This paper investigates different strategies for end-to-end flight software development that support having both desktop and embedded environments while minimizing the existing gap between them, in order to facilitate reiteration back and forth of the flight application. For desktop prototyping, the use of Python as a user-facing language wrapping C/C++ algorithm source code is considered. The Basilisk software testbed is presented as a specific incarnation of this desktop development proposal. For embedded development and testing, two different approaches are reviewed and demonstrated: the use of NASA’s core Flight System, which is a well-known middleware layer, and the use of MicroPython, which is a new, lean, and efficient implementation of the Python 3 programming language optimized to run on constrained environments. The migration flow of flight algorithms from the Basilisk desktop environment into each of the considered embeddable targets is described and numerical results from embedded testing are shown. While the Basilisk–core Flight System strategy is explained through the experience of its use in an actual mission, the Basilisk-MicroPython strategy is proposed as a promising and novel strategy that is still under investigation.

本文研究了端到端飞行软件开发的不同策略，这些策略支持同时具有桌面环境和嵌入式环境，同时最大程度地减小了桌面环境和嵌入式环境之间的距离，以方便反复地反复进行飞行应用程序的开发。对于桌面原型，考虑使用Python作为包装C / C ++算法源代码的面向用户的语言。Basilisk软件测试平台作为此桌面开发建议的具体体现而呈现。对于嵌入式开发和测试，审查并演示了两种不同的方法：使用NASA的核心Flight System（这是一个著名的中间件层）和使用MicroPython（这是一种新的，精简且高效的Python实现） 3种编程语言经过优化，可在受限环境中运行。描述了飞行算法从Basilisk桌面环境到每个考虑的可嵌入目标的迁移流程，并显示了嵌入式测试的数值结果。虽然通过在实际任务中使用蛇怪-核心飞行系统策略进行了解释，但蛇怪-MicroPython策略被认为是一种有前途且新颖的策略，目前仍在研究中。