Vehicle Route Selection Using Static Safe Area Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving systems using deep learning for environmental recognition and route calculation are limited to a functional safety level of around ASIL-B, which is insufficient for achieving the higher ASIL-D level required for advanced autonomous driving applications.
Innovation Solution
An automotive arithmetic device that combines deep learning for vehicle external environment estimation with a target object recognition system using predetermined rules to set safe areas and determine motor vehicle motion, ensuring routes are within safe areas, thereby enhancing the functional safety level to ASIL-D.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If deep learning is used for vehicle external environment estimation and route calculation, then the autonomous driving function is enabled, but the functional safety level remains at ASIL-B which is insufficient for advanced autonomous driving applications
Solution Approach 1:
The system is divided into two independent recognition systems: a deep learning-based system for environmental estimation and a rule-based system for target object recognition. Each system operates independently to generate candidate routes, which are then integrated through a route selection mechanism. This segmentation allows each subsystem to be optimized for its specific function while maintaining overall system safety at ASIL-D level.
Solution Approach 2:
Different recognition methods are applied to different aspects of the environment: deep learning is used for general environmental estimation while rule-based methods are specifically applied to critical target object recognition. This local differentiation ensures that safety-critical functions use the most reliable methods appropriate for each specific task.
2Reliability
If a single deep learning system is used for route calculation, then the system is simpler to implement, but it cannot achieve the required ASIL-D functional safety level
Solution Approach 1:
The system merges deep learning-based environmental estimation with rule-based target object recognition into a unified autonomous driving system. Both systems process sensor data independently and their results are combined through a route selection mechanism that chooses the safest route from multiple candidates, achieving ASIL-D safety level through complementary strengths of different approaches.
Solution Approach 2:
A route selection mechanism acts as an intermediary between the deep learning system and rule-based system. This mediator receives candidate routes from both systems, evaluates them against safety criteria, and selects the final route that maximizes functional safety while maintaining system implementability.
3Reliability
If deep learning alone is used for environmental recognition, then processing is faster, but the functional safety level is insufficient for ASIL-D requirements
Solution Approach 1:
The system performs partial deep learning processing for environmental estimation while using rule-based methods for critical target recognition. This partial application of computationally intensive deep learning, combined with faster rule-based verification, achieves ASIL-D safety levels while maintaining acceptable route calculation speeds through selective use of processing methods.
Data Source
AI summary
An automotive arithmetic device includes circuitry that calculates a first candidate route based on a vehicle external environment estimated using deep learning; sets a static safe area (SA2) based on a result of recognition of a target object outside a vehicle according to a predetermined rule; and determines a target motion of the motor vehicle. The circuitry selects the first candidate route as a travel route of the motor vehicle under a condition the first candidate route is entirely within the static safe area (SA2), and does not select the first candidate route as the travel route of the motor vehicle when the first candidate route at least partially deviates from the static safe area (SA2).