Vehicle Sensor Fusion Control for Travel-Restricted Section Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems struggle to accurately recognize travel-restricted sections on roads, such as those affected by construction or accidents, which can compromise safety during automatic driving and driving assistance operations.
Innovation Solution
A control device equipped with a processor that acquires output information from both radar and camera sensors to perform dual processing for recognizing specific objects, like pylons, within predefined ranges. The processor determines the reliability of the recognition results by comparing the outputs from these sensors, thereby improving the accuracy of travel-restricted section detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single sensor is used to detect objects and recognize travel-restricted sections, then the device complexity is reduced, but the measurement precision and reliability of recognition results deteriorate
Solution Approach 1:
The patent combines multiple sensors (first sensor and second sensor) to perform simultaneous detection of the same object. The processor integrates detection results from both sensors to determine the presence of travel-restricted sections, thereby improving measurement precision while maintaining manageable device complexity through systematic integration.
Solution Approach 2:
The processor acts as an intermediary that receives detection results from multiple sensors and synthesizes them into a reliable recognition result. By using the processor to mediate between multiple sensor inputs and the final determination, the system achieves higher precision without proportionally increasing overall system complexity.
2Measurement precision
If multiple sensors are used to detect objects and improve recognition accuracy, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
Multiple sensors are configured to perform the same detection function (detecting objects in the surrounding region). This multi-functionality approach allows the system to improve measurement precision through redundant detection while managing complexity by having each sensor fulfill a universal detection role rather than requiring specialized different functions.
Solution Approach 2:
The processor uses detection results from the first sensor to inform and validate detection results from the second sensor. This feedback mechanism allows the system to cross-verify detections, improving precision while managing complexity through intelligent information processing rather than simple hardware addition.
3Reliability
If the recognition range is expanded to cover more areas, then the reliability of detecting travel-restricted sections improves, but the loss of information increases due to processing larger data volumes
Solution Approach 1:
The peripheral region is divided into multiple predetermined ranges, and the detection process is segmented into identifying objects in each range. This segmentation allows the system to expand the overall recognition coverage while managing information processing by handling smaller, divided data units rather than processing a single large data set.
Solution Approach 2:
The system performs detection in multiple predetermined ranges that may overlap or extend beyond the minimum necessary coverage. This partial or excessive action ensures that travel-restricted sections are reliably detected even when they appear at range boundaries, while the processor manages the resulting data volume through efficient integration of results from multiple ranges.
Data Source
AI summary
A control device includes a processor configured to acquire output information of a first type of a first sensor and output information of a second type of a second sensor each configured to detect an object in an around of a vehicle, the processor performing first processing of recognizing a first range including a specific object among a plurality of ranges set in a peripheral region of the vehicle based on the output information of the first sensor, and second processing of recognizing a second range including the specific object among the plurality of ranges based on the output information of the second sensor, selecting a third range from the plurality of ranges obtained by excluding the first range based on the first range, and determining reliability of a recognition result of the first range based on a recognition result of the second processing for the third range.