Publication

Human motion trajectory prediction using the Social Force Model for real-time and low computational cost applications

Conference Article

Conference

Iberian Robotics Conference (ROBOT)

Edition

6th

Pages

235–247

Doc link

https://doi.org/10.1007/978-3-031-58676-7_19

File

Download the digital copy of the doc pdf document

Abstract

Human motion trajectory prediction is a very important functionality for human-robot collaboration, specifically in accompany- ing, guiding, or approaching tasks, but also in social robotics, self-driving vehicles, or security systems. In this paper, a novel trajectory prediction model, Social Force Generative Adversarial Network (SoFGAN), is pro- posed. SoFGAN uses a Generative Adversarial Network (GAN) and So- cial Force Model (SFM) to generate different plausible people trajectories reducing collisions in a scene. Furthermore, a Conditional Variational Au- toencoder (CVAE) module is added to emphasize the destination learn- ing. We show that our method is more accurate in making predictions in UCY or BIWI datasets than most of the current state-of-the-art models and also reduces collisions in comparison to other approaches. Through real-life experiments, we demonstrate that the model can be used in real-time without GPU’s to perform good quality predictions with a low computational cost.

Categories

intelligent robots, learning (artificial intelligence), uncertainty handling.

Author keywords

Human Motion Prediction, Social Force Model, Generative Adversarial Network, Conditional Variational Autoencoder

Scientific reference

O. Gil and A. Sanfeliu. Human motion trajectory prediction using the Social Force Model for real-time and low computational cost applications, 6th Iberian Robotics Conference, 2023, Coimbra, Portugal, pp. 235–247, Springer.