Assurance Cases for Certification of Efficiently Learned Evaluated Rapidly and Automatically using Theory of Evidence (ACCELERATE)
Abstract
The Assurance Cases for Certification of Efficiently Learned Evaluated Rapidly and Automatically using Theory of Evidence (ACCELERATE) project was part of the DARPA Automated Rapid Certification of Software (ARCOS) Program. The ACCELERATE project completed 20 months of a 48month development effort. During this period, we focused on the design and implementation of a logic for evaluating assurance cases, as well as on the development of an initial prototype for the construction, scoring, and browsing of assurance cases by certifiers. In the initial assessment of the ACCELERATE project, we demonstrated prototype implementations of each individual module of ACCELERATE: the logic module, an argument template library, a module for the automatic construction of assurance cases from argument templates, and a module to aid in the comprehensibility of assurance cases. After this initial implementation, the ACCELERATE team focused development around a motivating example: a geofence cage ceiling monitor assurance case (geofence assurance case). This open source example allowed our team to more concretely define the format of templates, how they should be instantiated into cases, and the correct scoring of the result with the logic module of ACCELERATE.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 17, 2022
- Accession Number
- AD1185498
Entities
People
- Ken Schmidt
Organizations
- Johns Hopkins University Applied Physics Laboratory