Blind Spot Object Detection With Dynamic Vehicle Warning Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle sensor systems fail to adequately detect and warn about target objects in blind spot regions between observation areas, leading to continuous alarms even after the object has left the observation range.
Innovation Solution
An information processing device equipped with sensors and a processing unit that determines the presence of target objects in blind spot regions by analyzing data from multiple sensors, and outputs warning information when a target object is detected in these regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are installed to detect objects in observation regions, then object detection capability is improved, but blind spot regions between observation regions cannot be detected
Solution Approach 1:
The patent transitions from two-dimensional sensor coverage (front, rear, side cameras) to three-dimensional spatial awareness by inferring object positions in blind spot regions through coordinate transformation and positional relationships between multiple observation regions. This adds the dimension of blind spot coverage to the existing observation coverage.
Solution Approach 2:
The patent introduces an intermediary inference mechanism that uses detection data from observable regions as intermediate information to deduce the presence of objects in blind spot regions. The system acts as a mediator, translating known positions in observation regions into predicted positions in blind spot regions through spatial relationship analysis.
2Reliability
If alarm is continuously output when object is detected, then safety warning is improved, but unnecessary alarms occur after object leaves observation range
Solution Approach 1:
The patent implements feedback control by continuously monitoring object positions and dynamically adjusting alarm output based on real-time detection data. The system provides feedback to determine whether to maintain, start, or stop alarm output based on the object's current position and movement trajectory, ensuring alarms are only active when necessary.
Solution Approach 2:
The patent makes the alarm system dynamic by transitioning from static alarm conditions to dynamic alarm control based on real-time object position and movement analysis. The alarm state changes dynamically according to whether the object is in observation regions or blind spot regions and its predicted movement trajectory.
3Area of stationary object
If multiple sensors are used to cover observation regions, then detection coverage is improved, but blind spot regions remain undetected
Solution Approach 1:
The patent extends detection coverage from the two-dimensional areas covered by physical sensors to the three-dimensional space including blind spot regions through computational inference. By analyzing spatial relationships and transforming coordinates between observation regions, the system achieves awareness in regions without physical sensor coverage.
Solution Approach 2:
The patent makes the sensor system multi-functional by enabling it to perform both direct detection in observation regions and indirect inference in blind spot regions. The same sensor data serves dual purposes: direct object detection and inferential blind spot detection through positional relationship analysis.
Data Source
AI summary
An information processing device includes a determination unit and a first output control unit. The determination unit determines whether a target object is present in a blind spot region between observation regions around a mobility. The first output control unit outputs first warning information indicative of presence of the target object in the blind spot region in a case where the target object has been determined to be present in the blind spot region.