Data-driven simulation of pedestrian collision avoidance with a\n nonparametric neural network

2019/07/17 by R. Martín, Martin, Rafael F., Daniel R. Parisi +1
Computer Science · Engineering · #Anomaly Detection Techniques and Applications #Autonomous Vehicle Technology and Safety #FOS: Physical sciences #Physics and Society (physics.soc-ph) #Traffic Prediction and Management Techniques

paper · pdf · doi:10.48550/arxiv.1907.07702

openalex publication_date 2019/07/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Data-driven simulation of pedestrian dynamics is an incipient and promising\napproach for building reliable microscopic pedestrian models. We propose a\nmethodology based on generalized regression neural networks, which does not\nhave to deal with a huge number of free parameters as in the case of multilayer\nneural networks. Although the method is general, we focus on the one\npedestrian-one obstacle problem. Experimental data were collected in a motion\ncapture laboratory providing high-precision trajectories. The proposed model\nallows us to simulate the trajectory of a pedestrian avoiding an obstacle from\nany direction.\n

Related

• Time-continuous microscopic pedestrian models: an overview
• Pedestrian Prediction by Planning using Deep Neural Networks
• Prediction of Pedestrian Speed with Artificial Neural Networks
• SFMGNet: A Physics-based Neural Network To Predict Pedestrian Trajectories
• What the Constant Velocity Model Can Teach Us About Pedestrian Motion Prediction