TrafficBots V1.5: Traffic Simulation via Conditional VAEs and Transformers with Relative Pose Encoding

Zhejun Zhang, Christos Sakaridis, Luc Van Gool
6/16/2024
cs.ROcs.CV

Official code: zhejz/TrafficBotsV1.5 (42 stars, PyTorch, last commit 2024-06-18).

Abstract

In this technical report we present TrafficBots V1.5, a baseline method for the closed-loop simulation of traffic agents. TrafficBots V1.5 achieves baseline-level performance and a 3rd place ranking in the Waymo Open Sim Agents Challenge (WOSAC) 2024. It is a simple baseline that combines TrafficBots, a CVAE-based multi-agent policy conditioned on each agent's individual destination and personality, and HPTR, the heterogeneous polyline transformer with relative pose encoding. To improve the performance on the WOSAC leaderboard, we apply scheduled teacher-forcing at the training time and we filter the sampled scenarios at the inference time. The code is available at https://github.com/zhejz/TrafficBotsV1.5.

View on arXivView PDF

Code Implementations(1)

zhejz/TrafficBotsV1.5Official100%
PyTorch

TrafficBots V1.5: TrafficBots + HPTR. 3rd place solution for Waymo Open Sim Agent Challenge 2024.

425Shell, PythonMay 22, 20242 years agoNOASSERTION
autonomous-drivingcvpr2024motion-predictionpytorchtraffic-simulation+2 more

Similar Papers

Analyzing Closed-loop Training Techniques for Realistic Traffic Agent Models in Autonomous Highway Driving Simulations

83% match2024

CityFlowER: An Efficient and Realistic Traffic Simulator with Embedded Machine Learning Models

83% match1 repo262024

RLFTSim: Realistic and Controllable Multi-Agent Traffic Simulation via Reinforcement Learning Fine-Tuning

82% match1 repo12026

VissimRL: A Multi-Agent Reinforcement Learning Framework for Traffic Signal Control Based on Vissim

82% match1 repo52026

Toward Efficient and Robust Behavior Models for Multi-Agent Driving Simulation

82% match2025

Decoupled Intelligence: A Multi-Agent LLM Framework for Controllable Traffic Scenario Generation in SUMO

82% match2026

Cite this paper

@article{zhang2024trafficbots,
  title  = {TrafficBots V1.5: Traffic Simulation via Conditional VAEs and Transformers with Relative Pose Encoding},
  author = {Zhejun Zhang and Christos Sakaridis and Luc Van Gool},
  year   = {2024},
  eprint = {2406.10898},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2406.10898v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2406.10898.svg)](https://codeofpaper.com/paper/2406.10898)

Discussion