On this page

Scenario

Today’s vehicle has lots of sensors generating lots of different data which if captured could be used
for vehicle monitoring, predictive analytics, finding anomalies and taking corrective actions as needed
either locally, remotely or in self-healing manner

The various data points (sensors) are as follows;

time
car
coolant_temp
intake_air_temp
intake_air_flow_speed
battery_percentage
battery_voltage
current_draw
speed
engine_vibration_amplitude
throttle_pos
tire_pressure_1_1
tire_pressure_1_2
tire_pressure_2_1
tire_pressure_2_2
accelerometer_1_1_value
accelerometer_1_2_value
accelerometer_2_1_value
accelerometer_2_2_value
control_unit_firmware

In this demo, we will stream data for many cars into the system, each car sending one event per sec or more.
We will use BangDB agent to stream the data into BangDB, and use streaming analysis for these events with 1 day as sliding window length (continuous window).

Further we will use BangDB ML abstractions to train two models, one using existing file and other on data which have been streamed into the database. These models will help us find certain anomalies for further action

Was this article helpful to you? Yes No