Predictive Maintenance For Connected Vehicles
- Analytics & Modeling - Machine Learning
- Analytics & Modeling - Predictive Analytics
- Analytics & Modeling - Real Time Analytics
- Application Infrastructure & Middleware - Data Visualization
- Automotive
- Maintenance
- Predictive Maintenance
By 2025, Transport for London will have to meet strict emission-control regulations. This means buying and operating new fleets of hybrid or fully electric, zero-emission buses. As a consequence, many Original Equipment Manufacturers (OEMs) and operators will have to develop new technologies to help them get-to-market fast enough to meet demand.
Accelerating time-to-market and compliance controls
Vantage Power worked with Luxoft (an AWS Partner Network Advanced Consulting partner) to create VPVision - an innovative telemetry platform. VPVision brings the AWS cloud platform to each connected vehicle, providing Vantage Power customers with an overview of each vehicle’s powertrain components, including batteries, control systems, engines, motors, and electric generators.
VPVision is built around an IoT architecture (AWS IoT Core, AWS Greengrass, and AWS IoT Analytics), then combined with Amazon Simple Storage Services (Amazon S3) and AWS Lambda. This optimizes the processing of hundreds of thousands of data points per minute. In turn, it provides OEMs and operators with valuable insights, allowing them to create predictive maintenance models and implement real-time preventative action.
Here are just six of VPVision’s benefits:
1. The system monitors everything from vehicle speed to engine health and battery-pack-level diagnostics. It enables operators and OEMs to receive insight and reporting for over 6,000 data points from each vehicle.
2. VPVision collects, processes, stores, and presents real-time vehicle data, automatically, via the cloud.
3. It enables real-time geolocation visualization and two-way communication between fleet managers and drivers.
4. When routine maintenance is required, real-time alerts keep downtime to a minimum.
5. It leverages IoT components from AWS such as IoT Analytics, Greengrass ML, Sagemaker, and many more.
6. The team developed a machine learning model that used AWS IoT Sagemaker Notebooks to analyze ‘idle time’ and ‘vehicle location’ data, automatically.