Student Name: Shanmurugan Selvamurugan

Advisor: Dr. E. Glenn Lightsey

Milestone: PhD Thesis Proposal

Degree Program: Aerospace Engineering

Title: Distributed Multi-Agent Space Systems for Scalable On-Orbit Transportation

Abstract: Access to high-energy orbits—such as geostationary orbit, cislunar space, and other strategic destinations—remains constrained by high launch costs, limited mission opportunities, and schedule uncertainty. These challenges disproportionately affect small satellite missions, which typically rely on rideshare deployments tied to the priorities of primary payloads. Even after reaching low Earth orbit, spacecraft must still perform substantial orbit-raising maneuvers using either onboard propulsion—which increases wet mass and launch cost—or expendable last-mile delivery systems, which add integration complexity and consume valuable launch vehicle capacity. Emerging reusable, space-resident transport vehicles via on-orbit servicing (OOS) offer a promising alternative by enabling rendezvous, proximity operations, and point-to-point orbital transfers starting from low-cost LEO launch opportunities. However, current architectures almost universally center on a single, large servicer, limiting scalability, responsiveness, and system resilience. This proposal investigates whether a distributed network of smaller, reusable servicers—coordinated as a multi-agent space transportation system—can provide a more scalable and cost-effective alternative. By decomposing transport tasks across multiple agents and exploiting natural orbital mechanics alongside active propulsion, such architectures could offer flexible, on-demand mobility while reducing the mass and complexity required of any individual vehicle. The research is organized around three core contributions. First, as a proof of concept, it formulates and solves the time-optimal transport logistics problem for delivering a single payload from LEO to MEO and GEO using high-thrust (impulsive) chemical servicers, under simplifying assumptions of coplanar orbits and Keplerian dynamics, enabling analytical trajectory computation via the universal-variable Gauss method. Second, the analysis is significantly expanded by relaxing these assumptions: the work develops and solves both the time-constrained fuel-optimal and time-optimal transport problems for single-payload delivery to GEO using low-thrust electric and multimodal chemical–electric servicers, while accounting for inclination and RAAN changes (including J2-driven precession), and incorporating vehicle design parameters such as propellant mass fraction. This stage jointly optimizes servicer trajectories and depot placement, providing a more realistic and mission-relevant formulation. Third, the research will scale these methods to solve the time-constrained fuel-optimal logistics problem involving multiple payloads and multiple destinations, assuming fixed servicer and depot assets in space and enabling coordinated multi-agent transport planning. This contribution will also include a system-level comparative analysis evaluating how distributed multi-vehicle architectures perform relative to current transportation approaches—including single direct launch to target orbit, expendable kick stages, and single reusable space tugs—across both cost and time-to-delivery metrics. Together, these contributions establish a rigorous analytical foundation for distributed multi-agent space transportation networks that can deliver scalable, resilient, and economically sustainable on-orbit mobility for future commercial, civil, and scientific missions.

Date and time: 2026-02-12, 12PM

Location: MK-325

Committee:
Dr. E. Glenn Lightsey (advisor), School of Aerospace Engineering
Dr. Koki Ho, School of Aerospace Engineering
Dr. Brian Gunter, School of Aerospace Engineering
,