A New Neural Network Feature Importance Method: Application to Mobile Robots Controllers Gain Tuning

Ashley Hill, Eric Lucet, Roland Lenain

2020

Abstract

This paper proposes a new approach for feature importance of neural networks and subsequently a methodology using the novel feature importance to determine useful sensor information in high performance controllers, using a trained neural network that predicts the quasi-optimal gain in real time. The neural network is trained using the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) algorithm, in order to lower a given objective function. The important sensor information for robotic control are determined using the described methodology. Then a proposed improvement to the tested control law is given, and compared with the neural network’s gain prediction method for real time gain tuning. As a results, crucial information about the importance of a given sensory information for robotic control is determined, and shown to improve the performance of existing controllers.

Paper Citation

in Harvard Style

Hill A., Lucet E. and Lenain R. (2020). A New Neural Network Feature Importance Method: Application to Mobile Robots Controllers Gain Tuning.In Proceedings of the 17th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO, ISBN 978-989-758-442-8, pages 188-194. DOI: 10.5220/0009888501880194

in Bibtex Style

@conference{icinco20,
author={Ashley Hill and Eric Lucet and Roland Lenain},
title={A New Neural Network Feature Importance Method: Application to Mobile Robots Controllers Gain Tuning},
booktitle={Proceedings of the 17th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,},
year={2020},
pages={188-194},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0009888501880194},
isbn={978-989-758-442-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,
TI - A New Neural Network Feature Importance Method: Application to Mobile Robots Controllers Gain Tuning
SN - 978-989-758-442-8
AU - Hill A.
AU - Lucet E.
AU - Lenain R.
PY - 2020
SP - 188
EP - 194
DO - 10.5220/0009888501880194