Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks

1855 Ogledov
1008 Prenosov
Izvoz citacije: ABNT
HEGEDÜS, Ferenc ;BÉCSI, Tamás ;ARADI, Szilárd ;GÁSPÁR, Péter .
Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks. 
Strojniški vestnik - Journal of Mechanical Engineering, [S.l.], v. 65, n.3, p. 148-160, march 2019. 
ISSN 0039-2480.
Available at: <https://www.sv-jme.eu/sl/article/motion-planning-for-highly-automated-road-vehicles-with-a-hybrid-approach-using-nonlinear-optimization-and-artificial-neural-networks/>. Date accessed: 26 jan. 2021. 
doi:http://dx.doi.org/10.5545/sv-jme.2018.5802.
Hegedüs, F., Bécsi, T., Aradi, S., & Gáspár, P.
(2019).
Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks.
Strojniški vestnik - Journal of Mechanical Engineering, 65(3), 148-160.
doi:http://dx.doi.org/10.5545/sv-jme.2018.5802
@article{sv-jmesv-jme.2018.5802,
	author = {Ferenc  Hegedüs and Tamás  Bécsi and Szilárd  Aradi and Péter  Gáspár},
	title = {Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks},
	journal = {Strojniški vestnik - Journal of Mechanical Engineering},
	volume = {65},
	number = {3},
	year = {2019},
	keywords = {automated driving; motion planning; trajectory planning, vehicle control; nonlinear optimization; artificial neural networks},
	abstract = {Over the last decade, many different algorithms were developed for the motion planning of road vehicles due to the increasing interest in the automation of road transportation. To be able to ensure dynamical feasibility of the planned trajectories, nonholonomic dynamics of wheeled vehicles must be considered. Nonlinear optimization based trajectory planners are proven to satisfy this need, however this happens at the expense of increasing computational effort, which jeopardizes the real-time applicability of these methods. This paper presents an algorithm which offers a solution to this problematic with a hybrid approach using artificial neural networks (ANNs). First, a nonlinear optimization based trajectory planner is presented which ensures the dynamical feasibility with the model-based prediction of the vehicle’s motion. Next, an artificial neural network is trained to reproduce the behavior of the optimization based planning algorithm with the method of supervised learning. The generation of training data happens off-line, which eliminates the concerns about the computational requirements of the optimization-based method. The trained neural network then replaces the original motion planner in on-line planning tasks which significantly reduces computational effort and thus run-time. Furthermore, the output of the network is supervised by the model based motion prediction layer of the original optimization-based algorithm and can thus always be trusted. Finally, the performance of the hybrid method is benchmarked with computer simulations in terms of dynamical feasibility and run-time and the results are investigated. Examinations show that the computation time can be significantly reduced while maintaining the feasibility of resulting vehicle motions.},
	issn = {0039-2480},	pages = {148-160},	doi = {10.5545/sv-jme.2018.5802},
	url = {https://www.sv-jme.eu/sl/article/motion-planning-for-highly-automated-road-vehicles-with-a-hybrid-approach-using-nonlinear-optimization-and-artificial-neural-networks/}
}
Hegedüs, F.,Bécsi, T.,Aradi, S.,Gáspár, P.
2019 March 65. Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks. Strojniški vestnik - Journal of Mechanical Engineering. [Online] 65:3
%A Hegedüs, Ferenc 
%A Bécsi, Tamás 
%A Aradi, Szilárd 
%A Gáspár, Péter 
%D 2019
%T Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks
%B 2019
%9 automated driving; motion planning; trajectory planning, vehicle control; nonlinear optimization; artificial neural networks
%! Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks
%K automated driving; motion planning; trajectory planning, vehicle control; nonlinear optimization; artificial neural networks
%X Over the last decade, many different algorithms were developed for the motion planning of road vehicles due to the increasing interest in the automation of road transportation. To be able to ensure dynamical feasibility of the planned trajectories, nonholonomic dynamics of wheeled vehicles must be considered. Nonlinear optimization based trajectory planners are proven to satisfy this need, however this happens at the expense of increasing computational effort, which jeopardizes the real-time applicability of these methods. This paper presents an algorithm which offers a solution to this problematic with a hybrid approach using artificial neural networks (ANNs). First, a nonlinear optimization based trajectory planner is presented which ensures the dynamical feasibility with the model-based prediction of the vehicle’s motion. Next, an artificial neural network is trained to reproduce the behavior of the optimization based planning algorithm with the method of supervised learning. The generation of training data happens off-line, which eliminates the concerns about the computational requirements of the optimization-based method. The trained neural network then replaces the original motion planner in on-line planning tasks which significantly reduces computational effort and thus run-time. Furthermore, the output of the network is supervised by the model based motion prediction layer of the original optimization-based algorithm and can thus always be trusted. Finally, the performance of the hybrid method is benchmarked with computer simulations in terms of dynamical feasibility and run-time and the results are investigated. Examinations show that the computation time can be significantly reduced while maintaining the feasibility of resulting vehicle motions.
%U https://www.sv-jme.eu/sl/article/motion-planning-for-highly-automated-road-vehicles-with-a-hybrid-approach-using-nonlinear-optimization-and-artificial-neural-networks/
%0 Journal Article
%R 10.5545/sv-jme.2018.5802
%& 148
%P 13
%J Strojniški vestnik - Journal of Mechanical Engineering
%V 65
%N 3
%@ 0039-2480
%8 2019-03-27
%7 2019-03-27
Hegedüs, Ferenc, Tamás  Bécsi, Szilárd  Aradi, & Péter  Gáspár.
"Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks." Strojniški vestnik - Journal of Mechanical Engineering [Online], 65.3 (2019): 148-160. Web.  26 Jan. 2021
TY  - JOUR
AU  - Hegedüs, Ferenc 
AU  - Bécsi, Tamás 
AU  - Aradi, Szilárd 
AU  - Gáspár, Péter 
PY  - 2019
TI  - Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks
JF  - Strojniški vestnik - Journal of Mechanical Engineering
DO  - 10.5545/sv-jme.2018.5802
KW  - automated driving; motion planning; trajectory planning, vehicle control; nonlinear optimization; artificial neural networks
N2  - Over the last decade, many different algorithms were developed for the motion planning of road vehicles due to the increasing interest in the automation of road transportation. To be able to ensure dynamical feasibility of the planned trajectories, nonholonomic dynamics of wheeled vehicles must be considered. Nonlinear optimization based trajectory planners are proven to satisfy this need, however this happens at the expense of increasing computational effort, which jeopardizes the real-time applicability of these methods. This paper presents an algorithm which offers a solution to this problematic with a hybrid approach using artificial neural networks (ANNs). First, a nonlinear optimization based trajectory planner is presented which ensures the dynamical feasibility with the model-based prediction of the vehicle’s motion. Next, an artificial neural network is trained to reproduce the behavior of the optimization based planning algorithm with the method of supervised learning. The generation of training data happens off-line, which eliminates the concerns about the computational requirements of the optimization-based method. The trained neural network then replaces the original motion planner in on-line planning tasks which significantly reduces computational effort and thus run-time. Furthermore, the output of the network is supervised by the model based motion prediction layer of the original optimization-based algorithm and can thus always be trusted. Finally, the performance of the hybrid method is benchmarked with computer simulations in terms of dynamical feasibility and run-time and the results are investigated. Examinations show that the computation time can be significantly reduced while maintaining the feasibility of resulting vehicle motions.
UR  - https://www.sv-jme.eu/sl/article/motion-planning-for-highly-automated-road-vehicles-with-a-hybrid-approach-using-nonlinear-optimization-and-artificial-neural-networks/
@article{{sv-jme}{sv-jme.2018.5802},
	author = {Hegedüs, F., Bécsi, T., Aradi, S., Gáspár, P.},
	title = {Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks},
	journal = {Strojniški vestnik - Journal of Mechanical Engineering},
	volume = {65},
	number = {3},
	year = {2019},
	doi = {10.5545/sv-jme.2018.5802},
	url = {https://www.sv-jme.eu/sl/article/motion-planning-for-highly-automated-road-vehicles-with-a-hybrid-approach-using-nonlinear-optimization-and-artificial-neural-networks/}
}
TY  - JOUR
AU  - Hegedüs, Ferenc 
AU  - Bécsi, Tamás 
AU  - Aradi, Szilárd 
AU  - Gáspár, Péter 
PY  - 2019/03/27
TI  - Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks
JF  - Strojniški vestnik - Journal of Mechanical Engineering; Vol 65, No 3 (2019): Strojniški vestnik - Journal of Mechanical Engineering
DO  - 10.5545/sv-jme.2018.5802
KW  - automated driving, motion planning, trajectory planning, vehicle control, nonlinear optimization, artificial neural networks
N2  - Over the last decade, many different algorithms were developed for the motion planning of road vehicles due to the increasing interest in the automation of road transportation. To be able to ensure dynamical feasibility of the planned trajectories, nonholonomic dynamics of wheeled vehicles must be considered. Nonlinear optimization based trajectory planners are proven to satisfy this need, however this happens at the expense of increasing computational effort, which jeopardizes the real-time applicability of these methods. This paper presents an algorithm which offers a solution to this problematic with a hybrid approach using artificial neural networks (ANNs). First, a nonlinear optimization based trajectory planner is presented which ensures the dynamical feasibility with the model-based prediction of the vehicle’s motion. Next, an artificial neural network is trained to reproduce the behavior of the optimization based planning algorithm with the method of supervised learning. The generation of training data happens off-line, which eliminates the concerns about the computational requirements of the optimization-based method. The trained neural network then replaces the original motion planner in on-line planning tasks which significantly reduces computational effort and thus run-time. Furthermore, the output of the network is supervised by the model based motion prediction layer of the original optimization-based algorithm and can thus always be trusted. Finally, the performance of the hybrid method is benchmarked with computer simulations in terms of dynamical feasibility and run-time and the results are investigated. Examinations show that the computation time can be significantly reduced while maintaining the feasibility of resulting vehicle motions.
UR  - https://www.sv-jme.eu/sl/article/motion-planning-for-highly-automated-road-vehicles-with-a-hybrid-approach-using-nonlinear-optimization-and-artificial-neural-networks/
Hegedüs, Ferenc, Bécsi, Tamás, Aradi, Szilárd, AND Gáspár, Péter.
"Motion Planning for Highly Automated Road Vehicles with a Hybrid Approach Using Nonlinear Optimization and Artificial Neural Networks" Strojniški vestnik - Journal of Mechanical Engineering [Online], Volume 65 Number 3 (27 March 2019)

Avtorji

Inštitucije

  • Robert Bosch 1
  • Budapest University of Technology and Economics, Department of Control for Transportation and Vehicle Systems 2
  • Hungarian Academy of Sciences, Computer and Automation Research Institute 3

Informacije o papirju

Strojniški vestnik - Journal of Mechanical Engineering 65(2019)3, 148-160

https://doi.org/10.5545/sv-jme.2018.5802

Over the last decade, many different algorithms were developed for the motion planning of road vehicles due to the increasing interest in the automation of road transportation. To be able to ensure dynamical feasibility of the planned trajectories, nonholonomic dynamics of wheeled vehicles must be considered. Nonlinear optimization based trajectory planners are proven to satisfy this need, however this happens at the expense of increasing computational effort, which jeopardizes the real-time applicability of these methods. This paper presents an algorithm which offers a solution to this problematic with a hybrid approach using artificial neural networks (ANNs). First, a nonlinear optimization based trajectory planner is presented which ensures the dynamical feasibility with the model-based prediction of the vehicle’s motion. Next, an artificial neural network is trained to reproduce the behavior of the optimization based planning algorithm with the method of supervised learning. The generation of training data happens off-line, which eliminates the concerns about the computational requirements of the optimization-based method. The trained neural network then replaces the original motion planner in on-line planning tasks which significantly reduces computational effort and thus run-time. Furthermore, the output of the network is supervised by the model based motion prediction layer of the original optimization-based algorithm and can thus always be trusted. Finally, the performance of the hybrid method is benchmarked with computer simulations in terms of dynamical feasibility and run-time and the results are investigated. Examinations show that the computation time can be significantly reduced while maintaining the feasibility of resulting vehicle motions.

automated driving; motion planning; trajectory planning, vehicle control; nonlinear optimization; artificial neural networks