Dual-Controller Autonomous Driving Failover for ASIL D Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing single-architecture autonomous driving systems cannot achieve ASIL D-level functional safety and reliability due to the lack of backup redundancy, leading to potential failures when any controller or actuation mechanism fails at a single point.
Innovation Solution
The implementation of a dual-control system architecture comprising a main control system and a backup control system, where the main control system monitors its status in real-time and sends failure notifications to the backup control system to execute backup control instructions, ensuring continuous operation of autonomous driving vehicles even if a module fails.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single ECU and architecture is used in autonomous driving system, then device complexity is reduced, but reliability cannot achieve ASIL D-level functional safety
Solution Approach 1:
The autonomous driving control system is segmented into multiple independent ECUs (first ECU, second ECU, third ECU) with distinct functions: perception decision-making, motion planning control, and actuation. This segmentation allows failure isolation and enables ASIL D-level safety through functional distribution, resolving the contradiction between reliability and complexity by organizing complexity into manageable, safety-critical segments.
Solution Approach 2:
The system performs preliminary safety checks and monitoring before failures occur. The first ECU continuously monitors the operating state of actuators and communicates with other ECUs to detect potential failures early. This preliminary detection and communication mechanism ensures that safety protocols are activated before critical failures impact system reliability, achieving ASIL D-level functional safety.
2Reliability
If backup redundancy is added to achieve ASIL D-level safety, then reliability is improved, but device complexity increases
Solution Approach 1:
Instead of adding redundant backup systems throughout, the architecture segments safety-critical functions into separate ECUs with defined communication protocols. Each ECU is responsible for specific functions with built-in monitoring, reducing overall system complexity while maintaining ASIL D-level safety through functional segmentation and failure isolation.
Solution Approach 2:
The system implements continuous feedback mechanisms where ECUs monitor each other's operating states and communicate through defined protocols. The first ECU monitors actuator states and communicates with the motion planning control ECU, which in turn communicates with actuation ECUs. This feedback loop enables real-time safety verification without requiring complex redundant backup systems, resolving the contradiction between reliability improvement and complexity increase.
3Reliability
If controller monitors operating state in real-time and communicates with other controllers, then reliability is improved, but loss of time for communication increases
Solution Approach 1:
ECUs continuously monitor operating states and maintain ready-to-transmit communication channels before failures occur. Safety-critical parameters are pre-formatted and communication protocols are pre-established, enabling immediate response when failures are detected without time-consuming setup or protocol negotiation, thus minimizing communication delay while maintaining high reliability.
Solution Approach 2:
The system implements continuous feedback loops with defined communication protocols between ECUs. Monitoring data is continuously exchanged through optimized communication channels that prioritize safety-critical information. This real-time feedback mechanism ensures rapid failure detection and response while minimizing communication time through efficient data exchange protocols and continuous monitoring states.
Data Source
Figure 1~2
AI summary
An autonomous driving control system, comprising a main control system and a backup control system. The main control system comprises a main control module and main execution modules, and the backup control system comprises a backup control module and backup execution modules; the main control module is configured to monitor an operating status of the main control system in real time; the main control module is further configured to send, when detecting that a failure occurs in the main control system, a failure notification to the backup control module, and send a response termination control instruction to each of the main execution modules, the response termination control instruction being a control instruction for instructing each of the main execution modules not to respond to any control over a vehicle; and the backup control module is configured to control, after receiving the failure notification, the backup execution modules to start to execute a backup control instruction. The system can solve the problem that an autonomous driving control system cannot operate normally when any controller or actuation mechanism fails at a single point.