Autonomous Driving ODD Ontology for Precise Scene Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing frameworks for characterizing Operational Design Domains (ODDs) in autonomous driving are inadequate for precise definition and application, leading to potential safety risks due to complex interactions between various factors and reliance on expert knowledge, which can result in mischaracterization and fatal consequences.
Innovation Solution
A formal ontology-based system is introduced to precisely define and assess ODDs using a common ontology language, enabling efficient matching of driving scenes with ODD definitions through a computer system that processes scene data and determines whether a scene is within or outside the defined ODD.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a formal ontology-based system is used to precisely define and assess ODDs, then measurement precision and reliability are improved, but device complexity increases
Solution Approach 1:
The ODD characterization system is segmented into distinct modular components: an ontology definition module that defines the formal language and structure, a scene processing module that extracts and represents driving scenes, and a matching module that compares scenes against ODD definitions. This segmentation allows each component to be developed and maintained independently while achieving precise ODD characterization through their coordinated interaction.
Solution Approach 2:
A formal ontology serves as an intermediary layer between raw driving scene data and ODD definitions. The ontology provides a standardized intermediate representation that enables precise matching and comparison, acting as a mediator that transforms complex real-world driving scenarios into structured, machine-readable formats that can be rigorously evaluated against operational design domain specifications.
2Device complexity
If existing descriptive frameworks are used to characterize ODDs, then device complexity is reduced, but measurement precision and reliability deteriorate
Solution Approach 1:
The system transforms the characterization of ODDs from qualitative, descriptive parameters to quantitative, formal parameters. By defining ODDs and driving scenes in terms of a formal ontology with explicit parameters and constraints, the system enables precise, unambiguous assessment. This parameter transformation allows for automated, consistent evaluation while maintaining manageable complexity through the structured nature of the ontology.
3Adaptability or versatility
If expert knowledge is relied upon to characterize ODDs, then adaptability is improved, but measurement precision deteriorates due to human error
Solution Approach 1:
The formal ontology-based system enables automated self-service ODD characterization without requiring continuous expert intervention. Once the ontology framework is established, the system can automatically process driving scenes, extract relevant features, and assess them against ODD definitions. This self-service capability maintains high precision by eliminating human error while preserving adaptability through the configurable nature of the ontology, allowing experts to define custom ODDs once and have them automatically applied and validated.
Data Source
AI summary
A computer system for analysing driving scenes in relation to an autonomous vehicle (AV) operational design domain (ODD), the computer system comprising: an input configured to receive a definition of the ODD in a formal ontology language; a scene processor configured to receive data of a driving scene and extract a scene representation therefrom, the data comprising an ego trace, at least one agent trace, and environmental data about an environment in which the traces were captured or generated, wherein the scene representation is an ontological representation of both static and dynamic elements of the driving scene extracted from the traces and the environmental data, and expressed in the same formal ontology language as the ODD; and a scene analyzer configured to match the static and dynamic elements of the scene representation with corresponding elements of the ODD, and thereby determine whether or not the driving scene is within the defined ODD.