DeepAI AI Chat
Log In Sign Up

Generating Reliable and Efficient Predictions of Human Motion: A Promising Encounter between Physics and Neural Networks

by   Alessandro Antonucci, et al.

Generating accurate and efficient predictions for the motion of the humans present in the scene is key to the development of effective motion planning algorithms for robots moving in promiscuous areas, where wrong planning decisions could generate safety hazard or simply make the presence of the robot "socially" unacceptable. Our approach to predict human motion is based on a neural network of a peculiar kind. Contrary to conventional deep neural networks, our network embeds in its structure the popular Social Force Model, a dynamic equation describing the motion in physical terms. This choice allows us to concentrate the learning phase in the aspects, which are really unknown (i.e., the model's parameters) and to keep the structure of the network simple and manageable. As a result, we are able to obtain a good prediction accuracy with a small synthetically generated training set, and the accuracy remains acceptable even when the network is applied in scenarios quite different from those for which it was trained. Finally, the choices of the network are "explainable", as they can be interpreted in physical terms. Comparative and experimental results prove the effectiveness of the proposed approach.


page 11

page 12


Hybrid Physics and Deep Learning Model for Interpretable Vehicle State Prediction

Physical motion models offer interpretable predictions for the motion of...

Probabilistic Human Motion Prediction via A Bayesian Neural Network

Human motion prediction is an important and challenging topic that has p...

Human Motion Prediction using Adaptable Neural Networks

Human motion prediction is an important component to facilitate human ro...

Planning Robot Motion using Deep Visual Prediction

In this paper, we introduce a novel framework that can learn to make vis...

Pairwise coupling of convolutional neural networks for better explicability of classification systems

We examine several aspects of explicability of a classification system b...

Constants of motion network

The beauty of physics is that there is usually a conserved quantity in a...

Integrating Deep-Learning-Based Image Completion and Motion Planning to Expedite Indoor Mapping

The challenge of autonomous indoor mapping is addressed. The goal is to ...