Imitating a Safe Human Driver Behaviour in Roundabouts Through Deep Learning
Roundabouts provide safe and fast circulation as well as many environmental advantages, but drivers adopting unsafe behaviours while circulating through them may cause safety issues, provoking accidents. In this paper we propose a way of training an autonomous vehicle in order to behave in a human a...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | Russian |
| Published: |
Belarusian National Technical University
2020-02-01
|
| Series: | Nauka i Tehnika |
| Subjects: | |
| Online Access: | https://sat.bntu.by/jour/article/view/2285 |
| _version_ | 1811185373561225216 |
|---|---|
| author | A. S. J. Cervera F. J. Alonso F. S. García A. D. Alvarez |
| author_facet | A. S. J. Cervera F. J. Alonso F. S. García A. D. Alvarez |
| author_sort | A. S. J. Cervera |
| collection | DOAJ |
| description | Roundabouts provide safe and fast circulation as well as many environmental advantages, but drivers adopting unsafe behaviours while circulating through them may cause safety issues, provoking accidents. In this paper we propose a way of training an autonomous vehicle in order to behave in a human and safe way when entering a roundabout. By placing a number of cameras in our vehicle and processing their video feeds through a series of algorithms, including Machine Learning, we can build a representation of the state of the surrounding environment. Then, we use another set of Deep Learning algorithms to analyze the data and determine the safest way of circulating through a roundabout given the current state of the environment, including nearby vehicles with their estimated positions, speeds and accelerations. By watching multiple attempts of a human entering a roundabout with both safe and unsafe behaviours, our second set of algorithms can learn to mimic the human’s good attempts and act in the same way as him, which is key to a safe implementation of autonomous vehicles. This work details the series of steps that we took, from building the representation of our environment to acting according to it in order to attain safe entry into single lane roundabouts. |
| first_indexed | 2024-04-11T13:28:28Z |
| format | Article |
| id | doaj.art-ff6996a8dd6f420fabd4006fda559aab |
| institution | Directory Open Access Journal |
| issn | 2227-1031 2414-0392 |
| language | Russian |
| last_indexed | 2024-04-11T13:28:28Z |
| publishDate | 2020-02-01 |
| publisher | Belarusian National Technical University |
| record_format | Article |
| series | Nauka i Tehnika |
| spelling | doaj.art-ff6996a8dd6f420fabd4006fda559aab2022-12-22T04:21:58ZrusBelarusian National Technical UniversityNauka i Tehnika2227-10312414-03922020-02-01191858810.21122/2227-1031-2020-19-1-85-882038Imitating a Safe Human Driver Behaviour in Roundabouts Through Deep LearningA. S. J. Cervera0F. J. Alonso1F. S. García2A. D. Alvarez3Universidad Politécnica de Madrid, ETSI de Sistemas InformáticosUniversidad Politécnica de Madrid, Escuela Técnica Superior de Ingenieros IndustrialesUniversidad Politécnica de Madrid, ETSI de Sistemas InformáticosUniversidad Politécnica de Madrid, ETSI de Sistemas InformáticosRoundabouts provide safe and fast circulation as well as many environmental advantages, but drivers adopting unsafe behaviours while circulating through them may cause safety issues, provoking accidents. In this paper we propose a way of training an autonomous vehicle in order to behave in a human and safe way when entering a roundabout. By placing a number of cameras in our vehicle and processing their video feeds through a series of algorithms, including Machine Learning, we can build a representation of the state of the surrounding environment. Then, we use another set of Deep Learning algorithms to analyze the data and determine the safest way of circulating through a roundabout given the current state of the environment, including nearby vehicles with their estimated positions, speeds and accelerations. By watching multiple attempts of a human entering a roundabout with both safe and unsafe behaviours, our second set of algorithms can learn to mimic the human’s good attempts and act in the same way as him, which is key to a safe implementation of autonomous vehicles. This work details the series of steps that we took, from building the representation of our environment to acting according to it in order to attain safe entry into single lane roundabouts.https://sat.bntu.by/jour/article/view/2285deep learningneural networksdriver behaviourroundaboutsimitation learning |
| spellingShingle | A. S. J. Cervera F. J. Alonso F. S. García A. D. Alvarez Imitating a Safe Human Driver Behaviour in Roundabouts Through Deep Learning Nauka i Tehnika deep learning neural networks driver behaviour roundabouts imitation learning |
| title | Imitating a Safe Human Driver Behaviour in Roundabouts Through Deep Learning |
| title_full | Imitating a Safe Human Driver Behaviour in Roundabouts Through Deep Learning |
| title_fullStr | Imitating a Safe Human Driver Behaviour in Roundabouts Through Deep Learning |
| title_full_unstemmed | Imitating a Safe Human Driver Behaviour in Roundabouts Through Deep Learning |
| title_short | Imitating a Safe Human Driver Behaviour in Roundabouts Through Deep Learning |
| title_sort | imitating a safe human driver behaviour in roundabouts through deep learning |
| topic | deep learning neural networks driver behaviour roundabouts imitation learning |
| url | https://sat.bntu.by/jour/article/view/2285 |
| work_keys_str_mv | AT asjcervera imitatingasafehumandriverbehaviourinroundaboutsthroughdeeplearning AT fjalonso imitatingasafehumandriverbehaviourinroundaboutsthroughdeeplearning AT fsgarcia imitatingasafehumandriverbehaviourinroundaboutsthroughdeeplearning AT adalvarez imitatingasafehumandriverbehaviourinroundaboutsthroughdeeplearning |