SBIR Phase I: Optimizing Safety and Fuel Efficiency in Autonomous Rendezvous and Proximity Operations (RPO) of Uncooperative Objects

这个小企业创新研究（SBIR）第一阶段项目将实现一种新型的空间接近操作。这项研究不仅具有维持和改善太空行动的潜力，而且还具有加强国家安全并促进太空经济繁荣的潜力。预期的进展包括卫星检查、维修/升级、寿命终止服务、碎片修复甚至制造和组装操作的复杂能力。该项目强调任务的安全性、稳健性和自主性，最终也为更安全的人类航天飞行操作铺平了道路，并有助于减缓碎片和避免碰撞等重要领域。这一努力还延伸到探索前沿技术，如小行星采矿。该项目的方法创造了商业机会，解锁了在轨服务、装配和制造价值链。SBIR第一阶段项目将合成神经Lyapunov函数，该函数可以集成到任何类型的控制系统的滤波方案中，接受来自多传感器测量的状态反馈。本研究的主要目的是使检测和捕获不合作、不受控制和未准备的物体成为可能。这种能力是通过融合来自多个传感器的数据并应用基于Neural Lyapunov理论的屏障函数来实现的，以确保在对接和“组合堆栈”阶段（即服务端与客户端航天器对接时）在驱动限制和状态约束下的安全性。此外，该技术还开发了路径规划算法，使用实时光学测量来计算客户卫星的坠落率，确保更安全的检查和对接操作。这些步骤对于确保在对接阶段和组合堆栈机动期间的安全自主操作至关重要。本研究的最终结果是开发一种任务设计、分析和规划工具，以帮助运营商考虑涉及在轨近距离操作的不同任务场景，同时分析安全保证与燃料优化之间的权衡。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 Small Business Innovation Research (SBIR) Phase I project will enable a novel class of in-space proximity operations. This research has the potential not only to sustain and improve space operations, but also to strengthen national security and result in a thriving economy in space. The expected advances include sophisticated capabilities for satellite inspection, repair/upgrade, end-of-life servicing, debris remediation, and even manufacturing and assembly operations. This project's emphasis on the safety, robustness, and autonomy of missions also ultimately paves the way for safer human spaceflight operations and contributes to vital areas like debris mitigation and collision avoidance. This effort also extends to the exploration of frontier technologies such as asteroid mining. This project's approach creates commercial opportunities and unlocks the in-orbit servicing, assembly, and manufacturing value chain.This SBIR Phase I project will synthesize Neural Lyapunov functions, which can be integrated into filter schemes for any type of control system that accepts state feedback from multi-sensor measurements. The primary objective of this study is to enable the inspection and capture of uncooperative, uncontrolled, and unprepared objects. This ability is achieved by fusing data from multiple sensors and applying barrier functions, rooted in Neural Lyapunov theory, to ensure safety within actuation limits and state constraints during the docking and 'combined stack' phases (i.e., when a servicer is docked with a client spacecraft). Furthermore, this technology developed path planning algorithms that use real-time optical measurements to account for the detumbling rates of client satellites, ensuring safer inspection and docking maneuvers. These steps are critical for ensuring safe autonomous operations during the docking phase and combined stack maneuvers. The final outcome of this research is to develop a mission design, analysis, and planning tool to help operators account for different mission scenarios involving in-orbit proximity operations, while analyzing tradeoffs of safety assurance versus fuel optimization.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.