Redundant Data Processing Networks for Real-Time Safety Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance and automated driving systems face challenges in achieving high computing power while meeting stringent safety requirements due to the limitations of conventional microcontrollers and the lack of internal optimizers in off-the-shelf processors, leading to inefficiencies and increased latency in data processing.
Innovation Solution
A data processing network utilizing a software lockstep approach with two separate data processing modules and a comparator module on trusted hardware, allowing for flexible execution order and efficient utilization of high-performance processors while ensuring ASIL-D compliance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional microcontrollers are used to meet safety requirements, then reliability is improved, but computing power deteriorates
Solution Approach 1:
The system is divided into multiple independent data processing modules (first module, second module, third module) that each perform specific data processing tasks. This segmentation allows the use of multiple lower-power microcontrollers instead of a single high-power processor, maintaining safety through redundancy while distributing computational load across multiple units with limited individual computing power.
Solution Approach 2:
A comparator module acts as an intermediary between the first and second data processing modules. It receives control parameters from both modules, compares them for consistency, and generates synchronized control parameters. This intermediary mechanism enables coordination and validation between parallel processing units, ensuring safety compliance while allowing the system to leverage multiple processing paths.
2Productivity
If high-end processors are used to increase computing power, then productivity is improved, but device complexity deteriorates
Solution Approach 1:
Rather than using a single complex high-end processor, the system segments computing tasks across multiple simpler microcontrollers. Each module handles specific data processing functions, and the overall system achieves high productivity through parallel processing and coordinated operation of these simpler units.
Solution Approach 2:
Multiple data processing modules perform similar data processing functions, providing functional redundancy and universality. The comparator module serves multiple purposes: receiving parameters from different modules, comparing them for validation, and generating synchronized control signals. This multi-functionality approach increases productivity while managing complexity through standardized, reusable components.
3Reliability
If redundant data processing modules are deployed to ensure safety, then reliability is improved, but loss of time deteriorates
Solution Approach 1:
The comparator module continuously monitors and compares control parameters from parallel data processing modules in real-time, performing validation actions preliminarily before final output is generated. This ongoing comparison ensures safety validation occurs during processing rather than as a separate post-processing step, minimizing additional latency while maintaining reliability.
Solution Approach 2:
The system maintains continuous operation of multiple data processing modules in parallel, with the comparator module continuously comparing parameters and generating synchronized outputs. This continuous parallel processing ensures that safety validation does not interrupt the data flow but occurs concurrently, reducing time loss while ensuring reliability through ongoing validation.
Data Source
AI summary
A data processing network is for performing a plurality of successive data processing steps in a redundant and validated manner. The data processing steps are each used to generate output data from input data. At least some output data from a first data processing step are at the same time input data of a further data processing step. At least a first data processing module and a second data processing module are provided for performing each data processing step. The data processing network includes a comparator module. The first data processing module and the second data processing module are configured to perform the data processing steps, optionally in a first working mode with parallel operation, or in a second working mode with an upstream data processing module and a downstream data processing module.