TY - GEN

T1 - ForceFormer

T2 - 34th IEEE Intelligent Vehicles Symposium, IV 2023

AU - Zhang, Weicheng

AU - Cheng, Hao

AU - Johora, Fatema T.

AU - Sester, Monika

N1 - Funding information: 1Weicheng Zhang and Monika Sester are with the Institute of Cartography and Geoinformatics, Leibniz University Hannover, Appelstr. 9, 30167 Hannover, Germany weicheng.zhang@stud.uni-hannover.de, Monika.Sester@ikg.uni-hannover.de 2Hao Cheng is with the Faculty of Geo-Information Science and Earth Observation, University of Twente, 7500 AE Enschede, The Netherlands. Cheng is funded by MSCA European Postdoctoral Fellowships under the 101062870 VeVuSafety project. h.cheng-2@utwente.nl 2Fatema T. Johora is with the Department of Informatics, Clausthal University of Technology, Julius-Albert-Str. 4, 38678 Clausthal-Zellerfeld, Germany fatema.tuj.johora@tu-clausthal.de *Corresponding author

PY - 2023

Y1 - 2023

N2 - Predicting trajectories of pedestrians based on goal information in highly interactive scenes is a crucial step toward Intelligent Transportation Systems and Autonomous Driving. The challenges of this task come from two key sources: (1) complex social interactions in high pedestrian density scenarios and (2) limited utilization of goal information to effectively associate with past motion information. To address these difficulties, we integrate social forces into a Transformer-based stochastic generative model backbone and propose a new goal-based trajectory predictor called ForceFormer. Differentiating from most prior works that simply use the destination position as an input feature, we leverage the driving force from the destination to efficiently simulate the guidance of a target on a pedestrian. Additionally, repulsive forces are used as another input feature to describe the avoidance action among neighboring pedestrians. Extensive experiments show that our proposed method achieves on-par performance measured by distance errors with the state-of-the-art models but evidently decreases collisions, especially in dense pedestrian scenarios on widely used pedestrian datasets.

AB - Predicting trajectories of pedestrians based on goal information in highly interactive scenes is a crucial step toward Intelligent Transportation Systems and Autonomous Driving. The challenges of this task come from two key sources: (1) complex social interactions in high pedestrian density scenarios and (2) limited utilization of goal information to effectively associate with past motion information. To address these difficulties, we integrate social forces into a Transformer-based stochastic generative model backbone and propose a new goal-based trajectory predictor called ForceFormer. Differentiating from most prior works that simply use the destination position as an input feature, we leverage the driving force from the destination to efficiently simulate the guidance of a target on a pedestrian. Additionally, repulsive forces are used as another input feature to describe the avoidance action among neighboring pedestrians. Extensive experiments show that our proposed method achieves on-par performance measured by distance errors with the state-of-the-art models but evidently decreases collisions, especially in dense pedestrian scenarios on widely used pedestrian datasets.

UR - http://www.scopus.com/inward/record.url?scp=85168003419&partnerID=8YFLogxK

U2 - 10.48550/arXiv.2302.07583

DO - 10.48550/arXiv.2302.07583

M3 - Conference contribution

AN - SCOPUS:85168003419

SN - 979-8-3503-4692-3

T3 - IEEE Intelligent Vehicles Symposium

BT - 2023 IEEE Intelligent Vehicles Symposium (IV)

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 4 June 2023 through 7 June 2023

ER -