Performance Analysis of Deep Neural Network Controller for Autonomous Driving Learning from a Nonlinear Model Predictive Control Method
Nonlinear model predictive control (NMPC) is based on a numerical optimization method considering the target system dynamics as constraints. This optimization process requires large amount of computation power and the computation time is often unpredictable which may cause the control update rate to...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-03-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/10/7/767 |
_version_ | 1797540224884015104 |
---|---|
author | Taekgyu Lee Yeonsik Kang |
author_facet | Taekgyu Lee Yeonsik Kang |
author_sort | Taekgyu Lee |
collection | DOAJ |
description | Nonlinear model predictive control (NMPC) is based on a numerical optimization method considering the target system dynamics as constraints. This optimization process requires large amount of computation power and the computation time is often unpredictable which may cause the control update rate to overrun. Therefore, the performance must be carefully balanced against the computational time. To solve the computation problem, we propose a data-based control technique based on a deep neural network (DNN). The DNN is trained with closed-loop driving data of an NMPC. The proposed "DNN control technique based on NMPC driving data" achieves control characteristics comparable to those of a well-tuned NMPC within a reasonable computation period, which is verified with an experimental scaled-car platform and realistic numerical simulations. |
first_indexed | 2024-03-10T12:57:06Z |
format | Article |
id | doaj.art-7194de025d504d19952a2f1de7340508 |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-10T12:57:06Z |
publishDate | 2021-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-7194de025d504d19952a2f1de73405082023-11-21T11:48:45ZengMDPI AGElectronics2079-92922021-03-0110776710.3390/electronics10070767Performance Analysis of Deep Neural Network Controller for Autonomous Driving Learning from a Nonlinear Model Predictive Control MethodTaekgyu Lee0Yeonsik Kang1Graduate School of Automotive Engineering, Kookmin University, Seoul 02707, KoreaGraduate School of Automotive Engineering, Kookmin University, Seoul 02707, KoreaNonlinear model predictive control (NMPC) is based on a numerical optimization method considering the target system dynamics as constraints. This optimization process requires large amount of computation power and the computation time is often unpredictable which may cause the control update rate to overrun. Therefore, the performance must be carefully balanced against the computational time. To solve the computation problem, we propose a data-based control technique based on a deep neural network (DNN). The DNN is trained with closed-loop driving data of an NMPC. The proposed "DNN control technique based on NMPC driving data" achieves control characteristics comparable to those of a well-tuned NMPC within a reasonable computation period, which is verified with an experimental scaled-car platform and realistic numerical simulations.https://www.mdpi.com/2079-9292/10/7/767data-driven controlmodel predictive controlartificial neural networkautonomous drivingdeep neural network controlartificial intelligence |
spellingShingle | Taekgyu Lee Yeonsik Kang Performance Analysis of Deep Neural Network Controller for Autonomous Driving Learning from a Nonlinear Model Predictive Control Method Electronics data-driven control model predictive control artificial neural network autonomous driving deep neural network control artificial intelligence |
title | Performance Analysis of Deep Neural Network Controller for Autonomous Driving Learning from a Nonlinear Model Predictive Control Method |
title_full | Performance Analysis of Deep Neural Network Controller for Autonomous Driving Learning from a Nonlinear Model Predictive Control Method |
title_fullStr | Performance Analysis of Deep Neural Network Controller for Autonomous Driving Learning from a Nonlinear Model Predictive Control Method |
title_full_unstemmed | Performance Analysis of Deep Neural Network Controller for Autonomous Driving Learning from a Nonlinear Model Predictive Control Method |
title_short | Performance Analysis of Deep Neural Network Controller for Autonomous Driving Learning from a Nonlinear Model Predictive Control Method |
title_sort | performance analysis of deep neural network controller for autonomous driving learning from a nonlinear model predictive control method |
topic | data-driven control model predictive control artificial neural network autonomous driving deep neural network control artificial intelligence |
url | https://www.mdpi.com/2079-9292/10/7/767 |
work_keys_str_mv | AT taekgyulee performanceanalysisofdeepneuralnetworkcontrollerforautonomousdrivinglearningfromanonlinearmodelpredictivecontrolmethod AT yeonsikkang performanceanalysisofdeepneuralnetworkcontrollerforautonomousdrivinglearningfromanonlinearmodelpredictivecontrolmethod |