CAREER: Topological Planning in Information Space for Intelligent Robotic Systems

该项目设想移动机器人系统能够检索世界上任何物理位置的流信息，以造福科学家、急救人员和公众。在过去的十年中，许多海洋和空中系统在环境监测，搜救，监视和科学探索等应用中出现。然而，目前操作员要么直接远程操作这些车辆，要么在固定计划中预先指定路点位置。这些方法通过限制机器人车辆的能力(1)随着环境、任务和队友的变化继续有效地运行，以及(2)纳入人类操作员给出的高水平目标，导致严重限制了性能。该项目旨在弥合当前系统与智能系统之间的差距，这些系统天真地遵循路径点，能够适应环境、团队能力和操作人员的目标。该项目的最终目标是开发基于同伦和持久同伦的拓扑规划技术，该技术可以对环境条件（例如，信息区域，干扰区域和风险区域）的紧凑语义表示进行推理。通过对简洁的语义图进行推理，所提出的方法将使自动驾驶汽车团队能够在物理世界中就信息收集、成本最小化和目标实现做出决策。开发的算法将通过开源分发向公众提供，并通过与海洋科学家的持续合作过渡到实地测试。该项目的教育部分将通过海洋机器人证书课程将机器人车辆规划整合到俄勒冈州立大学的研究生课程中，并通过俄勒冈州立大学本科机器人俱乐部设计的水下机器人编程模块整合到俄勒冈州立高中。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This project envisions mobile robotic systems capable of retrieving streaming information about any physical location in the world for the benefit of scientists, first responders, and the general public. In the past decade, a number of marine and aerial systems have emerged in applications such as environmental monitoring, search and rescue, surveillance, and scientific exploration. However, currently an operator either teleoperates these vehicles directly or pre-specifies waypoint locations in a fixed plan. These approaches lead to severely limited performance by restricting the capabilities of robotic vehicles to (1) continue operating effectively as their environment, mission, and teammates change and (2) incorporate high-level goals given by human operators. This project seeks to bridge the gap between current systems that naively follow waypoints and intelligent systems capable of adapting to their environment, their team's capabilities, and to the goals of the operator. The ultimate objective of this project is to develop topological planning techniques based on homotopy and persistent homology that reason over a compact semantic representation of environmental conditions (e.g., informative areas, disturbances, and risky areas). By reasoning over concise semantic maps, the proposed methods will enable teams of autonomous vehicles to make decisions regarding information collection, cost minimization, and goal achievement in the physical world. The developed algorithms will be made publicly available through open source distribution and transitioned to field tests through ongoing collaborations with ocean scientists. The educational component of this project will integrate robotic vehicle planning into the graduate curriculum at Oregon State University through a Marine Robotics Certificate program and into Oregon high schools through an underwater robot programming module designed by the Oregon State University undergraduate Robotics Club.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.