AI Accelerator Image Processing Redundancy for Vehicle Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In autonomous and assisted driving scenarios, the use of artificial intelligence accelerators for image processing is prone to errors due to hardware failures, leading to reduced safety in vehicle driving.
Innovation Solution
A method and apparatus for image processing using an artificial intelligence accelerator, which involves determining algorithm model information for a to-be-processed image, selecting the appropriate AI accelerator, and executing algorithm model instructions to obtain a processing result. This approach includes redundancy by using multiple storage spaces and AI accelerators to mitigate errors in case of hardware failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single AI accelerator and storage space are used for image processing, then device complexity is reduced, but reliability deteriorates due to hardware failure causing continuous errors
Solution Approach 1:
The system segments the image processing function across multiple storage spaces and multiple AI accelerators. Each frame of image is processed by a different combination of storage spaces and accelerators, dividing the overall processing task into independent segments that can tolerate individual component failures without affecting overall system reliability.
Solution Approach 2:
The system changes the operational parameters by rotating which storage spaces and AI accelerators are used for processing different frames. This parameter rotation ensures that no single component is continuously stressed and that failures in one component do not propagate to subsequent frames, thereby improving reliability without requiring a fundamentally more complex system architecture.
2Reliability
If multiple storage spaces and AI accelerators are used for redundancy, then reliability improves, but device complexity increases
Solution Approach 1:
The system divides the redundancy resources into multiple storage spaces and multiple AI accelerators, assigning different segments of the processing workload to different components. This segmentation allows the system to leverage redundancy for reliability while managing complexity through structured resource allocation rather than monolithic design.
Solution Approach 2:
The system performs preliminary allocation of multiple storage spaces and AI accelerators in advance, establishing a rotation schedule before processing begins. This preliminary action prepares the redundant resources for reliable operation while avoiding the need for complex real-time decision-making during processing, thereby managing system complexity.
3Reliability
If the same storage space and AI accelerator are continuously used, then ease of operation is maintained, but reliability deteriorates due to hardware failure
Solution Approach 1:
The system implements periodic rotation of storage spaces and AI accelerators used for processing different frames. This periodic action maintains reliability by preventing continuous use of the same components, while the regular, predictable pattern of rotation keeps the operation simple and easy to manage through established cycles rather than complex adaptive logic.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
Disclosed are a method and an apparatus for image processing by using an artificial intelligence accelerator, and a chip. The method includes: determining algorithm model information corresponding to a to-be-processed image; determining, based on the algorithm model information, an artificial intelligence accelerator that needs to execute algorithm model instructions corresponding to the algorithm model information; and reading the algorithm model instructions from a first preset storage space storing the algorithm model instructions by using the artificial intelligence accelerator, and executing the algorithm model instructions, to obtain a processing result for the to-be-processed image. According to the embodiments of this disclosure, a probability of image processing errors caused by an accelerator hardware failure or a memory hardware failure can be effectively reduced, thereby greatly improving safety of a vehicle during driving.