Research
from the Galactic Bioware team.

Technical papers on AI-controlled life support, semi-closed-circuit breathing and the control theory behind it. Each paper is hosted by its publisher.

arXiv:2603.26697 · eess.SY

Physicochemical-Neural Fusion for Semi-Closed-Circuit Respiratory Autonomy in Extreme Environments

Galactic Bioware's Life Support System — a semi-closed-circuit breathing apparatus for a positive-pressure firefighting suit, governed by an AI control system. Part I develops the physicochemical foundations from first principles; Part II introduces a model-predictive and reinforcement-learning control architecture behind a control-barrier-function safety filter, demonstrating 18-34% endurance improvement in simulation over PID baselines.

Phillip Kingston, Nicholas Johnston 16 March 2026 ResearchGate:10.13140/RG.2.2.14309.36325 · Preprint

Exact Viability and Gas Allocation in a Capacity-Limited Venting Two-Gas Atmosphere

An exact analytic treatment of resource allocation in a reduced venting two-gas atmosphere. A scalar comparison result yields deterministic closed-form conditions for finite-horizon viability, maximum duration, viable initial compositions and optimal initialization. The analysis extends to the regime where finite actuator capacities induce mandatory oxygen and diluent flows.

Phillip Kingston 2026

Research

Physicochemical-Neural Fusion for Semi-Closed-Circuit Respiratory Autonomy in Extreme Environments

16 March 2026 Phillip Kingston, Nicholas Johnston

Abstract

This paper introduces Galactic Bioware's Life Support System, a semi-closed-circuit breathing apparatus designed for integration into a positive-pressure firefighting suit and governed by an AI control system. The breathing loop incorporates a soda lime CO2 scrubber, a silica gel dehumidifier, and pure O2 replenishment with finite consumables. One-way exhaust valves maintain positive pressure while creating a semi-closed system in which outward venting gradually depletes the gas inventory.

Part I develops the physicochemical foundations from first principles, including state-consistent thermochemistry, stoichiometric capacity limits, adsorption isotherms, and oxygen-management constraints arising from both fire safety and toxicity. Part II introduces an AI control architecture that fuses three sensor tiers, external environmental sensing, internal suit atmosphere sensing (with triple-redundant O2 cells and median voting), and firefighter biometrics. The controller combines receding-horizon model-predictive control (MPC) with a learned metabolic model and a reinforcement learning (RL) policy advisor, with all candidate actuator commands passing through a final control-barrier-function safety filter before reaching the hardware. This architecture is intended to optimize performance under unknown mission duration and exertion profiles.

In this paper we introduce an 18-state, 3-control nonlinear state-space formulation using only sensors viable in structural firefighting, with triple-redundant O2 sensing and median voting. Finally, we introduce an MPC framework with a dynamic resource scarcity multiplier, an RL policy advisor for warm-starting, and a final control-barrier-function safety filter through which all actuator commands must pass, demonstrating 18-34% endurance improvement in simulation over PID baselines while maintaining tighter physiological and fire-safety margins.

arXiv
2603.26697 · eess.SY
Cite as
arXiv:2603.26697 [eess.SY]
Submitted
16 March 2026
Authors
Phillip Kingston, Nicholas Johnston

View on arXiv → Download PDF ↓

Research

Exact Viability and Gas Allocation in a Capacity-Limited Venting Two-Gas Atmosphere

2026 Phillip Kingston

Abstract

We solve a scalar state-constrained allocation problem with two finite resources and pressure-coupled actuator bounds, motivated by a reduced venting two-gas atmosphere. The atmosphere is isothermal and perfectly mixed; its total inventory is held constant by exact pressure regulation, and oxygen fraction is confined to a prescribed band.

For constant known loads, nonbinding actuators, and a prescribed initial composition, a scalar comparison result gives the exact minimum cumulative draw of each species. Conservation and convexity determine the complete cumulative oxygen-allocation interval, whose intersection with the reserve constraints yields exact deterministic open-loop formulas for finite-horizon viability, maximum duration, viable initial compositions, and optimal initialization. Free-initial duration, reserve regimes, and reserve shadow prices follow as corollaries.

For finite actuator capacities satisfying the global pointwise-authority conditions, the capacity limits induce mandatory oxygen and diluent flows. The same scalar lemma gives exact shifted allocation endpoints, an explicit viability kernel within this authority-admissible regime, and a piecewise closed-form maximum-duration formula. A residual-flexibility identity links instantaneous actuator authority to the asymptotic width growth of the cumulative-allocation set.

Discretized linear programs provide deterministic same-model checks of authority cases, viability boundaries, initialization, and reduction to the full allocation range when both capacities are nonbinding. The resulting formulas provide an exact analytic benchmark for the stated scalar finite-resource allocation model.

ResearchGate
10.13140/RG.2.2.14309.36325 · Preprint
Cite as
DOI:10.13140/RG.2.2.14309.36325
Submitted
2026
Authors
Phillip Kingston

View on ResearchGate →