Section: New Results

Decision-making for automated vehicles adapting human-like behavior

Participants : Pierre de Beaucorps, Thomas Streubel, Anne Verroust-Blondet, Fawzi Nashashibi.

Learning from human driver’s strategies for solving complex and potentially dangerous situations including interaction with other road users has the potential to improve decision-making methods for automated vehicles. In [37], we focus on simple unsignalized intersections and roundabouts in presence of another vehicle. We propose a human-like decision-making algorithm for these scenarios built up from human drivers recordings. The algorithm includes a risk assessment to avoid collisions in the intersection area. Three road topologies with different interaction scenarios were presented to human participants on a previously developed simulation tool. The same scenarios have been used to validate our decision-making process. We obtained promising results with no collisions in all setups and the ability to successfully determine to go before or after another vehicle.

A further study was conducted in [36] to assess the acceptability of the approach by human drivers.