English

Collision Avoidance Detour for Multi-Agent Trajectory Forecasting

Computer Vision and Pattern Recognition 2023-06-21 v1 Robotics

Abstract

We present our approach, Collision Avoidance Detour (CAD), which won the 3rd place award in the 2023 Waymo Open Dataset Challenge - Sim Agents, held at the 2023 CVPR Workshop on Autonomous Driving. To satisfy the motion prediction factorization requirement, we partition all the valid objects into three mutually exclusive sets: Autonomous Driving Vehicle (ADV), World-tracks-to-predict, and World-others. We use different motion models to forecast their future trajectories independently. Furthermore, we also apply collision avoidance detour resampling, additive Gaussian noise, and velocity-based heading estimation to improve the realism of our simulation result.

Keywords

Cite

@article{arxiv.2306.11638,
  title  = {Collision Avoidance Detour for Multi-Agent Trajectory Forecasting},
  author = {Hsu-kuang Chiu and Stephen F. Smith},
  journal= {arXiv preprint arXiv:2306.11638},
  year   = {2023}
}

Comments

3rd place award, 2023 Waymo Open Dataset Challenge - Sim Agents, Workshop on Autonomous Driving of The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR Workshop) 2023

R2 v1 2026-06-28T11:09:48.735Z