VA-FastNavi-MARL: Real-Time Robot Control with Multimedia-Driven Meta-Reinforcement Learning
VA-FastNavi-MARL:基于多媒体驱动的元强化学习的实时机器人控制
机构 * School of Artificial Intelligence, Hubei University(湖北大学人工智能学院) ; Department of Mechanical and Aerospace Engineering, University of Missouri(密苏里大学机械与航空航天工程系) ; School of Construction Machinery, Chang’an University(长安大学工程机械学院) ; Key Laboratory of Intelligent Sensing System and Security (Ministry of Education), Hubei University(湖北大学智能感知系统与安全教育部重点实验室)
专题命中 音频语音多模态 :audio-visual(abstract)
AI总结 本文提出VA-FastNavi-MARL框架,通过元强化学习将异步音频视觉输入统一为潜在表示,实现快速适应未知指令,提升实时控制性能。
Comments Accepted to the 2026 IEEE International Conference on Multimedia and Expo (ICME 2026)
Journal ref 2026 IEEE International Conference on Multimedia and Expo (ICME)