SGER: Dynamic Fault Tolerance Methods for Robotics

本研究在探索性研究小额拨款模式下，对机器人容错领域进行了研究。这一活动产生了一系列具有潜在价值的个人风气。该项目的目标是开发一种严格的机器人容错方法的基础。该研究涉及为机器人合成保护性操作系统外壳环境，并将该环境纳入可证明的容错机器人系统的新方法中。操作系统外壳的设计是为了保护机器人免受下级机械手系统和上级计划中错误的影响。该外壳由多个处于离散决策层的容错任务组成。在探索工作的一年中，将为三连杆平面机器人确定并概述这些任务。该机器人将在每个关节上增强两个传感器，并在运动学上冗余地执行飞机上的定位任务。这种级别的硬件冗余将为机器人提供容错能力。将为这个简单的例子机器人开发一个正式的模型，并用来展示这种研究方法的潜在好处。这项工作将为验证机器人系统的低级和高级子系统的容错性提供形式化的方法。关键的总体目标是将机器人容错的各种技术结合到一种可证明是正确的方法中。这种方法的形式化工具的开发对于证明关于机器人容错的严格结果将是重要的。在SGER项目的一年中，一系列基本的例子将展示上述方法如何显著影响这一新兴研究领域。

This research, in the Small Grants for Exploratory Research mode, investigates the area of fault tolerance for robotics. This activity has produced a series of potentially valuable individual ethods. The goal of this project is to develop the foundations for a rigorous approach to robotic fault tolerance. The research involves synthesis of a protective operating system shell environment for robots, and incorporation of this environment into a new approach for provably fault tolerant robotic systems. The operating system shell is designed to protect the robot from the effects of errors in the lower level manipulator system and the higher level plan. This shell is composed of a number of fault tolerance tasks at a discrete decision making level. These tasks will be identified and outlined for a three link planar robot during the year of exploratory work. The robot will be enhanced with two sensors per joint and will be kinematically redundant for positioning tasks in the plane. This level of hardware redundancy will provide fault tolerance capabilities to the robot. A formal model will be developed for this simple example robot and used to demonstrate the potential benefits of the research approach. This work will provide formal methods to verify fault tolerance for lower and higher level subsystems of a robot system. The key overall goal is to combine diverse techniques for robotic fault tolerance into a provably correct methodology. The development of formal tools for the approach will be important in proving rigorous results about robot fault tolerance. In the one year of the SGER project, a series of fundamental examples will demonstrate how the above approach can significantly impact this emerging research area.