Nowadays, cars are equipped with hundreds of sensors that support a variety of features ranging from basic functionalities to advanced driver assistance systems. The communication protocol of automotive signals is defined in DBC files, yet, signal descriptions are typically ambiguous and vary across manufacturers.   
In this work, we address the problem of extracting the semantics of DBC files, which is then managed in an Automotive Signal Knowledge Graph (ASKG). 
We developed a semi-automatic tool that automatically extracts signals from DBC files and compute candidate links to the ontology.  
These candidates can then be revised by experts who can also extend the ontology to accommodate new signal types in a pay-as-you-go manner. 
The knowledge provided by the experts is stored in the ASKG and exploited by the tool thereafter.
We conducted an evaluation of the tool based on a targeted experiment with automotive experts and report on first lessons learned from the usage of the tool in the context of the Bosch automotive data lake.
The results show that our solution can correctly populate the ASKG and that the expert effort is reduced over time.

Pay-as-you-go Population of an Automotive Signal Knowledge Graph

Making an informed and right decision poses huge challenges for drivers in day-to-day traffic situations. This task vastly depends on many subjective and objective factors, including the current driver state, her destination, personal preferences and abilities as well as surrounding environment. In this paper, we present CoSI (Context and Situation Intelligence), a Knowledge Graph (KG)-based approach for fusing and organizing heterogeneous types and sources of information. The KG serves as a coherence layer representing information in the form of entities and their inter-relationships augmented with additional semantic axioms. Harnessing the power of axiomatic rules and reasoning capabilities enables inferring additional knowledge from what is already encoded. Thus, dedicated components exploit and consume the semantically enriched information to perform tasks such as situation classification, difficulty assessment, and trajectory prediction.  Further, we generated a synthetic dataset to simulate real driving scenarios with a large range of driving styles and vehicle configurations.  We use KG embedding techniques based on a Graph Neural Network (GNN) architecture for a classification task of driving situations and achieve over 95% accuracy whereas vector-based approaches achieve only 75% accuracy for the same task.  The results suggest that the KG-based information representation combined with GNN are well suited for situation understanding tasks as required in driver assistance and automated driving systems.

ESWC 2021 Conference In-Use Track Submissions

Pay-as-you-go Population of an Automotive Signal Knowledge Graph

A Knowledge Graph-based Approach for Situation Comprehension in Driving Scenarios