Skip to Main content Skip to Navigation
Conference papers

A new neural network feature importance method: Application to mobile robots controllers gain tuning

Ashley Hill 1 Eric Lucet 1 Roland Lenain 2
1 LRI - Laboratoire de Robotique Interactive
DIASI - Département Intelligence Ambiante et Systèmes Interactifs : DRT/LIST/DIASI
Abstract : This paper proposes a new approach for feature importance of neural networks and subsequently a methodology 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.
Complete list of metadata

https://hal-cea.archives-ouvertes.fr/cea-03314585
Contributor : Eric Lucet <>
Submitted on : Thursday, August 5, 2021 - 10:30:23 AM
Last modification on : Tuesday, September 7, 2021 - 3:44:09 PM

File

2020_ICINCO_Hill.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : cea-03314585, version 1

Citation

Ashley Hill, Eric Lucet, Roland Lenain. A new neural network feature importance method: Application to mobile robots controllers gain tuning. ICINCO 2020, 17th International Conference on Informatics in Control, Automation and Robotics, Jul 2020, Paris, France. ⟨cea-03314585⟩

Share

Metrics

Record views

43

Files downloads

30