MTGNet: Multi-Agent End-to-End Motion Trajectory Prediction with Multimodal Panoramic Dynamic Graph

With the rapid development of autonomous driving technology, multi-agent trajectory prediction has become the core foundation of autonomous driving algorithms. Efficiently and accurately predicting the future trajectories of multiple agents is key to evaluating the reliability and safety of autonomo...

Full description

Saved in:
Bibliographic Details
Main Authors: Yinfei Dai, Yuantong Zhang, Xiuzhen Zhou, Qi Wang, Xiao Song, Shaoqiang Wang
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/10/5244
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850258249719218176
author Yinfei Dai
Yuantong Zhang
Xiuzhen Zhou
Qi Wang
Xiao Song
Shaoqiang Wang
author_facet Yinfei Dai
Yuantong Zhang
Xiuzhen Zhou
Qi Wang
Xiao Song
Shaoqiang Wang
author_sort Yinfei Dai
collection DOAJ
description With the rapid development of autonomous driving technology, multi-agent trajectory prediction has become the core foundation of autonomous driving algorithms. Efficiently and accurately predicting the future trajectories of multiple agents is key to evaluating the reliability and safety of autonomous driving vehicles. Recently, numerous studies have focused on capturing agent interactions in complex traffic scenarios. While most methods adopt agent-centric scene construction, they often rely on fixed scene sizes and incur significant computational overhead. Based on this, we propose the multimodal transformer graph convolution neural network (MTGNet) framework. The MTGNet framework can not only construct a panoramic, fully connected dynamic traffic map for agents but also dynamically adjust the size of traffic scenes. Additionally, it enables accurate and efficient multi-modal multi-agent trajectory prediction. In addition, we utilize the graph convolutional neural network (GCN) to process graph-structured data. This approach not only captures global relationships but also enhances the focus on local features within the scene, thereby improving the model’s sensitivity to local information. Our framework was tested on the Argoverse 2.0 dataset and compared with nine state-of-the-art vehicle trajectory prediction methods, achieving the best performance across all three selected metrics.
format Article
id doaj-art-c084f65459ba475992e56bf9cc80434a
institution OA Journals
issn 2076-3417
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-c084f65459ba475992e56bf9cc80434a2025-08-20T01:56:13ZengMDPI AGApplied Sciences2076-34172025-05-011510524410.3390/app15105244MTGNet: Multi-Agent End-to-End Motion Trajectory Prediction with Multimodal Panoramic Dynamic GraphYinfei Dai0Yuantong Zhang1Xiuzhen Zhou2Qi Wang3Xiao Song4Shaoqiang Wang5College of Computer Science and Technology, Changchun University, Changchun 130022, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaWith the rapid development of autonomous driving technology, multi-agent trajectory prediction has become the core foundation of autonomous driving algorithms. Efficiently and accurately predicting the future trajectories of multiple agents is key to evaluating the reliability and safety of autonomous driving vehicles. Recently, numerous studies have focused on capturing agent interactions in complex traffic scenarios. While most methods adopt agent-centric scene construction, they often rely on fixed scene sizes and incur significant computational overhead. Based on this, we propose the multimodal transformer graph convolution neural network (MTGNet) framework. The MTGNet framework can not only construct a panoramic, fully connected dynamic traffic map for agents but also dynamically adjust the size of traffic scenes. Additionally, it enables accurate and efficient multi-modal multi-agent trajectory prediction. In addition, we utilize the graph convolutional neural network (GCN) to process graph-structured data. This approach not only captures global relationships but also enhances the focus on local features within the scene, thereby improving the model’s sensitivity to local information. Our framework was tested on the Argoverse 2.0 dataset and compared with nine state-of-the-art vehicle trajectory prediction methods, achieving the best performance across all three selected metrics.https://www.mdpi.com/2076-3417/15/10/5244multimodalmulti-agent trajectory predictiongraph neural networktransformer
spellingShingle Yinfei Dai
Yuantong Zhang
Xiuzhen Zhou
Qi Wang
Xiao Song
Shaoqiang Wang
MTGNet: Multi-Agent End-to-End Motion Trajectory Prediction with Multimodal Panoramic Dynamic Graph
Applied Sciences
multimodal
multi-agent trajectory prediction
graph neural network
transformer
title MTGNet: Multi-Agent End-to-End Motion Trajectory Prediction with Multimodal Panoramic Dynamic Graph
title_full MTGNet: Multi-Agent End-to-End Motion Trajectory Prediction with Multimodal Panoramic Dynamic Graph
title_fullStr MTGNet: Multi-Agent End-to-End Motion Trajectory Prediction with Multimodal Panoramic Dynamic Graph
title_full_unstemmed MTGNet: Multi-Agent End-to-End Motion Trajectory Prediction with Multimodal Panoramic Dynamic Graph
title_short MTGNet: Multi-Agent End-to-End Motion Trajectory Prediction with Multimodal Panoramic Dynamic Graph
title_sort mtgnet multi agent end to end motion trajectory prediction with multimodal panoramic dynamic graph
topic multimodal
multi-agent trajectory prediction
graph neural network
transformer
url https://www.mdpi.com/2076-3417/15/10/5244
work_keys_str_mv AT yinfeidai mtgnetmultiagentendtoendmotiontrajectorypredictionwithmultimodalpanoramicdynamicgraph
AT yuantongzhang mtgnetmultiagentendtoendmotiontrajectorypredictionwithmultimodalpanoramicdynamicgraph
AT xiuzhenzhou mtgnetmultiagentendtoendmotiontrajectorypredictionwithmultimodalpanoramicdynamicgraph
AT qiwang mtgnetmultiagentendtoendmotiontrajectorypredictionwithmultimodalpanoramicdynamicgraph
AT xiaosong mtgnetmultiagentendtoendmotiontrajectorypredictionwithmultimodalpanoramicdynamicgraph
AT shaoqiangwang mtgnetmultiagentendtoendmotiontrajectorypredictionwithmultimodalpanoramicdynamicgraph