Multimodal Trajectory Prediction for Diverse Vehicle Types in Autonomous Driving with Heterogeneous Data and Physical Constraints
The accurate prediction of vehicle behavior is crucial for autonomous driving systems, impacting their safety and efficiency in complex urban environments. To address the challenge of multi-agent trajectory prediction, we propose a novel model integrating multiple input modalities, including histori...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/22/7323 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846152411391983616 |
|---|---|
| author | Maoning Ge Kento Ohtani Ming Ding Yingjie Niu Yuxiao Zhang Kazuya Takeda |
| author_facet | Maoning Ge Kento Ohtani Ming Ding Yingjie Niu Yuxiao Zhang Kazuya Takeda |
| author_sort | Maoning Ge |
| collection | DOAJ |
| description | The accurate prediction of vehicle behavior is crucial for autonomous driving systems, impacting their safety and efficiency in complex urban environments. To address the challenge of multi-agent trajectory prediction, we propose a novel model integrating multiple input modalities, including historical trajectories, map data, vehicle features, and interaction information. Our approach employs a Conditional Variational Autoencoder (CVAE) framework with a decoder that predicts control actions using the Gaussian Mixture Model (GMM) and then converts these actions into dynamically feasible trajectories through a bicycle model. Evaluated on the nuScenes dataset, the model achieves great performance across key metrics, including minADE<sub>5</sub> of 1.26 and minFDE<sub>5</sub> of 2.85, demonstrating robust performance across various vehicle types and prediction horizons. These results indicate that integrating multiple data sources, physical models, and probabilistic methods significantly improves trajectory prediction accuracy and reliability for autonomous driving. Our approach generates diverse yet realistic predictions, capturing the multimodal nature of future outcomes while adhering to Physical Constraints and vehicle dynamics. |
| format | Article |
| id | doaj-art-4aedf912c87945dbaf7c7b55b038b913 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-4aedf912c87945dbaf7c7b55b038b9132024-11-26T18:21:32ZengMDPI AGSensors1424-82202024-11-012422732310.3390/s24227323Multimodal Trajectory Prediction for Diverse Vehicle Types in Autonomous Driving with Heterogeneous Data and Physical ConstraintsMaoning Ge0Kento Ohtani1Ming Ding2Yingjie Niu3Yuxiao Zhang4Kazuya Takeda5Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-Ward, Nagoya 464-8601, JapanGraduate School of Informatics, Nagoya University, Furo-cho, Chikusa-Ward, Nagoya 464-8601, JapanZhejiang Fubang Technology Inc., Ningbo R&D Campus Block A, Ningbo 315048, ChinaGraduate School of Informatics, Nagoya University, Furo-cho, Chikusa-Ward, Nagoya 464-8601, JapanRoboSense Technology Co., Ltd., 701 Block B, 800 Naxian Road, Pudong, Shanghai 200131, ChinaGraduate School of Informatics, Nagoya University, Furo-cho, Chikusa-Ward, Nagoya 464-8601, JapanThe accurate prediction of vehicle behavior is crucial for autonomous driving systems, impacting their safety and efficiency in complex urban environments. To address the challenge of multi-agent trajectory prediction, we propose a novel model integrating multiple input modalities, including historical trajectories, map data, vehicle features, and interaction information. Our approach employs a Conditional Variational Autoencoder (CVAE) framework with a decoder that predicts control actions using the Gaussian Mixture Model (GMM) and then converts these actions into dynamically feasible trajectories through a bicycle model. Evaluated on the nuScenes dataset, the model achieves great performance across key metrics, including minADE<sub>5</sub> of 1.26 and minFDE<sub>5</sub> of 2.85, demonstrating robust performance across various vehicle types and prediction horizons. These results indicate that integrating multiple data sources, physical models, and probabilistic methods significantly improves trajectory prediction accuracy and reliability for autonomous driving. Our approach generates diverse yet realistic predictions, capturing the multimodal nature of future outcomes while adhering to Physical Constraints and vehicle dynamics.https://www.mdpi.com/1424-8220/24/22/7323multi-agent trajectory predictionmultimodal learningConditional Variational AutoencoderGaussian Mixture Modelautonomous driving |
| spellingShingle | Maoning Ge Kento Ohtani Ming Ding Yingjie Niu Yuxiao Zhang Kazuya Takeda Multimodal Trajectory Prediction for Diverse Vehicle Types in Autonomous Driving with Heterogeneous Data and Physical Constraints Sensors multi-agent trajectory prediction multimodal learning Conditional Variational Autoencoder Gaussian Mixture Model autonomous driving |
| title | Multimodal Trajectory Prediction for Diverse Vehicle Types in Autonomous Driving with Heterogeneous Data and Physical Constraints |
| title_full | Multimodal Trajectory Prediction for Diverse Vehicle Types in Autonomous Driving with Heterogeneous Data and Physical Constraints |
| title_fullStr | Multimodal Trajectory Prediction for Diverse Vehicle Types in Autonomous Driving with Heterogeneous Data and Physical Constraints |
| title_full_unstemmed | Multimodal Trajectory Prediction for Diverse Vehicle Types in Autonomous Driving with Heterogeneous Data and Physical Constraints |
| title_short | Multimodal Trajectory Prediction for Diverse Vehicle Types in Autonomous Driving with Heterogeneous Data and Physical Constraints |
| title_sort | multimodal trajectory prediction for diverse vehicle types in autonomous driving with heterogeneous data and physical constraints |
| topic | multi-agent trajectory prediction multimodal learning Conditional Variational Autoencoder Gaussian Mixture Model autonomous driving |
| url | https://www.mdpi.com/1424-8220/24/22/7323 |
| work_keys_str_mv | AT maoningge multimodaltrajectorypredictionfordiversevehicletypesinautonomousdrivingwithheterogeneousdataandphysicalconstraints AT kentoohtani multimodaltrajectorypredictionfordiversevehicletypesinautonomousdrivingwithheterogeneousdataandphysicalconstraints AT mingding multimodaltrajectorypredictionfordiversevehicletypesinautonomousdrivingwithheterogeneousdataandphysicalconstraints AT yingjieniu multimodaltrajectorypredictionfordiversevehicletypesinautonomousdrivingwithheterogeneousdataandphysicalconstraints AT yuxiaozhang multimodaltrajectorypredictionfordiversevehicletypesinautonomousdrivingwithheterogeneousdataandphysicalconstraints AT kazuyatakeda multimodaltrajectorypredictionfordiversevehicletypesinautonomousdrivingwithheterogeneousdataandphysicalconstraints |