{"title":"A Multimodal Trajectory Prediction Method for Pedestrian Crossing Considering Pedestrian Motion State","authors":"Zhuping Zhou, Bowen Liu, Changji Yuan, Ping Zhang","doi":"10.1109/mits.2023.3331817","DOIUrl":null,"url":null,"abstract":"Predicting pedestrian crossing trajectories has become a primary task in aiding autonomous vehicles to assess risks in pedestrian–vehicle interactions. As agile participants with changeable behavior, pedestrians are often capable of choosing from multiple possible crossing trajectories. Current research lacks the ability to predict multimodal trajectories with interpretability, and it also struggles to capture low-probability trajectories effectively. Addressing this gap, this article proposes a multimodal trajectory prediction model that operates by first estimating potential motion trends to prompt the generation of corresponding trajectories. It encompasses three sequential stages. First, pedestrian motion characteristics are analyzed, and prior knowledge of pedestrian motion states is obtained using the Gaussian mixture clustering method. Second, a long short-term memory model is employed to predict future pedestrian motion states, utilizing the acquired prior knowledge as input. Finally, the predicted motion states are discretized into various potential motion patterns, which are then introduced as prompts to the Spatio-Temporal Graph Transformer model for trajectory prediction. Experimental results on the Euro-PVI and BPI datasets demonstrate that the proposed model achieves cutting-edge performance in predicting pedestrian crossing trajectories. Notably, it significantly enhances the diversity, accuracy, and interpretability of pedestrian crossing trajectory predictions.","PeriodicalId":48826,"journal":{"name":"IEEE Intelligent Transportation Systems Magazine","volume":null,"pages":null},"PeriodicalIF":4.3000,"publicationDate":"2023-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Intelligent Transportation Systems Magazine","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1109/mits.2023.3331817","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0
Abstract
Predicting pedestrian crossing trajectories has become a primary task in aiding autonomous vehicles to assess risks in pedestrian–vehicle interactions. As agile participants with changeable behavior, pedestrians are often capable of choosing from multiple possible crossing trajectories. Current research lacks the ability to predict multimodal trajectories with interpretability, and it also struggles to capture low-probability trajectories effectively. Addressing this gap, this article proposes a multimodal trajectory prediction model that operates by first estimating potential motion trends to prompt the generation of corresponding trajectories. It encompasses three sequential stages. First, pedestrian motion characteristics are analyzed, and prior knowledge of pedestrian motion states is obtained using the Gaussian mixture clustering method. Second, a long short-term memory model is employed to predict future pedestrian motion states, utilizing the acquired prior knowledge as input. Finally, the predicted motion states are discretized into various potential motion patterns, which are then introduced as prompts to the Spatio-Temporal Graph Transformer model for trajectory prediction. Experimental results on the Euro-PVI and BPI datasets demonstrate that the proposed model achieves cutting-edge performance in predicting pedestrian crossing trajectories. Notably, it significantly enhances the diversity, accuracy, and interpretability of pedestrian crossing trajectory predictions.
期刊介绍:
The IEEE Intelligent Transportation Systems Magazine (ITSM) publishes peer-reviewed articles that provide innovative research ideas and application results, report significant application case studies, and raise awareness of pressing research and application challenges in all areas of intelligent transportation systems. In contrast to the highly academic publication of the IEEE Transactions on Intelligent Transportation Systems, the ITS Magazine focuses on providing needed information to all members of IEEE ITS society, serving as a dissemination vehicle for ITS Society members and the others to learn the state of the art development and progress on ITS research and applications. High quality tutorials, surveys, successful implementations, technology reviews, lessons learned, policy and societal impacts, and ITS educational issues are published as well. The ITS Magazine also serves as an ideal media communication vehicle between the governing body of ITS society and its membership and promotes ITS community development and growth.