TSS-INSID Project
In-Silico Driving Assurance for Safe Autonomous Vehicles
In-Silico Driving Assurance Using Machine-Learning-Powered Verification and Synthesis for Safe Autonomous Vehicles (TSS-INSID)
A Mälardalen University project within the Trusted Smart Systems initiative, concerned with assuring the safety of automated driving in simulation.
Key Focus:
- Scenario-based testing of Automated Driving Systems
- Generating executable simulation scenarios from natural-language descriptions
- Combining large language models with structured compilation and deterministic validation
My Contribution:
TSS-INSID co-funded HASCO, a Hybrid AI Simulation Compiler that translates natural-language accident reports into executable OpenSCENARIO scenarios for the esmini simulator. The work was carried out together with Prover Technology, Volvo Cars, and Chalmers University of Technology, and published at AEiC 2026.
Read the HASCO paper or see the publications page.