AI-based Signal Integrity Monitoring for Integrated Vehicle Health Management
2022-01-0256
03/29/2022
- Event
- Content
- The increasing complexity of vehicle electronics and software is bringing an abundance of vehicle health related challenges, including quality issues and increasing costs of warranty claims, recalls, maintenance, and downtime. This negatively impacts both OEM and fleet profitability, user experience, and end customer costs. In order to reduce OEM costs and the total cost of ownership for consumers and fleets, new methods are needed to detect, predict, and diagnose vehicle health issues. Existing vehicle health management solutions rely on diagnostics trouble codes (DTC) and limited amounts of telematics data. These solutions can detect known failure modes using hard-coded signal behavior validation rules that are frequently based on thresholds. They also provide alerts based on pre-defined error codes. However, they are unable to detect and diagnose unforeseen failure modes that do not have hard-coded rules, nor can they prognose future vehicle health issues. By using deep learning technology to analyze the thousands of parameters that are available onboard the vehicle in real-time, AI-based Integrated Vehicle Health Management solutions can not only predict vehicle health issues, but also estimate the remaining useful life of components, detect performance degradation, and determine the root cause of problems. This paper will describe how deep learning based Signal Integrity Monitoring can be used to detect complex anomalous patterns and provide signal fault isolation. It will describe how unsupervised deep learning technology can simplify and automate the data collection and AI training processes. By integrating AI-based Signal Integrity Monitoring into embedded vehicle electronics, automakers and fleet owners can provide: • Early indicators of known faults before a DTC is triggered • Detection of system-level faults that do not have DTCs • Detection of complex fault patterns that are difficult to define explicitly • Root cause analysis of DTCs using signal fault isolation • Reduced No-Trouble-Found cases by correlating DTCs to anomalies • Reduced repair time with enhanced diagnostics report
- Citation
- Apartsin, S., Stein, H., Reiter, G., and Williams, K., "AI-based Signal Integrity Monitoring for Integrated Vehicle Health Management," SAE Technical Paper 2022-01-0256, 2022, .