HIT academician Xibin Cao's research team: Dynamic verification of space missions via flexible model-based co-simulation with systems modeling language and spacesim

Space missions are complex, multidisciplinary tasks that involve high risk and high cost. Systems engineering (SE) technology is an emerging discipline used to manage project complexity and ensure mission success . As technology advances and systems become more complex and volatile, SE needs to accommodate the constant reassessment, upgrading, and development of systems . Traditional SE relies on a large number of decentralized documents that cannot keep up with the changes in the system. Therefore, SE is transforming toward digitalization and has led to model-based systems engineering (MBSE), which provides a way to address SE challenges and is emerging as a paradigm and principle of SE. According to the International Council on Systems Engineering, MBSE is “the formalized application of modeling to support system requirements, design, analysis, verification and validation activities beginning in the conceptual design phase and continuing throughout development and later life cycle phase”.

In a recently published study in Space: Science & Technology, the research team led by Academician Xibin Cao at Harbin Institute of Technology proposed a flexible co-simulation framework. This framework integrates the Systems Modeling Language (SysML) with the independently developed orbital analysis tool SpaceSim, establishing an executable, traceable, and rapidly iterative dynamic verification environment for spacecraft systems during the conceptual design phase.

First of all, this study proposes a flexible space system analysis framework that enables design and verification capabilities during the early stages of space system development. This framework fully leverages the MBSE environment by tightly integrating system architecture—encompassing requirements, structure, and behavior—with co-simulation to achieve dynamic system verification. In this study, the satellite system is considered the system of interest (SoI). Through the integration and standardization of a co-simulation meta-model within the MBSE environment, interoperability between SysML models and the SpaceSim tool is established, facilitating the generation of the satellite’s operational timeline. Based on this, an executable architecture model of the satellite system is developed, which uses the computational results from SpaceSim to drive system operations. This enables the dynamic verification of communication links and energy balance in a virtual environment, thereby allowing for the evaluation of overall system performance.

The satellite system dynamic simulation presented here focuses on synchronized simulation of energy flow and information flow within a state-driven framework, where the satellite transitions between states and executes the corresponding operational mode workflows. To support this process, a co-simulation first produces the satellite’s state sequence, and the architecturelevel dynamic simulation then computes the time evolution of critical system parameters. The overall workflow and interfaces are shown in Fig. 1. The analysis environment comprises 2 components, an MBSE tool (MagicDraw 2022 in this study) and the orbital-domain analysis tool SpaceSim. SysML models maintained in the MBSE environment capture all system and simulation definitions, including the satellite system architecture, the co-simulation architecture model, a graphical user interface (GUI) for data display and simulation control, and the required data and communication interfaces.

In addition, the proposed approach incorporates hierarchical and modular principles during the architecture development phase to enhance the flexibility of system analysis. Since the complete SysML model encompasses both system architecture and co-simulation components, the method is abstracted into a structured system analysis process supported by 2 meta-models: the SoI meta-model (Fig. 2) and the SpaceSim co-simulation meta-model (Fig. 3). These meta-models adopt distinct strategies to respectively enhance the flexibility of architectural dynamic verification and co-simulation. A meta-model defines the elements and relationships reflected in the instantiated model, including object types, their attributes, and the rules governing their composition. This meta-model-based approach addresses the limited domain specificity of SysML in applications such as spacecraft systems. By instantiating the meta-models, system engineers can obtain architecture models tailored to specific engineering problems, enabling higher-fidelity verification at earlier stages and supporting rapid iterations during the evolving conceptual design process—ultimately accelerating the overall design cycle.

Subsequently, the authors demonstrated the proposed methodology through a simulated space mission and its modification scenarios, illustrating how it facilitates system analysis, defines mission objectives, and validates requirements, while highlighting the flexibility of system model updates.

This study uses a space remote sensing mission as a case study. The mission objective was to develop a satellite equipped with a hyperspectral camera capable of remotely sensing major shipping routes to enhance navigation safety. Additionally, the satellite needed to transmit remote sensing data to ground stations as it passed over them. The expected satellite lifetime was 5 years. Following a preliminary mission analysis, the orbital engineers selected a 700-km-high Sun-synchronous orbit for the mission. Given the use of the MBSE approach, defining a satellite system architecture model became essential. The goal was to verify the daily balance of energy and information flows. The results are presented in Fig. 4 . Specifically, Fig. 4 A and B show the satellite’s operational mode and solar visibility timelines, respectively, while Fig. 4 C and D depict data storage and battery power curves, respectively. The results demonstrate that the satellite maintains energy balance throughout a 24-h period (approximately 14.6 orbital cycles), with power levels remaining above the DoD threshold. Additionally, the satellite successfully downlinks data to ground stations in a timely manner.

On this basis, the present study adopts flexibility as a fundamental architectural principle and demonstrates the effectiveness of the proposed method by analyzing the steps involved in model modification and the corresponding time consumption.

(1) In case A (Fig. 5), the SpaceSim meta-model captures essential knowledge for using the analysis tool, and its hierarchical and functionally modular architecture substantially reduces the difficulty of modifying models. Shared components within the meta-model and the simulation architecture allow system engineers—regardless of familiarity with SpaceSim—to easily define or modify co-simulation scenarios. Compared to traditional approaches, SysML models representing system design can be directly transformed into analysis models without the need for redundant manual redefinition.

(2) Case B (Fig. 6) involves modifications to the behavioral aspects of the system architecture. As with conventional methods, system engineers edit system behavior at the logical level. However, the proposed method ensures the executability of the overall architecture by encapsulating executable behaviors at the computational layer with well-defined interfaces. In contrast, approaches that do not follow modeling guidelines often mix system operation and simulation logic, resulting in behavior models that are difficult to understand and maintain. A hierarchical approach to behavior modeling addresses this problem by clearly separating layers, thereby reducing the complexity of behavior updates.

(3) Case C (Fig. 7) examines changes to system design parameters. While modifying parameters in MBSE is relatively straightforward, such modifications occur frequently throughout system development. Compared to non-MBSE approaches, the proposed method enables changes to propagate across the entire model and allows direct verification through simulation, eliminating the need for manually updating dispersed parameter code. The case study also underscores the rationale for decoupling SpaceSim simulations from spacecraft system simulations. Joint simulation is primarily used to compute satellite operation timelines, whereas system architectures typically undergo several revisions before converging on a feasible solution. If the 2 simulations were tightly coupled, each flow balance analysis would require re-execution of the full collaborative simulation, dramatically decreasing analysis efficiency.

Finally, the authors summarize the overall contribution of this study. By defining a collaborative simulation domain model in SysML, the research achieves standardized descriptions of system design models, simulation scenarios, and execution logic, enabling analysis models to be automatically generated based on design models. Consequently, system engineers no longer need to manually define SpaceSim commands or rewrite scenario files when building simulation tasks, significantly reducing modeling costs and professional barriers. The modular and hierarchical structure of the metamodel ensures flexibility in the modeling process, allowing modifications to scenarios, behaviors, and parameters at minimal cost. Meanwhile, the collaborative simulation framework enables automatic communication and result feedback between multiple software systems, enhancing continuity between design, simulation, and verification. Overall, this paper not only proposes an operational MBSE simulation framework but also lays the foundation for digital modeling, simulation automation, and the development of future digital twin platforms in spacecraft systems.