Session
Technical Poster Session 4
Location
Utah State University, Logan, UT
Abstract
The modern space mission landscape requires consideration of many trade variables to meet growing cost and performance constraints. In the defense sector, threats continue to proliferate, driving the need for multi-layered constellations that are robust to myriad potential conflicts. In the civil sector, stakeholders seek science data with ever increasing timeliness, coverage, and resolution to monitor the pulse of Earth's changing environment. Across the commercial arena, companies compete to deliver the best space products and services, bringing cutting edge business opportunities into the realm of possibility for the first time. Now more than ever, all three sectors demand optimal mission performance at low cost.
For years academic methods have existed to address such challenges through complex system-of-systems analyses. However, widespread practical application of these methods to multi-satellite mission design remains uncommon. It is in the best interest of industry to evaluate, adopt, deploy, and standardize these methods as they have the capability to deliver unbiased, quick-turn constellation assessments, improving responsiveness to customer investigations of next-generation space missions of all types. Such capabilities lay the groundwork for rapid discovery of highly capable, low cost, future space architectures.
This paper presents an approach to identify promising smallsat architectures through integration of parameterized engineering and cost estimation models. This approach simultaneously explores design drivers at multiple levels of mission architecture including payload, bus, orbit, and launch vehicle by employing proven statistical, data science, and machine learning techniques. When deployed at the early stages of constellation development, this analysis approach delivers two main benefits: It informs stakeholders of mission performance and cost sensitivity to a variety of design variables, leading to better decision making earlier in the acquisition timeline; and it uncovers promising regions of large design spaces to be examined further by teams of experts, increasing the efficiency of engineering design cycles.
SSC23-P4-29 Poster
Surrogate-Based SmallSat Architecture Design Leveraging Integrated Parametric Mission Models
Utah State University, Logan, UT
The modern space mission landscape requires consideration of many trade variables to meet growing cost and performance constraints. In the defense sector, threats continue to proliferate, driving the need for multi-layered constellations that are robust to myriad potential conflicts. In the civil sector, stakeholders seek science data with ever increasing timeliness, coverage, and resolution to monitor the pulse of Earth's changing environment. Across the commercial arena, companies compete to deliver the best space products and services, bringing cutting edge business opportunities into the realm of possibility for the first time. Now more than ever, all three sectors demand optimal mission performance at low cost.
For years academic methods have existed to address such challenges through complex system-of-systems analyses. However, widespread practical application of these methods to multi-satellite mission design remains uncommon. It is in the best interest of industry to evaluate, adopt, deploy, and standardize these methods as they have the capability to deliver unbiased, quick-turn constellation assessments, improving responsiveness to customer investigations of next-generation space missions of all types. Such capabilities lay the groundwork for rapid discovery of highly capable, low cost, future space architectures.
This paper presents an approach to identify promising smallsat architectures through integration of parameterized engineering and cost estimation models. This approach simultaneously explores design drivers at multiple levels of mission architecture including payload, bus, orbit, and launch vehicle by employing proven statistical, data science, and machine learning techniques. When deployed at the early stages of constellation development, this analysis approach delivers two main benefits: It informs stakeholders of mission performance and cost sensitivity to a variety of design variables, leading to better decision making earlier in the acquisition timeline; and it uncovers promising regions of large design spaces to be examined further by teams of experts, increasing the efficiency of engineering design cycles.
