Vehicle Surround View Failover Using Dual Hardware Accelerators
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle surround view systems are vulnerable to failures due to camera malfunctions or software issues in high-level operating systems, leading to operational hindrances and safety concerns.
Innovation Solution
Implement a fail-safe mechanism that switches to a second hardware accelerator, such as an image signal processor or digital signal processor, to render surround view images when the primary hardware accelerator, typically a graphics processing unit, fails or crashes, and compensates for camera failures by adjusting blend lines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple cameras are used to provide surround views, then the safety and maneuverability assistance is improved, but the system complexity and vulnerability to camera failures increase
Solution Approach 1:
The system divides the surround view rendering function into multiple independent hardware accelerators (first hardware accelerator and second hardware accelerator). Each accelerator can independently process camera feeds and generate surround views, allowing the system to segment the critical rendering function across multiple hardware components to improve reliability while managing complexity.
Solution Approach 2:
The system pre-configures multiple hardware accelerators and establishes failover protocols before any failure occurs. The second hardware accelerator is prepared in advance as a backup, and the system has predetermined procedures for switching between accelerators, eliminating the need for complex real-time decision-making during failures.
2Ease of operation
If software for stitching and rendering runs in the context of a high level operating system, then the ease of operation and development is improved, but the vulnerability to software issues and system failures increases
Solution Approach 1:
The hardware accelerators serve as intermediaries between the camera input and the display output, performing the critical stitching and rendering functions in dedicated hardware rather than relying solely on high-level operating system software. This intermediary hardware layer provides a more reliable, deterministic execution environment for safety-critical surround view generation.
Solution Approach 2:
The system replaces software-based stitching and rendering (which runs in the high-level operating system context) with hardware-based accelerators that perform these functions in dedicated hardware circuits. This substitution of hardware for software eliminates many software-related reliability issues while maintaining the ease of operation benefits through hardware abstraction.
3Device complexity
If a single hardware accelerator is used for rendering surround views, then the device complexity is reduced, but the vulnerability to hardware failures and loss of function increases
Solution Approach 1:
The surround view rendering function is segmented across multiple hardware accelerators rather than concentrated in a single unit. The first hardware accelerator handles primary rendering while the second hardware accelerator serves as a backup, dividing the critical function across independent hardware components to eliminate single points of failure.
Solution Approach 2:
The system implements prior cushioning by having the second hardware accelerator ready as a backup before any failure occurs. This preemptive redundancy ensures that if the first hardware accelerator fails, the system can immediately switch to the second accelerator without loss of function, cushioning against hardware failures.
Data Source
AI summary
A technique including capturing, by one or more cameras of a set of cameras disposed about a vehicle, one or more images, wherein a surround view system of the vehicle is configured to render a surround view image using a first hardware accelerator based on the one or more images, determining that a first hardware accelerator is unavailable, and rendering the surround view image using a second hardware accelerator based on the captured one or more images.