Streamlined Process for Cloud Based Diagnostics Using Amazon Web Services

2021-01-0159

04/06/2021

Event
SAE WCX Digital Summit
Authors Abstract
Content
In the age of 5G, the cloud constitutes a massive computational resource. Such capability is greatly underutilized, especially for the purpose of vehicle diagnostics and prognostics. Diagnostics and prognostics run mostly in the limited and cost sensitive electronic module of the vehicle. Utilizing vehicle connectivity, along with the massive capability of the cloud would allow the deployment of smarter algorithms that provide improved vehicle performance and operation management. In this paper, a streamlined process to develop and deploy off-board diagnostics is presented. The process included developing multiphysics digital twins and running the diagnostics off-board. It was demonstrated on a fleet of virtual Hybrid Electric Vehicles (HEV). The Digital Twin replica was created using Simulink® and Simscape®. The microcontroller used to demonstrate the diagnostic is a Raspberry Pi hardware running in real time. Furthermore, a Digital Twin of the same HEV model was developed and deployed into the Amazon Cloud Services (AWS). A total of 1000 vehicles were simulated to prove the effectiveness of the proposed process. The diagnostic data of these 1000 vehicles were sent in real time to AWS Digital Twin. The Digital Twin ran a diagnostic algorithm based on vehicle speed, motor, generator, engine speeds and battery SOC data received from the hardware. Failures were randomly introduced in the 1000 vehicles. The Digital Twin Diagnostics effectively detected the introduced failures and notified the drivers via Email and Text messages.
Meta TagsDetails
DOI
https://doi.org/10.4271/2021-01-0159
Pages
5
Citation
Khaled, N., "Streamlined Process for Cloud Based Diagnostics Using Amazon Web Services," SAE Technical Paper 2021-01-0159, 2021, https://doi.org/10.4271/2021-01-0159.
Additional Details
Publisher
Published
Apr 6, 2021
Product Code
2021-01-0159
Content Type
Technical Paper
Language
English