Skip to main content
Aller à la page d’accueil de la Commission européenne (s’ouvre dans une nouvelle fenêtre)
français fr
CORDIS - Résultats de la recherche de l’UE
CORDIS

Liability Allocation for Autonomous Vehicles: From Driving Modes to Insurance Pricing

Objectif

The proposed project, LiAVbility, introduces a novel, operational liability-sharing framework to address one of the most critical barriers to the large-scale deployment of autonomous vehicles (AVs): the lack of a scalable, publicly acceptable model for economic liability allocation. Rather than relying on traditional legal or ethics-led approaches, this project adopts a data-driven, engineering-based perspective to allocate accident liability based on who makes which driving decisions. Under this framework, AV users select a preferred driving mode (e.g. safe, eco, comfort, sport), while the system executes the corresponding algorithm developed by the AV manufacturer (OEM). The OEM assumes baseline liability for the safest mode, and users bear an incremental insurance premium if they opt for riskier modes, proportionate to the mode's ex ante crash risk.

To implement this, the project combines three interdisciplinary work packages. WP1 derives human driving preferences from naturalistic datasets using Inverse Soft Q-learning, clusters them, and trains distinct AV control policies via deep reinforcement learning. WP2 builds high-fidelity SUMO simulation networks calibrated against real-world traffic and accident data to generate accurate crash metrics for each mode. WP3 translates these metrics into actuarially grounded insurance premiums, further adjusted through large-scale public surveys to ensure fairness and social acceptability.

This project advances the state-of-the-art by linking technical AV control performance with societal deployment via a transparent and actionable liability-pricing model. Hosted at TU Delft with a secondment at VU Amsterdam, LiAVbility integrates control engineering, actuarial science, and behavioural economics to deliver a deployable platform supported by insurers and OEMs. It is expected to enhance trust, reduce litigation risk, and accelerate the safe, large-scale adoption of AVs across Europe.

Champ scientifique (EuroSciVoc)

Mots‑clés

Programme(s)

Thème(s)

Régime de financement

HORIZON-TMA-MSCA-PF-EF - HORIZON TMA MSCA Postdoctoral Fellowships - European Fellowships

Voir tous les projets financés dans le cadre de ce programme de financement

Appel à propositions

(s’ouvre dans une nouvelle fenêtre) HORIZON-MSCA-2025-PF

Voir tous les projets financés au titre de cet appel

Coordinateur

TECHNISCHE UNIVERSITEIT DELFT
Contribution nette de l'UE
€ 217 076,16
Coût total
Aucune donnée

Partenaires (1)