Feasibility Study of Distributed Space Antennas Using Electromagnetic Formation Flight

Distributed space antennas using electromagnetic formation flight (EMFF) can achieve long-term, high-performance communication systems. EMFF, which enables simultaneous control of relative positions and absolute attitudes of multiple objects, is a promising propulsion technology for distributed space antennas. Although distributed space antennas are dynamic systems, their shape directly affects communication performance. Therefore, evaluating control performance for shape maintenance, deployment, and reconfiguration is crucial. Previous EMFF design studies primarily focused on a few satellites, identifying trade-offs between coil size and control force. However, in distributed space antennas, the large number of satellites and the power requirements for maintaining communication performance make this trade-off significantly more complex. This paper proposes a novel design method for satellites and coil actuators that satisfies antenna performance requirements, contributing to practical design applications. By assuming a grid formation for the antenna shape, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$J_{2}$</tex> disturbances are compensated in a bucket-brigade manner. This grid assumption simplifies the derivation of antenna gain and effective isotropic radiated power (EIRP), while restricting directionality simplifies the antenna pattern. This approach enables identification of the maximum side lobe direction, allowing consideration of communication strength outside the main lobe. We applied the proposed method to three cases: two with satellite spacing of half a wavelength (one considering only disturbance compensation power and the other including deployment power) and one with 3/2-wavelength spacing considering only disturbance compensation power. The derived design solutions indicate that a larger antenna than the monolithic BlueWalker3 (8m, 1500kg) is achievable when considering only formation maintenance. When deployment power is included, the solution achieved a similar size, suggesting the potential for improved performance through scalability. To validate the feasibility of the secured deployment power, we performed deployment simulations using model predictive control (MPC). The results demonstrate that MPC achieves faster convergence to a stable orbit compared to feedback control. This validates the adequacy of the deployment power and serves as a practical example of control system design for distributed space antennas.