Embedded Files
Deep Reinforcement Learning-based Intelligent Mobility
Deep Reinforcement Learning-based Intelligent Mobility
Fig1. Deep Reinforcement Learning-based Intelligent Mobility
[Description]
We develop deep reinforcement learning (MAPPO, Safe RL) methods to optimize decision-making policies for cooperative perception message redundancy mitigation, communication-aware longitudinal platoon control, and hybrid energy management, under practical constraints such as communication quality, safety, and energy efficiency.
[Achievement]
[Int'l Conf.] K. Park and H.-S. Jo, "Multi-Agent Proximal Policy Optimization Based Redundancy Mitigation Rule for C-V2X Collective Perception," Sixteenth International Conference on Ubiquitous and Future Networks (ICUFN), Lisbon, Portugal, Jul. 2025.
[Domestic Conf.] S. Kim, K. Park, and H.-S. Jo, "Reinforcement Learning-Based Longitudinal Control for Platoon Follower Vehicles Using an Integrated Co-Simulation Platform," Proc. Korean Society of Automotive Engineers (KSAE) Fall Conference and Exhibition, Busan, South Korea, Nov. 2025.
[Domestic Conf.] H. J. Kim, K. Park, and H.-S. Jo, "Development of an Optimal Torque Distribution Map for Hybrid Electric Vehicles Based on Operating Modes Using ECMS and DDPG," Proc. Korean Society of Automotive Engineers (KSAE) Fall Conference and Exhibition, Busan, South Korea, Nov. 2025.
Connected and Autonomous Driving:
Deep Reinforcement Learning Empowered Congestion Control
Connected and Autonomous Driving:
Deep Reinforcement Learning Empowered Congestion Control
Fig2. Deep Reinforcement Learning Empowered Congestion Control
[Description]
We develop a deep reinforcement learning–empowered decentralized congestion control (DCC) scheme for C-V2X to mitigate channel congestion and improve reliability in dense urban intersections. Building on cooperative multi-agent DRL (QMIX), each road-zone RSU acts as an agent and autonomously adapts the packet generation interval using local congestion indicators (e.g., vehicle density, CBR, LOS) to meet a target QoS.
[Achievement]
[SCIE] W. Yang and H.-S. Jo, "Deep-Reinforcement-Learning-Based Range-Adaptive Distributed Power Control for Cellular-V2X," ICT Express, Jul. 2022.
[SCIE] W. Yang, T. Nam, B. Jeon, C. Mun, N. Jeong, and H.-S. Jo, "Deep Reinforcement Learning-based Cellular-V2X QoS Adaptive Distributed Congestion Control," IEEE Transactions on Intelligent Vehicles, Feb. 2023.
[KCI] B. Jeon, W. Yang, and H.-S. Jo, "Deep Reinforcement Learning-Based C-V2X Distributed Congestion Control for Real-Time Vehicle Density Response," Journal of the Institute of Korean Electrical and Electronics Engineers, vol. 27, no. 4, pp. 12–18, Dec. 2023.
[Int'l Conf.] S. Nam, T. Nam, and H.-S. Jo, "Multi-Agent Deep Reinforcement Learning for QoS-Adaptive Decentralized Congestion Control in C-V2X Networks," The 16th International Conference on ICT Convergence (ICTC), Jeju Island, Korea, Oct. 2025.
Google Sites
Report abuse