Abstract
Interstellar probe swarms must coordinate under extreme latency, intermittent connectivity, and heterogeneous energy budgets. We present Intent-CRDT with Contact-Plan DTN (ICCD), a distributed control framework that maintains coherent mission intent across million-agent populations separated by light-minutes to hours. ICCD encodes agent goals and summaries as compact conflict-free replicated data types (CRDTs) and schedules dissemination via a contact-plan delay-/disruption-tolerant networking (DTN) layer. Agent-level policies are implemented in Gossamer (v0.2) and executed in the Leviathan Engine (py-0.2.0) with the upgraded communication model (range-dependent path loss, bandwidth, latency, and Bernoulli loss accounted for inside the C++ core), while Maneuver.Map orchestrates multi-generation parameter sweeps and visualization. We explicitly scope ICCD to intra-cluster coordination — agents within a light-second-scale cluster — with inter-cluster intent flow handled by sparse high-power relays as part of the contact plan. In large-scale simulations (up to 1×$10^6$ agents over 3 AU), ICCD reduced critical-intent age-of-information (AoI) by 41% ± 3% and improved formation coherence ψ by 23% over a periodic-broadcast baseline at comparable per-bit energy. ICCD also outperforms a learned MAPPO relay-selection baseline trained against the same DTN constraints by 0.06 absolute on ψ at 4× lower bandwidth. Under 20% relay attrition and 0.3–1.5 hour one-way delay, ICCD sustained ≥92% task completion and 0.31 J/KB median intra-cluster energy cost through energy-aware relay rotation. Results indicate that CRDT-based intent combined with contact-aware scheduling preserves global coordination without centralized control, enabling feasible long-baseline exploration, survey, and construction missions.