Embedded Files
A self-navigating balloon driven by reinforcement learning in the stratosphere.
Project mission
Project mission
Project Stratus aims to build a multi-purpose weather balloon that can navigate itself horizontally through the stratosphere. We will achieve this by controlling the balloon’s altitude through negative feedback loops to ride air currents that flow in different horizontal directions at different altitudes. Our final goal is to employ a machine learning model to predict wind velocities based on data collected by a set of mounted sensors and use it to create a weather balloon that can serve as a cost-efficient and versatile alternative to low orbit satellites for weather monitoring and Earth observation imagery.
Electronics
Electronics
The Electronics team focuses on designing and integrating the balloon's core systems, including sensors, communication modules, and power management.
Artificial Intelligence
Artificial Intelligence
The AI team develops advanced reinforcement learning algorithms that enable the balloon to navigate autonomously. By analyzing sensor data and predicting wind patterns, the AI adapts in real-time to dynamic conditions, optimizing the balloon's altitude for efficient travel and data collection.
Orbital Dynamics
Orbital Dynamics
While primarily focused on high-altitude ballooning, the Orbital Dynamics team applies principles of orbital mechanics to model the trajectory and movement of the balloon in the stratosphere. Their work contributes to precision navigation and future scalability for orbital systems.
Structure and Design
Structure and Design
The Structural team is responsible for designing the physical framework of the balloon and its payload. Using lightweight yet durable materials, the team ensures the system can withstand stratospheric pressures, temperatures, and turbulence while maximizing payload efficiency and stability.
Page updated
Google Sites
Report abuse