Readiness-Aware AI Assistance for Human Drivers

Project Overview

The AI Assistance for Human Driver Readiness project studies how artificial intelligence systems can coordinate with human drivers under extreme time pressure, partial information, and changing risk. In shared-autonomy settings, drivers may need to interpret warnings, decide whether to trust AI recommendations, and take action within only a few seconds.

The project addresses a limitation of many current systems: they assume that humans are ready to respond at all times. In practice, readiness can change with attention, workload, stress, situation awareness, and trust. The research develops AI assistance that accounts for this changing and uncertain readiness when deciding when and how to interact with a driver.

Research

The project has three main research thrusts:

  1. Modeling human readiness: Developing computational models that estimate readiness from behavioral and non-invasive physiological measures, including response time, eye gaze, heart rate, and simulator interaction data.
  2. Decision-time interaction: Designing alerts, prompts, and low-burden cognitive probes that reduce uncertainty about readiness while limiting additional cognitive workload.
  3. Deployment-time regulation: Developing methods that maintain stable human-AI coordination as people adapt through repeated interactions with shared-autonomy systems.

Human-in-the-loop driving simulator studies, including scenarios involving advanced driver assistance systems such as forward collision warning, will evaluate the proposed methods under time-critical conditions.

Expected Impact

Expected outcomes include computational models for readiness-aware human-AI coordination, adaptive interaction algorithms, publicly available datasets and experimental resources, and evaluation methods for shared-autonomy systems operating under uncertainty and compressed decision horizons. The findings may support safer, more reliable, and more trustworthy AI systems in transportation, assistive robotics, healthcare, and other safety-critical domains.

Funding

Project Team