ECU Test Sequencing Using Neural Network Signal Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of software in modern electronic control units (ECUs) in vehicles requires extensive and thorough testing to ensure proper operation, but existing methods are inefficient and resource-intensive, especially in advanced vehicles like autonomous driving systems.
Innovation Solution
A computer-implemented method using a neural network model to capture input-output relationships of ECUs connected to a CAN bus, pruning fully-connected neural networks based on neuron weight thresholds, ranking signal importance, and generating an optimized test case execution sequence to identify flaws in the system, thereby reducing the number of test cases and computational resources needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If extensive and thorough testing is performed on ECUs to ensure proper operation, then reliability of ECU functionality is improved, but testing time and computational resources are increased
Solution Approach 1:
The patent segments the testing process by dividing ECUs into different priority groups based on their importance to vehicle safety and operation. Critical ECUs (e.g., those controlling braking, steering, or airbags) are tested more thoroughly and earlier in the testing sequence, while less critical ECUs are tested with reduced coverage or later. This segmentation allows the system to maintain high reliability for safety-critical functions while reducing overall testing time and resources.
Solution Approach 2:
The patent implements preliminary action by performing static analysis, code review, and simulation testing on ECU software before actual hardware testing. This allows many defects to be detected and corrected in the development phase, reducing the need for extensive physical testing later. The testing sequence is also predetermined based on ECU priority rankings, with critical ECUs tested first while the vehicle is in controlled states, optimizing resource utilization.
2Reliability
If extensive and thorough testing is performed on ECUs to ensure proper operation, then reliability of ECU functionality is improved, but computational resources required are increased
Solution Approach 1:
The patent segments the computational testing workload by dividing test cases into different levels of depth and resource requirements. Critical ECUs receive comprehensive testing with multiple test cases and detailed analysis, while non-critical ECUs receive streamlined testing with fewer test cases. This segmentation allows the testing system to allocate computational resources efficiently, maintaining high reliability for safety-critical functions while reducing overall computational resource consumption.
Solution Approach 2:
The patent applies local quality by tailoring the depth and intensity of testing to the specific characteristics and importance of each ECU. Rather than applying uniform comprehensive testing to all ECUs, the system adjusts testing rigor locally based on each ECU's functional criticality, communication protocol, and integration complexity. This ensures that computational resources are concentrated on ECUs where they provide the most value for reliability improvement.
3Reliability
If traditional testing methods are used on ECUs, then comprehensive coverage can be achieved, but device complexity of the testing system is increased
Solution Approach 1:
The patent introduces an intermediary ECU priority ranking system that automatically analyzes ECU importance based on predefined criteria (safety criticality, functional importance, communication protocol) and generates an optimized testing sequence. This intermediary system simplifies the overall testing complexity by automatically determining test priorities and sequences, eliminating the need for manual configuration of complex testing schedules while maintaining comprehensive coverage of critical functions.
Solution Approach 2:
The patent changes the parameter of testing sequence organization from traditional arbitrary or chronological ordering to priority-based ordering based on ECU criticality. By changing how tests are sequenced and organized, the system achieves comprehensive coverage of critical functions with a simpler, more systematic approach. The testing framework maintains flexibility to adjust test parameters based on ECU priority levels without requiring complex custom configurations for each ECU.
Data Source
AI summary
A computer-implemented method for implementing electronic control unit (ECU) testing optimization includes capturing, within a neural network model, input-output relationships of a plurality of ECUs operatively coupled to a controller area network (CAN) bus within a CAN bus framework, including generating the neural network model by pruning a fully-connected neural network model based on comparisons of maximum values of neuron weights to a threshold, reducing signal connections of a plurality of collected input signals and a plurality of collected output signals based on connection weight importance, ranking importance of the plurality of collected input signals based on the neural network model, generating, based on the ranking, a test case execution sequence for testing a system including the plurality of ECUs to identify flaws in the system, and initiating the test case execution sequence for testing the system.