A Knowledge Graph-Based Data Integration Framework Applied to Battery Data Management

Abstract

Today, the automotive and transportation sector is undergoing a transformation process to meet the requirements of sustainable and efficient operations. This transformation mainly reveals itself by electric vehicles, hybrid electric vehicles, and electric vehicle sharing. One significant, and the most expensive, component in electric vehicles is the batteries, and the management of batteries is crucial. It is essential to perform constant monitoring of behavior changes for operational purposes and quickly adjust components and operations to these changes. Thus, to address these challenges, we propose a knowledge graph-based data integration framework for simplifying access and analysis of data accumulated through the operations of vehicles and related transportation systems. The proposed framework aims to enable the effortless analysis and navigation of integrated knowledge and the creation of additional data sets from this knowledge to use during the application of data analysis and machine learning. The knowledge graph serves as a significant component to simplify the extraction, enrichment, exploration, and generation of data in this framework. We have developed it according to the human-centered design, and various roles of the data science and machine learning life cycle can use it. Its main objective is to streamline the exploration and interaction with the integrated data to maximize human productivity. Finally, we present a battery use case to show the feasibility and benefits of the proposed framework. The use case illustrates the usage of the framework to extract knowledge from raw data, navigate and enrich it with additional knowledge, and generate data sets.

Object Detection in Driving Datasets Using a High-Performance Computing Platform: A Benchmark Study

“…With these in mind, we aim to investigate the benefits of using HPC with high-performance big data storage and processing for object detection use case that relies on a knowledge graph-based data integration framework [9]. We will utilize the already trained YOLOv4 [16], CenterNet HourGlass104 512x512, EfficientDet D0 512x512, and Faster R-CNN ResNet50 V1 640x640 object detection models for the use case.…”

“…Neo4j graph database [57] stores the knowledge graph. The main reason to store data as a knowledge graph is to combine different data that are stored in a set of separate data sources into a single, unified view [4], [9]. Step-by-step data import process implemented by nuScenes adapter notebooks Hence, the knowledge graph-based data integration provides a flexible, standardized, and agile platform to access and manipulate the datasets through a unified, single access layer.…”

“…Thus, several DS/ML algorithms and systems have been built to automate the different stages of the DS/ML life cycle [6], [8]. In this respect, we utilize a knowledge graph-based data integration framework [9] to ease the extraction, enrichment, generation, and exploration of the driving data sets. This framework aims to simplify the exploration and querying, data sets creation, and algorithm application over these data sets.…”

“…This framework is developed to address the data access obstacles in the automotive and transportation sector. Therefore, the focus of the framework is on the data readiness, preprocessing, cleaning, feature engineering, model building and training, and model refinement stages of the DS/ML life cycle [9]. After all, automating the data science and machine learning pipeline as much as possible, together with skilled data engineering, can simplify the development, testing, maintenance, and the whole process [1], [6], [11], [12].…”

“…We have selected nuScenes dataset provided by the Motional and Level 5 Perception Dataset provided by Lyft for the demonstration. We implement the previously proposed knowledge graph-based data integration [9] to combine these two different public datasets to achieve a successful data integration for the easy application of object detection methods over EVOLVE platform. We apply the object detection task using YOLOv4 and three additional models provided from TensorFlow Hub [25] over three different hardware setups.…”