Intelligent Perception and Control of Unmanned Systems 

(无人系统智能感知与控制)

Summary:

Advanced perception and control methods are important for performances such as the performances of safety, reliability, security and robustness in intelligent unmanned systems. Advances in artificial intelligence, sensor fusion, and edge computing have enabled these systems to operate in complex, dynamic environments with minimal human intervention. However, achieving reliable autonomy demands breakthroughs in accurate detection, real-time planning, adaptive control, and multi-agent coordination, particularly under uncertainties like unpredictable obstacles, communication latency, or resource constraints. This session will explore the scope of intelligent perception, autonomous planning and control through interdisciplinary lenses, covering topics such as AI-driven path optimization, reinforcement learning for adaptive behaviors, model-based control, data-driven control, swarm intelligence, human-in-the-loop control architectures, and resilience against cyber-physical threats. By uniting experts from robotics, computer science, control theory, and industry, it is significant to catalyze innovations that push the boundaries of unmanned systems, enabling them to operate seamlessly in both structured and unstructured environments while addressing societal and technical challenges. The goal is to bridge theoretical innovations with practical applications, fostering collaboration among researchers, engineers, and policymakers to develop robust and scalable solutions for intelligent unmanned systems.

先进的感知和控制方法对于智能无人系统的安全性、可靠性、安全性和鲁棒性等性能至关重要。人工智能、传感器融合和边缘计算的进步使这些系统能够在复杂、动态的环境中运行，只需最少的人为干预。然而，实现可靠的自主性需要在精确检测、实时规划、自适应控制和多智能体协调方面取得突破，特别是在不可预测的障碍、通信延迟或资源限制等不确定性下。本专题拟通过跨学科的视角探讨智能感知、自主规划和控制的范围，涵盖人工智能驱动的路径优化、自适应行为的强化学习、基于模型的控制、数据驱动的控制、群体智能、人类在环控制架构以及应对网络物理威胁的弹性等主题。通过联合来自机器人、计算机科学、控制理论和工业的专家，催化突破无人系统边界的创新具有重要意义，使它们能够在结构化和非结构化环境中无缝运行，同时应对社会和技术挑战。目标是将理论创新与实际应用联系起来，促进研究人员、工程师和政策制定者之间的合作，为智能无人系统开发稳健和可扩展的解决方案。