Backend
Averting `Gravity': Launch Control and the Limits of Collision Avoidance in Low Earth Orbit Operations
arXiv eess.SY0 views
arXiv:2611.23607v1 Announce Type: cross Abstract: As large constellations fill low Earth orbit, the risk of a runaway collision cascade has become a central concern of space operations. Debris population models treat collision avoidance as a fixed input, although the capacity to screen potential collisions and plan maneuvers degrades as debris accumulates. Using a stylized multi-layer queueing network model, we show that collision avoidance capabilities that saturate as the in-custody fragment count increases, can make the orbital shell bistable, with a low-debris and a high-debris equilibrium, and hysteretic. This has two major operational consequences. First, leaving the high-debris equilibrium requires a large cut in launch rates. Second, in the low-debris equilibrium, fragment counts respond with a delay as the launch rate rises past a tipping point. Removing bistability is desirable, and we show that fragment count-indexed admission control can achieve this. The least restrictive such policy lengthens the lag, whereas an exponential tilt of this policy makes the fragment count track its equilibrium with a bounded lag. Furthermore, admitting more launches can reduce the rate of completed missions, and under a unimodality condition a threshold rule on the debris fragment count maximizes this rate at equilibrium.
Source:https://arxiv.org/abs/2611.23607PreviousPlug-and-Play Sensor Evolution in Software-Defined Vehicles through Asset Administration ShellsNextPhysical Layer Security for Digital Semantic Communication: A Bit-Level Perspective
Related posts
Bad Genius: Counterfactual-Guided Harness Evolution Beyond Task-Specific Shortcuts
0arXiv stat.ML
Kinetic Langevin Splitting Schemes for Constrained Sampling
0arXiv stat.ML
From Data to Action: Accelerating Refinery Optimization with AI
0arXiv stat.ML
Equivariant Modeling with Polynomial Features via Minimal Homogeneous Generators
0arXiv stat.ML
Latent Action Policies from Suboptimal Data
0arXiv stat.ML
SoK: Market Microstructure for Decentralized Prediction Markets (DePMs)
0arXiv cs.CR