A Radar-Nearest-Neighbor based data-driven approach for crowd simulation. (August 2021)

• Record Type: Journal Article
• Authors: Zhao, Xuedan; Zhang, Jun; Song, Weiguo
• Is Part Of: Transportation research. Volume 129(2021)
• Publication Date: 2021-08
• Subjects: Crowd simulation -- Artificial neural network -- Data-driven -- Pedestrian dynamics -- Radar-Nearest-Neighbor; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2021.103260
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: A learnable data-driven motion model is established for crowd simulation. The Radar-Nearest-Neighbor concept is given to find pedestrians' nearest neighbors. Novel pedestrian motion states are designed based on the proposed Radar-NN method. The movement information is learned by the motion model from the motion states. The model is validated to be effective for bidirectional pedestrian flow simulation. Abstract: In this work, a learnable data-driven motion model namely Multi-Feature Fusion Recursive Neural Network (MFF-RNN) is proposed. The model yields pedestrians' velocities by learning from the designed motion states consisting of the relative distances and velocities with neighbors, as well as individuals' previous velocity sequences. A novel Radar-Nearest-Neighbor (Radar-NN) method is developed to determine the nearest neighbors of a pedestrian by treating him/her as a radar and detecting the surrounding environment within a limited circular receptive field. Bidirectional flow scenarios are adopted to evaluate the performance of the proposed model and the lane formation phenomenon can be successfully reproduced. The simulation results coincide with that of experiments and are superior to the work of Ma et al. in pedestrian trajectories, distributions, as well as fundamental diagrams. By calculating five evaluation metrics, it shows that the errors of our model are reduced by 34.1–79.0% compared with their work.