INVESTIGADORES
PARISI Daniel Ricardo
artículos
Título:
Data-driven simulation of pedestrian collision avoidance with a nonparametric neural network
Autor/es:
MARTIN, RAFAEL F.; PARISI, DANIEL R.
Revista:
NEUROCOMPUTING
Editorial:
ELSEVIER SCIENCE BV
Referencias:
Lugar: Amsterdam; Año: 2020 vol. 379 p. 130 - 140
ISSN:
0925-2312
Resumen:
Data-driven simulation of pedestrian dynamics is an incipient and promisingapproach for building reliable microscopic pedestrian models. We propose amethodology based on generalized regression neural networks, which does nothave to deal with a huge number of free parameters as in the case of multilayerneural networks. Although the method is general, we focus on the one pedestrian- one obstacle problem. Experimental data were collected in a motion capturelaboratory providing high-precision trajectories. The proposed model allowsus to simulate the trajectory of a pedestrian avoiding an obstacle from anydirection.
