Fault-Tolerant Computer System for Autonomous Vehicle Maneuvering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous vehicle computer systems lack cost-effective fault-tolerance, making them unreliable in the presence of failures, which is critical for ensuring safety and functionality in autonomous maneuvering applications.
Innovation Solution
A fault-tolerant high-performance computer system is designed with two fault-containment regions, each containing both safety and performance processors, allowing for information exchange via a communication subsystem, enabling the system to downgrade performance tasks and maintain functionality even when one region fails, ensuring safe vehicle operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a fault-tolerant architecture with multiple processors and regions is implemented, then system reliability is improved, but device complexity increases
Solution Approach 1:
The system is divided into two independent fault-containment regions (first and second regions), each containing safety and performance processors. This segmentation isolates failures to specific regions, preventing system-wide failures while maintaining manageable complexity through modular design.
Solution Approach 2:
The patent introduces a spatial dimension to fault tolerance by organizing processors into distinct physical or logical regions. This dimensional organization allows independent failure containment and recovery, improving reliability without proportionally increasing overall system complexity.
2Productivity
If high-performance processors are used for autonomous maneuvering tasks, then productivity is improved, but reliability deteriorates due to lack of fault tolerance
Solution Approach 1:
Different processors within the same region are assigned different functional qualities: safety processors execute safety-critical tasks with high reliability, while performance processors handle computationally intensive autonomous maneuvering tasks. This local differentiation allows each processor to be optimized for its specific function while contributing to overall system reliability through diversity.
Solution Approach 2:
The system preemptively establishes redundant safety processors that can take over performance tasks if performance processors fail. This prior cushioning ensures that high-performance computing capability is maintained even when primary performance processors fail, bridging the reliability-performance gap.
3Reliability
If redundant safety processors are added to handle failures, then reliability is improved, but cost increases
Solution Approach 1:
Safety processors are designed with multi-functionality, capable of executing both safety-critical tasks and performance tasks related to autonomous maneuvering. This universality allows the system to maintain reliability through redundancy while reducing the total number of dedicated processors needed, as safety processors can dynamically assume performance roles during failures.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The invention relates to a computer system.