Enabling robotic autonomy in challenging environments; an algorithmic approach

在充满挑战的环境中实现机器人自主；

• 批准号: 356377-2008
• 负责人: Rekleitis, Ioannis
• 金额: $ 1.09万
• 依托单位: McGill University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2011
• 资助国家: 加拿大
• 起止时间: 2011-01-01 至 2012-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=483078
• 关键词: Enabling; robotic; autonomy; challenging; environments

My work is centred on deriving algorithmic solutions that enhance an agent's autonomy through intelligent decision-making, collaboration and information exchange with the other members of the computational group. The agents may be nodes in a sensor network, mobile robots, or robotic spacecraft. Autonomy capabilities are especially important when robots operate in challenging environments such as underwater or in space. In particular both planetary exploration and underwater robotics suffer from the same limitations as communication with a human operator is intermittent and with low bandwidth. In order to facilitate autonomy any robotic system must address the following fundamental problems: estimate their state with respect to the surrounding environment; model the world around them; and plan their tasks in an optimal manner. In the proposed research I am planning to extend my work on state estimation for robots moving in three dimensions, using a variety of techniques such as Particle filters and Extended Kalman Filters. I am also planning to develop new techniques for environment modelling using irregular triangular meshes. The previous two components are usually termed together as Simultaneous Localization and Mapping (SLAM) in the robotics community. Moreover, it has been demonstrated in my earlier work, as well as, by other researchers that collaboration between intelligent agents greatly enhances their capabilities. Therefore in my future plans is the extension of single robot planetary exploration algorithms to multiple robots.

我的工作集中在推导算法的解决方案，通过智能决策，协作和信息交流与计算组的其他成员，提高代理的自主性。代理可以是传感器网络中的节点、移动的机器人或机器人航天器。当机器人在水下或太空等具有挑战性的环境中工作时，自主能力尤为重要。特别是行星探索和水下机器人都受到同样的限制，因为与人类操作员的通信是间歇性的，带宽很低。为了促进自主性，任何机器人系统都必须解决以下基本问题：估计它们相对于周围环境的状态;对它们周围的世界进行建模;以及以最佳方式规划它们的任务。在拟议的研究中，我计划扩展我的工作状态估计机器人在三维移动，使用各种技术，如粒子滤波器和扩展卡尔曼滤波器。我还计划开发使用不规则三角形网格进行环境建模的新技术。前两个组件通常被称为机器人社区中的同步定位和映射（SLAM）。此外，在我早期的工作中已经证明，以及其他研究人员，智能代理之间的合作大大提高了他们的能力。 因此，我未来的计划是将单机器人行星探索算法扩展到多机器人。