SLEDGE: Synthesizing Driving Environments with Generative Models and Rule-Based Traffic
Kashyap Chitta, Daniel Dauner, and Andreas Geiger
University of Tübingen, Tübingen AI Center

European Conference on Computer Vision (ECCV), 2024

This repo contains SLEDGE, the first generative simulator for vehicle motion planning trained on real-world driving logs. We will be publicly releasing our code for simulation, evaluation, and training (including pre-trained checkpoints).

• 18 Aug, 2024: We released v0.1 of the SLEDGE code!
• 01 Jul, 2024: Our paper was accepted at ECCV 2024 🇮🇹
• 27 Mar, 2024: We released our paper on arXiv!

1. Getting started
2. Changelog
3. Contact
4. Citation
5. Other resources

• Installation and download
• Running the autoencoder
• Running the diffusion model
• Simulation and visualization

• [2024/08/18] SLEDGE v0.1 release
  • Scripts for pre-processing and downloads
  • Raster-vector autoencoder (training & latent caching)
  • Latent diffusion models (training & scenario generation)
  • Simple simulations
  • SledgeBoard

• Add videos and talks
• Release checkpoints
• Metrics & complete simulation code
• SLEDGE v0.1 & camera ready release
• Initial repository & preprint release

If you have any questions or suggestions, please feel free to open an issue or contact us (daniel.dauner@uni-tuebingen.de).

If you find SLEDGE useful, please consider giving us a star 🌟 and citing our paper with the following BibTeX entry.

@InProceedings{Chitta2024ECCV, 
	title = {SLEDGE: Synthesizing Driving Environments with Generative Models and Rule-Based Traffic}, 
	author = {Kashyap Chitta and Daniel Dauner and Andreas Geiger}, 
	booktitle = {European Conference on Computer Vision (ECCV)}, 
	year = {2024}, 
}

• Special thanks to Agniv Sharma for his reimplementation of HDMapGen which we used as a baseline!
• NAVSIM | tuPlan garage | CARLA garage | Survey on E2EAD
• PlanT | KING | TransFuser | NEAT

(back to top)