Commercial Vehicles Sensor Data

This post describes how to use Commercial Vehicles Sensor Data to Commercial Vehicles Status Classification through AI learning.

Table of Contents

1. Data Introduction
2. Data Visualization with Machbase Neo
3. Table Creation and Data Upload in Machbase Neo
4. Experimental Methodology
5. Experiment Code
6. Experimental Results

1. Data Introduction

• DataHub Serial Number: 2025-5.
• Data Name: Commercial Vehicles Sensor Data.
• Data Collection Methods: Collected through sensors installed on each dump truck, including accelerometers, speed sensors, and gyroscopes.
• Data Source: Link
• Raw data size and format: 156MB, CSV.
• Number of tags: 8.

• Data Time Range: 2025-01-01 00:00:00 to 2025-01-01 04:44:00.
• Number of data records collected: 13,599,864.
• CSV data URL: https://data.yotahub.com/2025-5/datahub-2025-5-commercial-vehicles.csv.gz
• Data Migration: Commercial Vehicles Sensor Data Migration

2. Data Visualization with Machbase Neo

• Data visualization is possible through the Tag Analyzer in Machbase Neo.
• Select desired tag names and visualize them in various types of graphs.

• Below, access the 2025-5 DataHub in real-time, select the desired tag names from the data of 8 tags, visualize them, and preview the data patterns.

DataHub Viewer

3. Table Creation and Data Upload in Machbase Neo

• In the DataHub directory, use setup.wrk located in the Commercial Vehicles Sensor Dataset folder to create tables and load data, as illustrated in the image below.

1) Table Creation

• The table is created immediately upon pressing the "Run" button in the menu.
• If the commercial_vehicles table exists, execute the first line and then the second. If it does not exist, start from the second line.

2) Data Upload

• Loading tables in two different ways.

Method 1) Table loading method using TQL in Machbase Neo (since machbase-neo v8.0.29-rc1

• Pros

  • Machbase Neo loads as soon as you hit the launch button.

• Cons

  • Slower table loading speed compared to other method.

Method 2) Loading tables using commands

• Pros

  • Fast table loading speed.

• Cons

  • The table loading process is cumbersome.
  • Run cmd window - Change machbase-neo path - Enter command in cmd window.

• If run the below script from the command shell, the data will be entered at high speed into the commercial_vehicles table.

curl http://data.yotahub.com/2025-5/datahub-2025-5-commercial-vehicles.csv.gz | machbase-neo shell import --input - --compress gzip --header --method append --timeformat ns commercial_vehicles

• If specify a separate username and password, use the --user and --password options (if not sys/manager) and add the options as shown below.

curl http://data.yotahub.com/2025-5/datahub-2025-5-commercial-vehicles.csv.gz | machbase-neo shell import --input - --compress gzip --header --method append --timeformat ns commercial_vehicles --user USERNAME --password PASSWORD

4. Experimental Methodology

• Model Objective: Commercial Vehicles Sensor Classification.
• Tags Used: gFx, gFy, gFz, label, speed, wx, wy, wz.
• Model Configuration: Transformer.

• Learning Method: supervised Learning.
  • Train: Model Training.
  • Test: Model Performance Evaluation Based on Commercial Vehicles Sensor Classification.
• Model Optimizer: Adam.
• Model Loss Function: CrossEntropyLoss.
• Model Performance Metric: F1 Score.
• Data Loading Method
  • Loading the Entire Dataset.
  • Loading the Fetch Dataset.
• Data Preprocessing
  • MinMax Scaling.

5. Experiment Code

• Composed of three methods.
• Data Information: Outputs general information about the data.
• Visual Information: correlation heatmap, plot, Decomposition about the data.
• Statistical Test: ADF Test, KPSS Test, PP Test, ljung box Test, Arch Test, VIF Test about the data.

• The entire code can be run through 5.Commercial_Vehicles_Sensor_EDA.

Commercial Vehicles Sensor Classification

• Below is the code for each of the two ways to get data from the database.
• If all the data can be loaded and trained at once without causing memory errors, then method 1 is the fastest and simplest.
• If the data is too large, causing memory errors, then the batch loading method proposed in method 2 is the most efficient.

Method 1) Loading the Entire Dataset

• The code below is implemented in a way that loads all the data needed for training from the database all at once.
• It is exactly the same as loading all CSV files (The only difference is that the data is loaded from Machbase Neo).
• Pros
  • Can use the same code that was previously utilizing CSVs (Only the loading process is different).
• Cons
  • Unable to train if trainable data size exceeds memory size.

• The entire code can be run through 5.Commercial_Vehicles_Sensor_Full.

Method 2) Loading the Fetch Dataset

• Method for loading data from the Machbase Neo for a buffer size.
• Pros
  • It is possible to train the model regardless of the data size, no matter how large it is.
• Cons
  • It takes longer to train compared to method 1.

• The entire code can be run through 5.Commercial_Vehicles_Sensor_Buffered_Fetch.

6. Experimental Results

Method 1) Loading the Entire Dataset Result

Method 2) Loading the Fetch Dataset Result

• The F1 scores of 0.58 and 0.55 indicate relatively low performance.
• There could be several reasons for the low performance, but they can largely be divided into two main factors.
  • The first reason is that the data was collected without distinguishing between individual trucks, which undermines consistency.
  • The second reason is that the data values do not differ significantly depending on the truck's operational state.
• Try improving the performance by applying various data preprocessing techniques!

※ Various datasets and tutorial codes can be found in the GitHub repository below.

datahub/dataset at main · machbase/datahub

All Industrial IoT DataHub with data visualization and AI source - machbase/datahub

GitHubmachbase

Back to Top