Planning-oriented autonomous driving

Y Hu, J Yang, L Chen, K Li, C Sima… - Proceedings of the …, 2023 - openaccess.thecvf.com
Modern autonomous driving system is characterized as modular tasks in sequential order,
ie, perception, prediction, and planning. In order to perform a wide diversity of tasks and …

Motionlm: Multi-agent motion forecasting as language modeling

A Seff, B Cera, D Chen, M Ng, A Zhou… - Proceedings of the …, 2023 - openaccess.thecvf.com
Reliable forecasting of the future behavior of road agents is a critical component to safe
planning in autonomous vehicles. Here, we represent continuous trajectories as sequences …

Scene transformer: A unified architecture for predicting multiple agent trajectories

J Ngiam, B Caine, V Vasudevan, Z Zhang… - arXiv preprint arXiv …, 2021 - arxiv.org
Predicting the motion of multiple agents is necessary for planning in dynamic environments.
This task is challenging for autonomous driving since agents (eg vehicles and pedestrians) …

Gameformer: Game-theoretic modeling and learning of transformer-based interactive prediction and planning for autonomous driving

Z Huang, H Liu, C Lv - Proceedings of the IEEE/CVF …, 2023 - openaccess.thecvf.com
Autonomous vehicles operating in complex real-world environments require accurate
predictions of interactive behaviors between traffic participants. This paper tackles the …

Rethinking integration of prediction and planning in deep learning-based automated driving systems: a review

S Hagedorn, M Hallgarten, M Stoll… - arXiv preprint arXiv …, 2023 - arxiv.org
Automated driving has the potential to revolutionize personal, public, and freight mobility.
Besides the enormous challenge of perception, ie accurately perceiving the environment …

Differentiable integrated motion prediction and planning with learnable cost function for autonomous driving

Z Huang, H Liu, J Wu, C Lv - IEEE transactions on neural …, 2023 - ieeexplore.ieee.org
Predicting the future states of surrounding traffic participants and planning a safe, smooth,
and socially compliant trajectory accordingly are crucial for autonomous vehicles (AVs) …

Deep interactive motion prediction and planning: Playing games with motion prediction models

JLV Espinoza, A Liniger, W Schwarting… - … for Dynamics and …, 2022 - proceedings.mlr.press
Abstract In most classical Autonomous Vehicle (AV) stacks, the prediction and planning
layers are separated, limiting the planner to react to predictions that are not informed by the …

MoST: Multi-modality Scene Tokenization for Motion Prediction

N Mu, J Ji, Z Yang, N Harada, H Tang… - Proceedings of the …, 2024 - openaccess.thecvf.com
Many existing motion prediction approaches rely on symbolic perception outputs to generate
agent trajectories such as bounding boxes road graph information and traffic lights. This …

Uncertainty-aware model-based offline reinforcement learning for automated driving

C Diehl, TS Sievernich, M Krüger… - IEEE Robotics and …, 2023 - ieeexplore.ieee.org
Offline reinforcement learning (RL) provides a framework for learning decision-making from
offline data and therefore constitutes a promising approach for real-world applications such …

InterSim: Interactive traffic simulation via explicit relation modeling

Q Sun, X Huang, BC Williams… - 2022 IEEE/RSJ …, 2022 - ieeexplore.ieee.org
Interactive traffic simulation is crucial to autonomous driving systems by enabling testing for
planners in a more scalable and safe way compared to real-world road testing. Existing …