on the first parking. An aspect of the notification in the first parking and the second parking are different when an undetected object, that was not detected during the first parking, is detected in the second parking.