Miniaturized Hypoxic Generator Based on Reinforcement Learning for Continuous Altitude Simulation

Appropriate altitude pre-acclimatization training can effectively reduce the risk of acute altitude sickness among personnel in plateau areas. However, the size of the current mainstream plateau hypoxic generator system is too large, which is not conducive to widespread application in large areas. This paper proposes a deep reinforcement learning model that integrates an anti-time variation training method, which can effectively simplify the structural composition of the plateau hypoxic generator system and improve its miniaturization, and only need to install the sensor detection device at the output end of the system. Meanwhile, the algorithm proposed in this paper improves the average steady-state error index by 68.3% compared with the traditional deep reinforcement learning algorithm and PID classic control algorithm. Finally, we apply our proposed algorithm to design a practically functioning miniaturized hypoxia generation system.