https://scholar.dgist.ac.kr/handle/20.500.11750/97 2026-04-04T19:42:27Z 2026-04-04T19:42:27Z Nie, Wendi Gao, Deya Liu, Chaofan Duan, Yaoxin Lee, Victor C. S. Liu, Kai Jason Xue, Chun Gui, Guan Hyuk Son, Sang https://scholar.dgist.ac.kr/handle/20.500.11750/58927 2025-12-18T02:41:12Z 2025-08-31T15:00:00Z

Title: A Multi-Agent DRL-Based Method for Cooperatively Determining Coordination and Lane-Change of Vehicles at Signal-Free Intersections With Free-Direction Lanes Author(s): Nie, Wendi; Gao, Deya; Liu, Chaofan; Duan, Yaoxin; Lee, Victor C. S.; Liu, Kai; Jason Xue, Chun; Gui, Guan; Hyuk Son, Sang Abstract: Owing to the growing population and rapid urbanization, intersections, where traffic converges from various directions, have become major bottlenecks for road capacity due to frequent congestion. Recent advances in Connected and Autonomous Vehicle (CAV) technology enable signal-free intersections, where CAVs collaborate to cross intersections without collisions. Most existing signal-free intersection control methods focus on accommodating conflicts among vehicles inside the intersection and fixed-direction lanes are commonly adopted. However, the use of fixed-direction lanes is a legacy from conventional signalized intersections, where turning lanes are predetermined and fixed, so as to direct vehicles with different turning intentions to different lanes and avoid collisions. In this paper, we aim to make full utilization of the capacity of signal-free intersections by making use of free-direction lanes, which allow vehicles to make right, straight or left turns from any lane. To this end, we propose a cooperative multi-agent Deep Reinforcement Learning (DRL)-based control method for signal-free intersections with free-direction lanes. Specifically, we first study the problem of cooperatively determining coordination of vehicles inside the intersection and lane changes of vehicles on the incoming arms. Then, a multi-agent DRL-based control method for cooperatively determining coordination and lane-change of vehicles for signal-free intersections with free-direction lanes, named CD-CLC, is proposed for maximizing non-conflicting vehicles crossing the intersection simultaneously while taking vehicle fairness into consideration, to minimize travel delays of vehicles and improve traffic efficiency. Extensive experiments have been conducted to compare CD-CLC with other state-of-the-art methods to demonstrate the effectiveness of the proposed approach. © 2014 IEEE.

2025-08-31T15:00:00Z Ding, Yi Guo, Baoshen Zheng, Lin Lu, Mingming Zhang, Desheng Wang, Shuai Son, Sang Hyuk He, Tian https://scholar.dgist.ac.kr/handle/20.500.11750/15826 2025-07-25T04:09:37Z 2021-08-31T15:00:00Z

Title: A City-Wide Crowdsourcing Delivery System with Reinforcement Learning Author(s): Ding, Yi; Guo, Baoshen; Zheng, Lin; Lu, Mingming; Zhang, Desheng; Wang, Shuai; Son, Sang Hyuk; He, Tian Abstract: The revolution of online shopping in recent years demands corresponding evolution in delivery services in urban areas. To cater to this trend, delivery by the crowd has become an alternative to the traditional delivery services thanks to the advances in ubiquitous computing. Notably, some studies use public transportation for crowdsourcing delivery, given its low-cost delivery network with millions of passengers as potential couriers. However, multiple practical impact factors are not considered in existing public-transport-based crowdsourcing delivery studies due to a lack of data and limited ubiquitous computing infrastructures in the past. In this work, we design a crowdsourcing delivery system based on public transport, considering the practical factors of time constraints, multi-hop delivery, and profits. To incorporate the impact factors, we build a reinforcement learning model to learn the optimal order dispatching strategies from massive passenger data and package data. The order dispatching problem is formulated as a sequential decision making problem for the packages routing, i.e., select the next station for the package. A delivery time estimation module is designed to accelerate the training process and provide statistical delivery time guarantee. Three months of real-world public transportation data and one month of package delivery data from an on-demand delivery platform in Shenzhen are used in the evaluation. Compared with existing crowdsourcing delivery algorithms and widely used baselines, we achieve a 40% increase in profit rates and a 29% increase in delivery rates. Comparison with other reinforcement learning algorithms shows that we can improve the profit rate and the delivery rate by 9% and 8% by using time estimation in action filtering. We share the data used in the project to the community for other researchers to validate our results and conduct further research.1 [1]. © 2021 ACM.

2021-08-31T15:00:00Z Jeong, Sang Soo Baek, Youngmi Son, Sang Hyuk https://scholar.dgist.ac.kr/handle/20.500.11750/15399 2025-07-25T04:10:02Z 2021-03-31T15:00:00Z

Title: Distributed urban platooning towards high flexibility, adaptability, and stability Author(s): Jeong, Sang Soo; Baek, Youngmi; Son, Sang Hyuk Abstract: Vehicle platooning reduces the safety distance between vehicles and the travel time of vehicles so that it leads to an increase in road capacity and to saving fuel consumption. In Europe, many projects for vehicle platooning are being actively developed, but mostly focus on truck platooning on the highway with a simpler topology than that of the urban road. When an existing vehicle platoon is applied to urban roads, many challenges are more complicated to address than highways. They include complex topology, various routes, traffic signals, intersections, frequent lane change, and communication interference depending on a higher vehicle density. To address these challenges, we propose a distributed urban platooning protocol (DUPP) that enables high mobility and maximizes flexibility for driving vehicles to conduct urban platooning in a decentralized manner. DUPP has simple procedures to perform platooning maneuvers and does not require explicit conforming for the completion of platooning maneuvers. Since DUPP mainly operates on a service channel, it does not cause negative side effects on the exchange of basic safety messages on a control channel. Moreover, DUPP does not generate any data propagation delay due to contention-based channel access since it guarantees sequential data transmission opportunities for urban platooning vehicles. Finally, to address a problem of the broadcast storm while vehicles notify detected road events, DUPP performs forwarder selection using an analytic hierarchy process. The performance of the proposed DUPP is compared with that of ENSEMBLE which is the latest European platooning project in terms of the travel time of vehicles, the lifetime of an urban platoon, the success ratio of a designed maneuver, the external cost and the periodicity of the urban platooning-related transmissions, the adaptability of an urban platoon, and the forwarder selection ratio for each vehicle. The results of the performance evaluation demonstrate that the proposed DUPP is well suited to dynamic urban environments by maintaining a vehicle platoon as stable as possible after DUPP flexibly and quickly forms a vehicle platoon without the support of a centralized node. © 2021 by the authors. Licensee MDPI, Basel, Switzerland.

2021-03-31T15:00:00Z 원명규 박태준 손상혁 https://scholar.dgist.ac.kr/handle/20.500.11750/13380 2025-07-25T04:25:45Z 2013-08-31T15:00:00Z

Title: 사이버물리시스템의 현재와 미래: 응용 어플리케이션 관점에서의 접근 Author(s): 원명규; 박태준; 손상혁 Abstract: 사이버물리시스템 (Cyber Physical Systems, CPS)은 우리가 살아가는 물리 세계와 센서, 엑츄에이터, 임베디드 컴퓨팅 시스템 등으로 구성된 사이버 세계와의 융합을 추구하는 새로운 패러다임이다. 본고에서는 CPS가 무엇인지, 왜 중요한지, 그리고 풀어야 할 숙제는 무엇인지에 대한 논의를 CPS 응용 어플리케이션의 관점에서 접근해본다. 특히 CPS 핵심 응용 분야 중 교통, 의료, 전력시스템과 관련하여 현재 대구경북과학기술원의 CPS글로벌센터에서 수행 되고 있는 연구와 그러한 연구의 기여도에 초점을 맞추어 논의를 진행한다. 즉 본고에서는 지능형 교통 시스템, 스마트 홈, 스마트 그리드 및 미래의 헬스케어 시스템에 관한 연구 소개를 통하여 CPS에 대한 실질적인 이해를 돕고 앞으로 CPS 연구와 관련하여 나아가야 할 방향에 대하여 논의한다.

2013-08-31T15:00:00Z