Fleet Analysis Backend

The E2E demo includes a Jakarta EE 11 backend service that provides fleet-level analytics on top of the telemetry stored by the Fleet Management Blueprint.

Overview

The fleet analysis backend runs alongside the Fleet Management stack and connects to the same InfluxDB instance. It provides REST APIs for:

• Computing summary statistics from vehicle telemetry snapshots
• Ingesting telemetry data into InfluxDB
• Reading periodically refreshed fleet statistics

Architecture

API Endpoints

POST /api/analysis/summary

Accepts a JSON array of vehicle snapshots and returns computed summary statistics.

Request body:

[
  { "vehicleId": "truck-01", "speedKph": 85.2, "batterySoc": 0.78, "brakeActive": false, "updatedAt": "2024-06-10T10:15:30Z" },
  { "vehicleId": "truck-02", "speedKph": 60.0, "batterySoc": 0.52, "brakeActive": true, "updatedAt": "2024-06-10T10:15:32Z" }
]

Response:

{
  "vehicleCount": 2,
  "averageSpeedKph": 72.6,
  "minBatterySoc": 0.52,
  "maxBatterySoc": 0.78,
  "brakingVehicles": 1
}

POST /api/telemetry/ingest

Writes header and/or snapshot measurements into InfluxDB. Used for ingesting telemetry data from external sources.

Request body:

{
  "vin": "truck-001",
  "trigger": "periodic",
  "createdDateTime": 1737940602000,
  "header": {
    "hrTotalVehicleDistance": 12345.6,
    "grossCombinationVehicleWeight": 18100.2,
    "totalEngineHours": 82.5,
    "engineTotalFuelUsed": 221.9,
    "driver1Id": "driver-01",
    "driver1IdCardIssuer": "fleet"
  },
  "snapshot": {
    "latitude": 37.7749,
    "longitude": -122.4194,
    "speed": 54.2,
    "positionDateTime": 1737940602,
    "wheelBasedSpeed": 53.7,
    "fuelLevel1": 0.42,
    "parkingBrakeSwitch": false
  }
}

GET /api/analysis/stats

Returns the latest fleet statistics snapshot from InfluxDB, periodically refreshed (default: every 30 seconds).

Response:

{
  "vehicleCount": 5,
  "headerCount": 120,
  "snapshotCount": 480,
  "totalCount": 600,
  "generatedAt": "2025-01-15T10:30:00Z"
}

Configuration

The service is configured via environment variables:

Variable	Default	Description
INFLUXDB_URI	http://influxdb:8086	InfluxDB connection URI
INFLUXDB_ORG	sdv	InfluxDB organization
INFLUXDB_BUCKET	demo	InfluxDB bucket
INFLUXDB_TOKEN	—	InfluxDB authentication token
INFLUXDB_TOKEN_FILE	—	Path to token file (alternative to INFLUXDB_TOKEN)
INFLUXDB_STATS_INTERVAL_SECONDS	30	Interval for refreshing fleet statistics
INFLUXDB_STATS_INITIAL_DELAY_SECONDS	10	Initial delay before the first stats run

Build and Run

Build with Maven

cd devices/backend-fleet-analysis-java
mvn package

Run Standalone (Payara Micro)

Download Payara Micro 6, then deploy:

java -jar payara-micro.jar \
  --deploy target/fleet-analysis-backend.war \
  --contextRoot /fleet-analysis

The API will be available at http://localhost:8080/fleet-analysis/api.

Build Docker Image

docker build -t fleet-analysis-backend:local devices/backend-fleet-analysis-java

Run with Docker Compose (Recommended)

The service is included in the Fleet Management Docker Compose stack. Start from the repository root:

docker compose \
  -f external/fleet-management/fms-blueprint-compose.yaml \
  -f external/fleet-management/fms-blueprint-compose-zenoh.yaml \
  up --detach

The analytics service will be available at http://<host>:8082/fleet-analysis/api.

Integration with Fleet Management

The fleet analysis backend integrates with the upstream Fleet Management Blueprint:

• Data source: Reads from the same InfluxDB instance where the FMS Consumer writes vehicle telemetry
• Network: Joins the fms-backend Docker network
• Token sharing: Uses the same InfluxDB token created by the Influx init job (mounted via INFLUXDB_TOKEN_FILE)
• RFID driver identification: The Vehicle.Driver.Identifier.Subject signal from the RFID door ECU is forwarded by the FMS Forwarder as the driver1Id field in the header measurement, visible in the Grafana "Driver Identifier (RFID)" panel