EcoSimLab
"The vision of the IMIS driving simulator is an integrated driving energy simulation in which energy interfaces can be developed and evaluated and user-energy interaction issues can be investigated. Our aim is therefore the realistic simulation of energy-relevant parameters in driver-vehicle interaction"
Prof. Dr. Thomas Franke
Hardware
The EcoSimLab driving simulator features a 180° field of view using
three 55-inch diagonal 120Hz displays, delivering stable 50fps
rendering at 12k resolution. Powered by a Ryzen Threadripper 3970X and
a GeForce RTX 4090, it provides reliable performance.
The steering wheel is adjusted in height and length by the driver,
supporting the full 920° rotation of the Renault ZOE. Precise input
resolution and above-spec force feedback are provided by a Fanatec
ClubSport DD+ Module. The pedals are adjusted in pedal-travel and
resistance force, with a load cell used for the brake pedal sensor to
offer improved braking characteristics.
The simrig is custom-built to match the Renault ZOE's interior
measurements and ergonomics, ensuring that participants experience
driving conditions as close to the real vehicle as possible. This
accurate ergonomic replication is crucial, as it allows participants
to perform complex maneuvers without added difficulty or
simplification, supporting realistic driver behavior during studies.
Integrated OEM parts and interfaces further enhance the authenticity
of the simulation.
Additionally, EcoSimHub includes a study operator workspace, a fully
remote setup for monitoring and controlling scenarios during studies.
This workspace enables a single operator to prepare the simulator and
manage the study, ensuring efficient study execution.
Software
BeamNG
At EcoSimLab, we leverage BeamNG as our driving simulation software.
Built on a highly accurate physics engine, BeamNG offers detailed,
high-fidelity rendering. Its flexible platform allows us to customize
vehicles or create new ones entirely, while the powerful API supports
seamless integration with our research tools.
EcoInterfaces
A core component of our research at EcoSimLab is EcoInterfaces,
focusing on human-energy interaction. These interfaces are designed to
enhance driver awareness and energy efficiency. Cross-platform
compatibility allows for their use not only in simulators but also for
testing in real vehicles, bridging the gap between virtual and
physical environments.
Research Vehicle
The Renault ZOE Q90 serves as the reference vehicle for the AMORi
project. Its reverse-engineered powertrain is precisely modeled in
BeamNG, with accurate loss and efficiency representation. Fine-tuned
controls, exact pedal mapping, and a customized chassis replicate
real-world aerodynamics and road behavior. Our open-source simulation
is so precise, it can be used as an objective metric in efficiency
studies.
EcoDriving Testpark
The EcoDriving Testpark addresses a key challenge in driving
efficiency studies: typical drives often lack the frequency of
energy-relevant maneuvers, wasting valuable study time on steady-speed
driving. Implemented as a custom map and scenario in BeamNG, the
Testpark includes as many energy-relevant maneuvers as possible with
minimal filler distances. Designed to differentiate driver
characteristics, it features a balanced spread of control complexity.
The project is open-source and continuously refined to support
advanced research in eco-driving.
Study Control App
The Study Control App transforms BeamNG into a robust platform for
user studies. With a built-in scenario editor, it enables precise
study setup and remote monitoring. The app’s backend connects
seamlessly with databases, MATLAB, EcoInterfaces, and eyetracking
systems, making it a versatile tool for data collection and analysis.
Open APIs ensure flexibility and integration with various research
modules.
EcoMPC
EcoMPC is an economic model predictive controller that powers our
EcoInterfaces. Built on an offline algorithm, it calculates the most
efficient driving behavior for predefined conditions and routes. By
comparing the current driving state with the optimal one, it provides
impulses for corrective action. With a dynamic prediction horizon and
real-time operation through linearization, this MATLAB-based system
ensures efficient and adaptive driving control.
Infrastructure
Our infrastructure is designed with flexibility in mind, enabling
bidirectional communication between the simulation and additional
modules like EcoInterfaces and EcoMPC. At its core, it integrates
BeamNG.tech, the driving data database, and the Study Control App.
This robust setup allows for the creation and execution of large-scale
driving simulator studies, supporting high participant counts. With
precise data collection, real-time monitoring, and seamless
integration with tools such as MATLAB, our infrastructure ensures that
both new and ongoing studies can be conducted efficiently and
effectively.
Research
Theses
Jonas Busch (2023)
Entwicklung und Anwendung eines Auswertungstools für Blickbewegungsdaten in Fahrsimulationsstudien mit Live-EnergieanzeigenTobias Harms (2023)
Integration eines MATLAB Eco-MPC in die Fahrsimulationssoftware BeamNG.tech im Kontext der nutzerzentrierten Energieeffizienzanzeigen Forschung in der ElektromobilitätTobias Reins (2023)
Entwicklung eines Systems zur Planung, Durchführung und Auswertung von Simulator-StudienJosua Potschien (2022)
Systematische Untersuchung von Metriken hinsichtlich der Bewertung der Energieeffizienz und Eignung in verschiedenen EnergieanzeigenIndrani Breier (2021)
Tipp-Basiertes Ecodriving Feedback: Entwicklung eines Post-Manöver-Assistenzsystems zum Training energieeffizienten Fahrverhaltens in batteriebetriebenen ElektrofahrzeugenAnton de Vries (2021)
Beschleunigungsdisplays im Fahrsimulator - Untersuchung des Einflusses der Displayart auf die EnergieeffizienzPublications
Gödker, M., Schmees, S., Bernhardt. L., Heidinger, J., Görges, D., and Franke, T. (2024)
Driving Simulation for Energy Efficiency Studies: Analyzing Electric Vehicle Eco-Driving With EcoSimLab and the EcoDrivingTestParkIn Proceedings of the 16th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (AutomotiveUI '24). Association for Computing Machinery, New York, NY, USA, 32–42
https://doi.org/10.1145/3640792.3675706Gödker, M., & Franke, T. (2024)
Assessing Energy-Related Situation Awareness Using Self-Controlled Occlusion During Electric Vehicle Driving Scenes148, 286-296
https://doi.org/10.54941/ahfe1005219Heidinger, J., Bernhardt, L., & Franke, T. (2023)
Ecosimlab - A Low-Cost Driving Simulator Environment for Examining Human Factors in Vehicle Energy EfficiencyAutomotiveUI '23 Adjunct: Adjunct Proceedings of the 15th International Conference on Automotive User Interfaces and Interactive Vehicular Applications, , Ingolstadt (Germany), 18.-21.09.2023, 66-70
https://doi.org/10.1145/3581961.3609881Moll, V. E., & Franke, T. (2021)
Biased energy efficiency perception based on instantaneous consumption displays–Indication for heuristic energy information processingApplied Ergonomics, 94, 103399
https://doi.org/10.1016/j.apergo.2021.103399Franke, T., Görges, D., & Arend, M. G. (2019, September)
The Energy Interface Challenge. Towards Designing Effective Energy Efficiency Interfaces for Electric VehiclesIn Proceedings of the 11th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (pp. 35-48)
https://doi.org/10.1145/3342197.3344526Bernhardt, L., Moll, V., & Franke, T. (2019)
Entwicklung und Validierung einer Fahrsimulationsumgebung für Interface-Testing im Kontext Elektromobilität3. Kongress der Fachgruppe Verkehrspsychologie, Saarbrücken, 05.-07.03.2019
https://doi.org/10.13140/RG.2.2.26178.32962/1Projects
Team
Project Management
Ph. D. Students
Students
