Deep imitation reinforcement learning for self‐driving by vision

Abstract Deep reinforcement learning has achieved some remarkable results in self‐driving. There is quite a lot of work to do in the area of autonomous driving with high real‐time requirements because of the inefficiency of reinforcement learning in exploring large continuous motion spaces. A deep i...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Qijie Zou, Kang Xiong, Qiang Fang, Bohan Jiang
Formato:	article
Lenguaje:	EN
Publicado:	Wiley 2021
Materias:	Computational linguistics. Natural language processing P98-98.5 Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/fe4e98da1afc4fd195e561d3feac3d0c
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:fe4e98da1afc4fd195e561d3feac3d0c
record_format	dspace
spelling	oai:doaj.org-article:fe4e98da1afc4fd195e561d3feac3d0c2021-11-17T03:12:43ZDeep imitation reinforcement learning for self‐driving by vision2468-232210.1049/cit2.12025https://doaj.org/article/fe4e98da1afc4fd195e561d3feac3d0c2021-12-01T00:00:00Zhttps://doi.org/10.1049/cit2.12025https://doaj.org/toc/2468-2322Abstract Deep reinforcement learning has achieved some remarkable results in self‐driving. There is quite a lot of work to do in the area of autonomous driving with high real‐time requirements because of the inefficiency of reinforcement learning in exploring large continuous motion spaces. A deep imitation reinforcement learning (DIRL) framework is presented to learn control policies of self‐driving vehicles, which is based on a deep deterministic policy gradient algorithm (DDPG) by vision. The DIRL framework comprises two components, the perception module and the control module, using imitation learning (IL) and DDPG, respectively. The perception module employs the IL network as an encoder which processes an image into a low‐dimensional feature vector. This vector is then delivered to the control module which outputs control commands. Meanwhile, the actor network of the DDPG is initialized with the trained IL network to improve exploration efficiency. In addition, a reward function for reinforcement learning is defined to improve the stability of self‐driving vehicles, especially on curves. DIRL is verified by the open racing car simulator (TORCS), and the results show that the correct control strategy is learned successfully and has less training time.Qijie ZouKang XiongQiang FangBohan JiangWileyarticleComputational linguistics. Natural language processingP98-98.5Computer softwareQA76.75-76.765ENCAAI Transactions on Intelligence Technology, Vol 6, Iss 4, Pp 493-503 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Computational linguistics. Natural language processing P98-98.5 Computer software QA76.75-76.765
spellingShingle	Computational linguistics. Natural language processing P98-98.5 Computer software QA76.75-76.765 Qijie Zou Kang Xiong Qiang Fang Bohan Jiang Deep imitation reinforcement learning for self‐driving by vision
description	Abstract Deep reinforcement learning has achieved some remarkable results in self‐driving. There is quite a lot of work to do in the area of autonomous driving with high real‐time requirements because of the inefficiency of reinforcement learning in exploring large continuous motion spaces. A deep imitation reinforcement learning (DIRL) framework is presented to learn control policies of self‐driving vehicles, which is based on a deep deterministic policy gradient algorithm (DDPG) by vision. The DIRL framework comprises two components, the perception module and the control module, using imitation learning (IL) and DDPG, respectively. The perception module employs the IL network as an encoder which processes an image into a low‐dimensional feature vector. This vector is then delivered to the control module which outputs control commands. Meanwhile, the actor network of the DDPG is initialized with the trained IL network to improve exploration efficiency. In addition, a reward function for reinforcement learning is defined to improve the stability of self‐driving vehicles, especially on curves. DIRL is verified by the open racing car simulator (TORCS), and the results show that the correct control strategy is learned successfully and has less training time.
format	article
author	Qijie Zou Kang Xiong Qiang Fang Bohan Jiang
author_facet	Qijie Zou Kang Xiong Qiang Fang Bohan Jiang
author_sort	Qijie Zou
title	Deep imitation reinforcement learning for self‐driving by vision
title_short	Deep imitation reinforcement learning for self‐driving by vision
title_full	Deep imitation reinforcement learning for self‐driving by vision
title_fullStr	Deep imitation reinforcement learning for self‐driving by vision
title_full_unstemmed	Deep imitation reinforcement learning for self‐driving by vision
title_sort	deep imitation reinforcement learning for self‐driving by vision
publisher	Wiley
publishDate	2021
url	https://doaj.org/article/fe4e98da1afc4fd195e561d3feac3d0c
work_keys_str_mv	AT qijiezou deepimitationreinforcementlearningforselfdrivingbyvision AT kangxiong deepimitationreinforcementlearningforselfdrivingbyvision AT qiangfang deepimitationreinforcementlearningforselfdrivingbyvision AT bohanjiang deepimitationreinforcementlearningforselfdrivingbyvision
_version_	1718426026831249408

Deep imitation reinforcement learning for self‐driving by vision

Ejemplares similares