面向救援任务的异构多机器人混杂系统的协同优化控制

项目名称： 面向救援任务的异构多机器人混杂系统的协同优化控制

项目编号： No.60875072

项目类型： 面上项目

立项/批准年度： 2009

项目学科： 金属学与金属工艺

项目作者： 秦世引

作者单位： 北京航空航天大学

项目金额： 34万元

中文摘要： 本项目针对灾难救援的实际应用背景和需求，设计了地空异构多机器人联合救援系统。进而采用混杂系统理论同智能自主控制相结合的方法，选择在救援过程中对任务协作和行为协调具有决定性意义的协同优化控制问题为主攻方向，重点研究了地空异构多机器人混杂系统的综合集成建模；复杂环境下地空异构多机器人系统的环境感知、目标检测、识别与跟踪；基于流形学习的多机器人同步定位于地图构建；基于视觉选择性注意的环境感知与目标识别；面向运动成像的视频稳像算法和高性能模糊图像恢复技术；面向救援任务的多机器人路径规划与智能自主导航；Ad hoc网络环境下地空异构多机器人移动通信系统的优化设计与实验验证；面向救援任务的异构多机器人编队过程协调优化控制；非连续通信条件下面向分布一致性的多无人机协同优化控制；无人机编队过程的优化控制；地空异构多机器人混杂系统智能协同优化控制的模拟实验等方面的关键性命题。在理论方法和应用实现技术等方面取得了一系列重要研究成果。这对丰富和发展大规模异构多机器人混杂系统的控制理论体系，推动我国的机器人勘察/救援事业的发展进程具有重要意义。

中文关键词： 异构多机器人；灾难救援；环境感知；目标识别；协同控制

英文摘要： According to practical applied background and technical requirements in the salvation of disasters, the organization framework of heterogeneous multi-robots rescue system with ground-and-air cooperation is designed，and then the methodology of combining hybrid system theory with intelligent autonomous controls is employed to aim the collaborative optimizing controls of task collaboration and behavior coordination during the rescue course as the key research direction which is provided with decisive significance for a successful salvation. Thus, those corresponding topics and propositions are researched in depth to achieve the consummation for this project, which may be listed as follows: synthetically integrative modeling of hybrid system with ground-and-air heterogeneous multi-robots; environment perception, target detection/recognition and tracking for heterogeneous multi-robots under complex circumstances; multi-robot simultaneous localization and mapping based on manifold learning; environment perception and target recognition based on visual selective attention; video stabilization and high performance restoration algorithms of blurring images for motional imaging process; path planning and intelligent autonomous navigation of multi-robot toward rescue tasks; optimizing design and test of ad hoc routing protocol for mobile communication of multi-robots; collaborative optimizing control of formation for heterogeneous multi-robots toward rescue tasks; cooperative control of multi-UAVs towards distributed consensus under switching communication; optimizing controls of formation process for UAVs; comprehensive simulation experiments for intelligent collaborative optimizing controls of hybrid system with ground-and-air heterogeneous multi-robots. A series of important research results have been achieved in both theoretical methods and applied implementing technologies for above propositions. The research achievements will develop and enrich the control theory of large scale hybrid system of heterogeneous multi-robots so as to promote the development progress of robot reconnnaissance and rescue technologies on our contry.

英文关键词： heterogeneous multi-robots; salvation of disaster; environment perception; target recognition; collaborative controls