IoT Asset Tracking

Trucking & Fleet Management

Transcript

1. IoT Asset Management: TRUCKING & FLEET Rohan Hubli
2. REAL-TIME PROCESSING REQUIREMENTS Data Processing & Store Geospatial, Temporal Data Sensor Monitoring Event Filtering T1...........T2…….T3…….T4…….T5……. GPS Data ≈ 5KB/s Sensor Data ≈ 2KB/s OBDII Data ≈ 1KB/s Fleet Size: 300-600 ROHAN HUBLI 2018
3. IOT ASSET TRACKING: ARCHITECTURE ROHAN HUBLI 2018 Electronic Logger Module Electronic Logger Module Paho MQTT Leshan LWM2M KAFKA Connector MQTT KAFKA CLUSTER STORM CLUSTER HBase HDFS SPARK CLUSTER BI Dashboard W E B S E R V I C E Published Events EVENTS API
4. FLEET TRACKING: ELECTRONIC LOGGER MODULE 2018 ROHAN HUBLI  Sensor Data Stream: MQTT  Connectivity: GSM  Location: GPS  Device Management: WebUSB, BLE  Vehicle Interface: CAN  Development Environment: MicroEJ Battery Microcontroller Memory PMM NFC User Authentication Network Interface LPWAN Ethernet Watch Dog DATA CHANNEL On-Board Analytics/Monitoring CAN Display ALARM & STATUS Device Management Interface WebUSB BLE DATA CHANNEL Sensor Interface Sensor #1 Sensor #N GPS Asset Tracking: Electronic Logger Module MQTT ROM LWM2M
5. REAL-TIME MONITORING: KAFKA + STORM PROCESSING spout Preprocess Filter Bolt-1 Bolt-2 Bolt-3 Bolt-4 Bolt-5 #N Event Handler Geo-Spatial Speed/Route Sensors Bolt-6 Aggregate KAFKA HBase ROHAN HUBLI 2018
6. ISOLATING WORKLOADS: REAL-TIME VS BATCH Predictive Analytics BI Dashboard & Reporting Web Services ROHAN HUBLI 2018 HBase Primary Replica HDFS HDFS
7. THANK YOU