Special Session Ⅸ
Safe Control and Collaborative Optimization for Autonomous Swarm Confrontation (Submission Deadline: July 31, 2026)
自主集群对抗环境下的安全控制与协同优化
| Chair: | Co-chairs: | |
 |
 |
 |
| Haibo Du Hefei University of Technology, China |
Yuanqiu Mo Southeast University, China |
Lanlin Yu Hefei University of Technology, China |
| Keywords: | ||
| · Autonomous Swarms (自主集群) · Safe Control (安全控制) · Collaborative Optimization (协同优化) · Pursuit-Evasion Games (追逃博弈) |
||
| Topics: | ||
| · Control Barrier Functions (CBFs) for safe multi-agent coordination (面向多智能体安全协同的控制障碍函数) · Collaborative optimization, formation tracking, and trajectory planning for UAV swarm confrontation (面向无人机集群对抗的协同优化、编队跟踪与轨迹规划) · Safe coordination, relative navigation, and maneuver control in multi-satellite pursuit-evasion games (多卫星追逃博弈中的安全协同、相对导航与机动控制) · Continuous-time hierarchical reinforcement learning for complex tactical decision-making and pursuit (面向复杂战术决策与追逃的连续时间分层强化学习) · Differential games, zero-sum engagements, and Nash equilibrium seeking in dynamic adversarial environments (动态对抗环境下的微分博弈、零和交战与纳什均衡寻优) · Safe Multi-Agent Reinforcement Learning (MARL) with guaranteed convergence in cooperative-competitive settings (合作-竞争环境下具备收敛性保证的安全多智能体强化学习) · Distributed resource allocation, dynamic target assignment, and cooperative entrapment for perimeter defense (面向周边防御的分布式资源分配、动态目标分配与协同合围) · Resilient cooperative control and fault tolerance against cyber-attacks in tactical networks (战术网络中抵御网络攻击或干扰的弹性协同控制) |
||
| Summary: | ||
| · As multi-agent systems are increasingly deployed in contested environments, such as tactical pursuit, evasion, and cooperative defense, the demand for guaranteed operational safety alongside high collaborative efficiency has become a paramount research challenge. Traditional coordination protocols often struggle to balance aggressive task execution with stringent safety constraints during dynamic, adversarial swarm engagements. This special session addresses this critical gap by convening experts across the intersections of safety-critical control, game theory, and multi-agent reinforcement learning. Participants will explore state-of-the-art methodologies, with a distinct focus on distributed optimization algorithms and resilient decision-making frameworks. By highlighting both fundamental theoretical breakthroughs and practical deployment strategies, the session will equip attendees with the tools necessary to drive the next generation of secure, adaptive, and autonomous multi-agent coordination. |
||
|
· 在追逃与协同防御等复杂对抗环境中，确保多智能体网络系统的绝对安全性与协同效率至关重要。本专题专为深耕控制理论、自主无人系统、博弈论及多智能体强化学习领域的学者量身打造。通过参与本次会议，参会者将全面掌握安全关键控制的前沿理论、分布式优化算法及其在实际任务中的创新应用，从而共同推动安全、弹性的多智能体协同技术的跨越式发展。 |
||