Skip to content
EcoSimLab
DeutschEnglish
Set

EcoSimLab (live)

Eco-driving studies in the IMIS driving simulator — realistic EV energy simulation with a live data flow.

7 included tools
The driving simulator of the lab: a seat with steering wheel and pedals in front of three screens with a driving scene.

What the set enables

EcoSimLab (live) is the toolkit's lab setup: the IMIS driving simulator with integrated driving-energy simulation. In it, energy interfaces are developed and evaluated, and questions of user–energy interaction are investigated. The aim is a realistic simulation of the energy-relevant parameters in the driver–vehicle interaction. The tools of the set — test track, Python orchestration, display library and study backend — mesh with one another over an MQTT/WebSocket data flow. EcoSimLab (live) is the live/lab counterpart to the video-based set VideoSimLab.

Data flow

Live operation follows the C4 architecture from Paper 1. The Python API (EcoSimLabPy) drives BeamNG.tech. The real-time data run over the MQTT broker (Eclipse Mosquitto) to the study control backend. From there, they continue via MySQL and WebSocket to EcoInterfaces and StudyControlApp. Events (e.g. a driving scenario restart) run back via MQTT. Methodologically, three elements from Paper 2 support the setup. They are a calibrated EV energy simulation, an optimisation-based structuring of eco-driving and a synthetic driver as a benchmark.

Hardware setup

The IMIS driving simulator offers a 180° field of view across three 55“ displays (120 Hz). It also has a height- and length-adjustable steering wheel with the full 920° range of rotation of the Renault Zoe (Fanatec ClubSport DD+). The pedals are customised, the brake pedal carries a load cell. The sim rig is designed specifically for the interior and the ergonomics of the Renault Zoe. For the experimenter there is a remote setup for monitoring and driving scenario control.

Included tools

7
EcoInterfaces

EcoInterfaces

Library of the toolkit's eco-driving displays — SolidJS components, fed from an input contract.

Role in the setrenders the eco-driving driver displays live from the data stream (MQTT)

activeusable standalone
TechnologySolidJS, D3.js, TailwindCSS, Vite, TypeScript, Bun
  • EcoDrivingTestPark provides the energy-relevant test tracks as a BeamNG map
  • EcoSimLabPy orchestrates the live simulation and distributes driving data via MQTT
  • StudyControlApp controls the studies as a backend (driving scenarios, randomisation, monitoring, export)
  • EcoMPC performs the longitudinal control (Economic MPC) in the live setup
  • Fahrzeugmodelle provides the physics-based reference EV models (Renault Zoe)
  • Steering-Controller performs the lateral control, i.e. steering (Stanley / Pure Pursuit)

Cite

  • Jan Heidinger, Lukas Bernhardt, Thomas Franke (2023).
    EcoSimLab – A Low-Cost Driving Simulation Environment for Examining Human Factors in Vehicle Energy Efficiency
    AutomotiveUI '23 Adjunct (15th Intl. Conf. on Automotive User Interfaces), Ingolstadt
    10.1145/3581961.3609881
  • Markus Gödker, Steffen Schmees, Lukas Bernhardt, Jan Heidinger, Daniel Görges, Thomas Franke (2024).
    Driving Simulation for Energy Efficiency Studies: Analyzing Electric Vehicle Eco-Driving With EcoSimLab and the EcoDrivingTestPark
    AutomotiveUI '24 (16th Intl. Conf. on Automotive User Interfaces), Stanford, CA
    10.1145/3640792.3675706