Vehicle Risk Map Comparison for ECU Abnormality Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of automatic driving control makes it difficult to determine whether the control of electronic control units (ECUs) for vehicle automatic driving is abnormal, as simple changes in operation amounts may not indicate abnormality, and distinguishing between ECU-related, processing-related, and data input-related abnormalities is challenging.
Innovation Solution
An abnormality diagnosis system that includes a risk information generation unit and a diagnosis unit, using sensor information to generate and compare risk information across multiple ECUs, detecting abnormalities in risk maps to determine if risks are overlooked, and triggering resets when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If simple change detection of operation amounts is used for ECU failure diagnosis, then the diagnosis method is simple, but it cannot accurately determine abnormalities in complex automatic driving control
Solution Approach 1:
The patent introduces risk maps as an intermediary diagnostic tool. Instead of directly monitoring complex operation amount changes, the system generates risk maps that represent the ECU's understanding of the driving environment and compares these maps between main and sub ECUs. This intermediary representation simplifies the diagnosis process while improving accuracy, as risk maps provide a standardized format for comparing ECU perceptions of the same environment.
2Reliability
If multiple risk information generation units are used to improve diagnosis accuracy, then abnormality detection capability is improved, but system complexity increases
Solution Approach 1:
The patent merges the diagnostic functionality into the existing dual-ECU architecture by having both main and sub ECUs generate and compare risk maps. This approach leverages the redundant ECUs already present in the system for safety-critical applications, transforming them into diagnostic resources without adding separate dedicated diagnostic hardware. The risk map comparison mechanism utilizes the existing parallel processing capability of dual ECUs.
3Measurement precision
If detailed risk map comparison is performed to accurately diagnose abnormalities, then diagnosis precision is improved, but processing time increases
Solution Approach 1:
The patent extracts only the critical comparison elements from full risk maps for diagnostic purposes. Instead of comparing entire risk maps in detail, the system identifies and compares specific key parameters and risk regions that are most indicative of ECU abnormalities. This selective extraction approach maintains diagnostic precision by focusing on the most informative aspects of risk maps while significantly reducing processing time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An abnormality related to control of automatic driving of a vehicle can be easily and appropriately diagnosed. In a vehicle control system 1000, a plurality of risk information generation units (CPUs 10A and 10B that execute risk map creation program 112A and 112B) that generates risk map which is used for automatic driving control of a vehicle when the vehicle moves based on sensor information related to an object around the vehicle is provided. Diagnosis units (CPUs 10A and 10B that execute diagnosis (risk map comparison) programs 113A and 113B)) that diagnose whether or not an abnormality occurs in the generated risk information based on a plurality pieces of risk information generated by the plurality of risk information generation units is provided.