GeoScenario Server

Overview

GeoScenario Server (GSS) is a road traffic scenario simulator with reactive agents (vehicles and pedestrians). GSS leverages Lanelet2 high-definition map format and a scenario modelling language called GeoScenario2. Agent behaviour is specified using Behaviour Trees.

Example use: Traffic recording and re-simulation

Traffic recording
Trafic re-simulation

Example use: Cut-in test scenario in GSS dashboard and UnrealEngine 5 co-simulation

Cut-in test scenario in GSS dash and co-simulation

Native GeoScenario Server simulation

By default, GSS simulations can be viewed using the built-in dashboard, which shows the map (top left), view focused on a selected agent (bottom left), and a table listing the state of all agents (top right).

Erb Erbsville Crescent traffic lights

Ring Road Columbia

Paradise Crescent all way stop

Ring Road clock-wise

Co-simulation with a high-fidelity simulator

Additionally, GSS can be integrated with a high-fidelity simulator based on UnrealEngine.

GeoScenario multi agent simulation demo with GSServer, UE5, and Carla

WISE Sim integration

GeoScenario Server has also been used as a traffic simulator together with WISE Sim for testing of WISE ADS running on UW Moose.

Resources

Source code repository

Papers

Queiroz, R., Sharma, D., Caldas, R., Czarnecki, K., García, S., Berger, T., Pelliccione, P., "A Driver-Vehicle Model for ADS Scenario-Based Testing" in IEEE Transactions on Intelligent Transportation Systems. 03/2024. 2024. Available in IEEEXplore

Larter, S., Queiroz, R., Sedwards, S., Sarkar, A., Czarnecki, K., A Hierarchical Pedestrian Behavior Model to Generate Realistic Human Behavior in Traffic Simulation. Aachen, Germany: IEEE, 2022. Available on IEEEXplore (Also available on Arxiv).

Theses

Rodrigo Queiroz, Scenario Modeling and Execution for Simulation Testing of Automated-Driving Systems. PhD, University of Waterloo, 2022.

Scott Larter, A Hierarchical Pedestrian Behaviour Model to Reproduce Realistic Human Behaviour in a Traffic Environment. MMath. 2022.

Scott Larter, " A Hierarchical Pedestrian Behavior Model to Generate Realistic Human Behavior in Traffic Simulation", MMath, 2022