Modular Vehicle Control Device for Adaptive Cruise Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems face challenges in adapting to technological advancements and expanding to various vehicle types while ensuring safety, as they struggle to integrate new technologies and camera configurations effectively.
Innovation Solution
A vehicle control device with a modular architecture, featuring independent signal processing, recognition processing, and judgment IC units, allowing for easy updates and adaptations, including the use of deep learning for enhanced recognition precision and a backup rule-based safety system for reliability and safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If autonomous driving systems use deep learning for route calculation and environment recognition, then recognition precision is improved, but system complexity and difficulty of adapting to technological changes increases
Solution Approach 1:
The system is divided into multiple independent ECUs: surrounding environment acquisition ECUs (cameras, radar), route calculation ECU, and integrated control ECU. Each ECU performs a specific function and can be independently updated or replaced, reducing overall system complexity while maintaining high recognition precision through specialized deep learning processing in the route calculation ECU.
Solution Approach 2:
The system enables dynamic switching between different control modes (autonomous driving control and manual control) and different route calculation methods (deep learning-based and rule-based). This flexibility allows the system to adapt to technological changes by updating individual ECUs without redesigning the entire system, thus managing complexity while maintaining precision.
2Measurement precision
If the system integrates multiple surrounding environment acquisition devices and processing units, then recognition precision is improved, but ease of manufacture and expansion to vehicle types decreases
Solution Approach 1:
The integrated control ECU is designed to receive inputs from multiple types of surrounding environment acquisition devices (cameras, radar, LIDAR) and perform multiple functions including route calculation, collision avoidance control, and emergency braking. This universal design allows the same ECU to be applied across different vehicle types and configurations, easing manufacturing and expansion while maintaining high recognition precision through multi-sensor fusion.
3Reliability
If the system uses rule-based control for safety backup, then functional safety is improved, but recognition precision and adaptability to technological changes decreases
Solution Approach 1:
The system implements a backup control mechanism where the integrated control ECU can switch to rule-based control methods when the deep learning-based route calculation ECU fails or is unavailable. This beforehand cushioning ensures functional safety by having a reliable fallback system, while the primary deep learning system maintains high recognition precision during normal operation. The dual-mode design allows the system to prioritize precision when possible while guaranteeing safety through rule-based backup.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
A vehicle control device includes: a signal processing integrated circuit (IC) unit 10 for performing image processing on an output from a camera mounted in a vehicle and outputting image data obtained through the image processing; a recognition processing IC unit 20 provided as another unit different from the signal processing IC unit 10, for performing recognition processing for recognizing an external environment of the vehicle based on the image data received from the signal processing IC unit 10 and outputting external environment data obtained through the recognition processing; and a judgment IC unit 30 provided as another unit different from the signal processing IC unit 10 and the recognition processing IC unit 20, for performing judgment processing for cruise control of the vehicle based on the external environment data received from the recognition processing IC unit 20 and outputting a cruise control signal based on the judgment processing result.