Embodied Intelligent Systems (具身智能系统)
Multimodal Perception (多模态感知)
Intelligent Decision Control (智能决策控制)
LLM Control (大模型控制)

Multimodal environmental perception for embodied intelligent systems (具身智能系统的多模态环境感知)
Cross-modal understanding and multi-source data fusion methods (跨模态信息理解与多源数据融合方法)
Spatiotemporal data fusion and state estimation (时空数据融合与状态估计)
Perception-driven intelligent control approaches (感知驱动的智能控制方法)
Perception–control coordination in embodied intelligent systems (具身智能系统的感知-控制协同机制)
Intelligent decision-making and control under uncertainty (不确定环境下的智能决策与控制)
Multi-sensor fusion and control for robotic systems (机器人系统的多传感器融合与控制)
Deep learning methods for perception-control integration (基于深度学习的感知与控制融合方法)
Large Language Model-Driven Intelligent Control (大模型语言驱动的智能控制)

With the rapid development of embodied intelligent systems in autonomous platforms and intelligent interaction, perception and control in complex environments have attracted increasing attention. In dynamic and open environments, single-source sensing information is often insufficient for reliable environment understanding and decision-making, making multimodal perception and multi-source data fusion essential. However, heterogeneous sensing data still face challenges in spatiotemporal consistency, uncertainty handling, and real-time fusion for control decisions. Therefore, this special session aims to bring together recent advances in multimodal perception, data fusion, and intelligent control for embodied intelligent systems, promoting perception-driven control theories and applications in robotics and autonomous platforms.

随着具身智能系统在自主平台及智能交互等领域的快速发展，复杂环境下的智能感知与控制问题日益受到关注。在动态和开放环境中，单一传感信息难以支撑高可靠性的环境理解与决策控制，多模态感知与多源数据融合成为提升系统智能水平的重要技术途径。然而，不同传感器数据在时空一致性、信息不确定性以及实时融合与控制决策方面仍面临诸多挑战。因此，本专题旨在汇集具身智能系统多模态感知、数据融合与智能控制领域的最新研究成果，促进感知驱动控制理论与方法的发展，并推动其在智能机器人与自主系统中的应用。