Environment Recognition Processor with Abnormality Detection Cores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing external environment recognition systems for mobile objects face increased circuit size and power consumption due to the need for redundancy in abnormality detection systems, which is inefficient and resource-intensive.
Innovation Solution
An external environment recognition device is designed with a recognition processor that includes multiple arithmetic cores, where one core performs recognition processing and another core detects abnormalities in the data processing system, allowing for abnormality detection without the need for redundant configurations, thereby reducing circuit size and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If redundancy is provided in the data processing system for abnormality detection, then abnormality detection capability is improved, but circuit size and power consumption increase
Solution Approach 1:
The data processing system is segmented into a recognition processing core for normal operations and a separate abnormality detection core for monitoring. This segmentation allows the abnormality detection function to operate independently without requiring complete redundancy of the entire system, thereby reducing power consumption while maintaining detection capability.
Solution Approach 2:
An abnormality detection core acts as an intermediary component that monitors the output of the recognition processing core. Instead of duplicating the entire processing system, this intermediary core receives processed data and performs abnormality detection, reducing the overall circuit size and power requirements compared to full redundancy.
2Reliability
If redundancy is provided in the data processing system for abnormality detection, then abnormality detection capability is improved, but circuit size increases
Solution Approach 1:
The system is divided into distinct functional segments: a recognition processing core for primary data processing and an abnormality detection core for monitoring. This segmentation eliminates the need for complete system redundancy, reducing circuit size while preserving abnormality detection functionality.
Solution Approach 2:
Instead of creating a full copy of the data processing system for redundancy, the patent implements a simplified abnormality detection core that copies only the essential monitoring functions. This selective copying approach reduces circuit complexity while maintaining the ability to detect abnormalities in the recognition processing output.
Data Source
AI summary
A recognition processor performs recognition processing to recognize an external environment of a mobile object, based on image data acquired by an imaging unit that takes an image of an external environment of the mobile object. The recognition processor includes a plurality of arithmetic cores. The plurality of arithmetic cores include a recognition processing core that performs the recognition processing, and an abnormality detection core that detects an abnormality of a data processing system including the imaging unit and the recognition processor, based on the abnormality of the output from the recognition processing core.