Intelligent guidance, navigation and control in autonomous and aerospace robotic systems

自主系统技术的快速发展创造了对新的传感器/数据融合和智能控制技术的不断需求。在快速增长的应用程序中，需要从多个来源集成信息的有效方法。在过去十年中，机器人技术的广泛研究表明，使用先进的融合算法集成测量和数据可以显著提高自主系统的性能。特别是，无人驾驶飞机和地面车辆的制导、导航和控制起着关键作用。这对柔性机械手也很重要。本提案基于申请人早期在传感器信息融合方法及其在动态系统导航和控制中的应用方面的工作。提出的研究旨在整合大量的数据和测量，以实现有效的定位，定位和绘图。传感器和数据融合算法将使用传感器数据关联技术，统计方法，粒子滤波和估计技术进行研究和开发。本文将致力于发展基于扩展卡尔曼滤波、集成滤波和自适应卡尔曼滤波的先进方法。这些技术将允许对带有彩色噪声的非线性系统进行大量数据融合，例如将图像与全球定位系统、惯性导航系统和地图进行融合。

Rapid advances in autonomous systems technologies create constant demand for new sensor/data fusion and intelligent control techniques. Efficient methods of information integration from several sources are needed in rapidly growing number of applications. Extensive research in robotics in the last decade demonstrated that integration of measurements and data using advanced fusion algorithms lead to significant improvement in performance of autonomous systems. In particular, guidance, navigation, and control of unmanned aerial and ground vehicles play a key role. It is also important for flexible manipulators. This proposal builds on applicant's earlier work on sensor information fusion methods and its application to navigation and control of dynamic systems. The proposed research is aimed at integration of large amount of data and measurements to allow efficient localization, positioning and mapping. Sensor and data fusion algorithms will be investigated and developed using sensor data association techniques, statistical methods, particle filtering and estimation techniques. The main effort will be devoted to the development of advanced methods based on extended, ensemble and adaptive Kalman filter. These techniques would allow fusion of large amount of data for nonlinear systems with coloured noise, e.g. fusion of images with global positioning systems, inertial navigation systems and mapping.