The Plan to execute and document learnings

The pilot project. Got some personal time today to take another look at the overall approach. Also had a new tracker fitted and the old alarm CPU removed. The guys that had maintained the car before me, did not do a good job. It was an electrical fire waiting to happen. Now the old system has been removed, the wiring is back to OEM standards and a new track installed, I have some time to write about the plan.

The high-level plan.

1. Purchase Hardware Components

   1. OBD-II Adapter with Internet Connectivity:

   2. I plan to use an OBD-II adapter that supports Wi-Fi and Bluetooth to gather telemetry data from the car. Over the summer I purchased a standard Bluetooth OBD adaptor to clear some fault codes.

   3. The OBDLink MX+ adapter will connect to the OBD-II port and stream data such as vehicle speed, location, engine performance, and more.

2. Select the Communication Stack and Telemetry Streaming

Amazon IoT Core:

• I will set up Amazon IoT Core as the central hub to receive telemetry data from the OBD-II device. This will allow me to securely connect the car’s telemetry to the cloud over HTTP, MQTT, or WebSockets.

• I will configure Amazon IoT Core to capture real-time data from the OBD-II adapter.

• OBD-II Adapter Configuration:

• I will configure the OBD-II device to send telemetry data via REST APIs or MQTT to the Amazon IoT Core instance. I may need to install specific libraries or use the SDKs provided by the device manufacturer for data transmission.

  
  

3. Choose Data Ingestion and Storage

• I am exploring Amazon Kinesis or Amazon IoT Analytics:

  • I plan to use Amazon Kinesis or IoT Analytics to process and filter the telemetry data. Kinesis would help manage the data flow and distribute it for storage and further processing.

• Amazon DynamoDB (for Storage):

  • The goal is to create an Amazon DynamoDB database or Amazon S3 bucket to store the raw telemetry data coming in from the OBD-II adapter

4. Integration with SolarWinds Observability

   SolarWinds API for Data Ingestion:

   • I plan to work with engineering to help set up up a RESTful service or webhook that communicates between Amazon Kinesis or DynamoDB and SolarWinds Observability’s data ingestion layer. The SolarWinds API would allow the data to be pulled from Amazon IoT Core or DynamoDB into the SolarWinds observability dashboard

     
     

   • Configure a SolarWinds Custom Dashboard:

     • Then I will configure a custom dashboard in SolarWinds Observability to visualize telemetry data (vehicle location, speed, engine parameters). SolarWinds supports custom data ingestion via API, so can use this to display my Subaru’s telemetry as metrics.

5. Monitoring and Reporting

   1. SolarWinds Alerts and Analytics:

   2. I plan to try to use the SolarWinds alerting system to create rules for specific telemetry triggers (e.g., speeding alerts, location-based triggers, or diagnostic trouble codes from the car).

   3. Then set up performance charts and historical data tracking for trends like speed variations, engine efficiency, and geographical movement over time.

   
   

6. Explore Security and Authentication options

   1. Device Certificates and Authentication:

   2. Use Amazon IoT's built-in security features like X.509 certificates for authenticating the OBD-II device.

   3. Set up role-based access control (RBAC) and encryption (TLS/SSL) to ensure that telemetry data is securely transmitted over the internet.

   
   

7. Testing and Calibration

• Once the hardware and software integration is complete:

  • Test the system by driving the car and monitoring if the telemetry data appears in SolarWinds observability.

  • Fine-tune the reporting intervals and alerts based on real-world performance.

  • Trial using the car as a demo

Key Components to budget for

1. OBD-II Device: OBDLink MX+ or CarLoop 4G/LTE

2. Amazon IoT Core: For secure device communication

3. Amazon Kinesis/IoT Analytics: For data processing

4. Amazon DynamoDB/S3: For data storage

5. SolarWinds Observability: For telemetry monitoring and reporting

6. APIs/Webhooks: For data ingestion between AWS and SolarWinds

By using an OBD-II adapter to capture telemetry, Amazon IoT Core for cloud communication, and SolarWinds for observability and analytics, I should have a robust setup that provides real-time monitoring of my car’s telemetry by the end of the year.

UPDATE- OBDLink MX+ is not wifi capable and instead of an expensive OBD device I have purchased a wifi enabled device at a lower cost to connect the raw data to the computer.