다중 교차로에서 협력적 교통신호제어에 대한 연구
A Study on Cooperative Traffic Signal Control at multi-intersection
0
조회
0
다운로드
3
인용
0
참고문헌
키워드
초록
도시의 교통 혼잡 문제가 심각해지면서 지능형 교통신호제어가 활발하게 연구되고 있다. 강화학습은 교통신호제어에 가장활발하게 사용되고 있는 알고리즘으로 최근에는 심층 강화학습 알고리즘이 관심을 끌고 있다. 또한 심층 강화학습 알고리즘이 다양한 분야에서 높은 성능을 보이면서 심층 강화학습의 확장 버전들이 빠른 속도로 등장했다. 하지만 기존 교통신호제어연구들은 대부분 단일 교차로 환경에서 진행되었으며, 단일 교차로의 교통 혼잡만 완화하는 방법은 도시 전체의 교통 상황을고려하지 못한다는 한계가 있다. 본 논문에서는 다중 교차로 환경에서 협력적 교통신호제어를 제안한다. 신호제어 알고리즘에는 심층 강화학습의 확장 버전들이 결합된 알고리즘을 적용했으며 다중 교차로를 효율적으로 제어하기 위해 인접한 교차로의 교통 상황을 고려하였다. 실험에서는 제안하는 알고리즘과 기존 심층 강화학습 알고리즘을 비교하였으며, 더 나아가 협력적 방법이 적용된 모델과 적용되지 않은 모델의 실험 결과를 보여줌으로써 높은 성능을 증명한다.
Abstract
As traffic congestion in cities becomes more serious, intelligent traffic control is actively being researched. Reinforcement learning is the most actively used algorithm for traffic signal control, and recently Deep reinforcementlearning has attracted attention of researchers. Extended versions of deep reinforcement learning have been emerged asdeep reinforcement learning algorithm showed high performance in various fields. However, most of the existing trafficsignal control were studied in a single intersection environment, and there is a limitation that the method at a singleintersection does not consider the traffic conditions of the entire city. In this paper, we propose a cooperative trafficcontrol at multi-intersection environment. The traffic signal control algorithm is based on a combination of extendedversions of deep reinforcement learning and we considers traffic conditions of adjacent intersections. In the experiment,we compare the proposed algorithm with the existing deep reinforcement learning algorithm, and further demonstrate thehigh performance of our model with and without cooperative method.
참고문헌 (21)
- 김우생, "XML 문서의 구조와 내용을 고려한 유사도 측정", 멀티미디어학회논문지, 11(8), pp.1043~1050 (2008), 한국멀티미디어학회
- Chen Cai, "Adaptive traffic signal control using approximate dynamic programming", Transportation Research Part C: Emerging Technologies, 17(5), pp.456~474 (2009), Elsevier BV[DOI]
- Kartik Pandit, "Adaptive Traffic Signal Control With Vehicular Ad hoc Networks", IEEE Transactions on Vehicular Technology, 62(4), pp.1459~1471 (2013), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Maslekar, Nitin, "VANET based adaptive traffic signal control", 2011 IEEE 73rd Vehicular Technology Conference (VTC Spring). IEEE (2011)
- Mannion, Patrick, "Autonomic Road Transport Support Systems", pp.47~66 (2016), Birkhäuser
- Sutton, Richard S., "Introduction to reinforcement learning. Vol.2. No.4" (1998), MIT press
- Kaelbling, Leslie Pack, "Reinforcement learning : A survey", Journal of artificial intelligence research, 4, pp.237~285 (1996)
- Li, Li, "Traffic signal timing via deep reinforcement learning", IEEE, 2016, pp.3(3), 247~254[DOI]
- Genders, Wade, "Using a deep reinforcement learning agent for traffic signal control", arXiv preprint arXiv:1611.01142, ~ (2016)
- Gao, Juntao, "Adaptive traffic signal control: Deep reinforcement learning algorithm with experience replay and target network", arXiv preprint arXiv:1705.02755, ~ (2017)
- Volodymyr Mnih, "Human-level control through deep reinforcement learning", Nature, 518(7540), pp.529~533 (2015), Springer Science and Business Media LLC[DOI]
- 조영태, "교통 패턴 학습 기반 반복적 교통 혼잡을 고려한 교차로 신호 제어", 정보과학회 컴퓨팅의 실제 논문지, 20(8), pp.450~465 (2014), 한국정보과학회
- 김광백, "퍼지 기법을 이용한 지능형 교통 신호 제어", 한국정보통신학회논문지, 16(8), pp.1593~1598 (2012), 한국정보통신학회[DOI]
- Chen, Ruey-Shun, "ACTAM : Cooperative multi-agent system architecture for urban traffic signal control", IEICE transactions on Information and Systems, 88(1), pp.119~126 (2005)
- Van Hasselt, Hado, "Deep reinforcement learning with double q-learning", Thirtieth AAAI conference on artificial intelligence (2016)
- Wang, Ziyu, "Dueling network architectures for deep reinforcement learning", arXiv preprint arXiv:1511.06581, ~ (2015)
- Schaul, Tom, "Prioritized experience replay", arXiv preprint arXiv:1511.05952, ~ (2015)
- Richard S. Sutton, "Learning to predict by the methods of temporal differences", Machine Learning, 3(1), pp.9~44 (1988), Springer Science and Business Media LLC[DOI]
- Bellemare, Marc G., "A distributional perspective on reinforcement learning", Proceedings of the 34th International Conference on Machine Learning, 70 (2017), JMLR. org
- Fortunato, Meire, "Noisy networks for exploration", arXiv preprint arXiv:1706.10295, ~ (2017)
- Krajzewicz, Daniel, "SUMO (Simulation of Urban MObility)-an open-source traffic simulation", Proceedings of the 4th middle East Symposium on Simulation and Modelling (MESM20002) (2002)