Introduction
🔥 What is Basilisk?
Basilisk (BSK) is a software framework for astrodynamics simulation, designed to generate high-fidelity spacecraft simulations. It enables the modeling of orbital maintenance and maneuvering in deep space missions, covering not only orbital and attitude dynamics but also subsystems such as reaction wheels and thrusters. This capability allows Monte Carlo simulations of repeatable spacecraft missions and further supports real-time hardware-in-the-loop (HIL) verification.
The software was jointly developed by the Autonomous Vehicle Simulation (AVS) Laboratory and the Laboratory for Atmospheric and Space Physics (LASP) at the University of Colorado. Its FSW core is implemented in C/C++ and made accessible through a Python wrapper for ease of use. In addition, Basilisk provides a visualization tool called Vizard, based on the Unity rendering engine, which enables interactive visualization of simulation states.
💡 Utilization
The proposed applications of Basilisk are as follows:
-
astrodynamics research to model complex spacecraft dynamical behaviors
-
developing new guidance, estimation and control solutions
-
supporting mission concept development
-
supporting flight software development
-
supporting hardware in the loop testing by simulating in realtime the spacecraft states
-
analysis of flight data and compare against expected behavior
-
supporting spacecraft AI based autonomy development
🌏 Associated Software Tool
The following are other software tools commonly used in the fields of orbital mechanics and astrodynamics.
STK (Systems Tool Kit) is widely regarded as an industry-standard tool for mission analysis and simulation, capable of modeling a wide range of aerospace systems, but its high cost makes it difficult for individuals or academic research groups to access. In contrast, GMAT (General Mission Analysis Tool), developed by NASA, is a free and open-source software package that is broadly used for orbit propagation and mission design. Another representative tool, Orekit, is a Java-based open-source astrodynamics library with strong capabilities in precise calculations such as coordinate and time transformations. Finally, CesiumJS is a web-based 3D visualization engine specialized in rendering orbital and space data in an intuitive and interactive manner.
💘 Why Basilisk?
Unlike these tools (such as GMAT, Orekit), Basilisk distinguishes itself as a control system simulator. It not only models orbital and attitude dynamics but also integrates subsystems such as reaction wheels, thrusters, and sensors, providing a more comprehensive spacecraft simulation environment. Moreover, through its Unity-based visualization tool Vizard, Basilisk enables interactive visualization of simulation states, offering a unique advantage for both research and development in spacecraft dynamics and control.
🚀 Objective: RL-Based Autonomous Orbital Maneuvering
The ultimate goal is to develop a reinforcement learning–based autonomous orbital maneuvering algorithm in the space environment and, through this, to establish a Guidance, Navigation, and Control (GNC) system capable of autonomous threat avoidance, strategic repositioning, and mission continuity. To achieve this, LEO, MEO, and GEO orbital environments will be configured, and debris-avoidance scenarios will be designed to conduct threat-responsive autonomous satellite maneuvering simulations. Basilisk-RL, provided within the Basilisk framework, will serve as the primary reinforcement learning framework, while Gymnasium and RLlib will be employed as supplementary tools.
Thanks for reading!
