Theoretical and experimental application of neural networks in spaceflight control systems

“Spaceflight systems can enable advanced mission concepts that can help expand our understanding of the universe. To achieve the objectives of these missions, spaceflight systems typically leverage guidance and control systems to maintain some desired path and/or orientation of their scientific instrumentation. A deep understanding of the natural dynamics of the environment in which these spaceflight systems operate is required to design control systems capable of achieving the desired scientific objectives. However, mitigating strategies are critically important when these dynamics are unknown or poorly understood and/or modelled. This research introduces two neural network methodologies to control the translation and rotation dynamics of spaceflight systems. The first method uses a neural network to perform nonlinear estimation in the control space for both translational and attitude control. The second method uses an observer with a neural network to perform estimation outside the control space, and input-output feedback linearization using the estimated dynamics for both translational and attitude control. The methods are demonstrated for attitude control through simulation and hardware testing on the Wallops Arc-Second Pointer, a high-altitude balloon-borne spaceflight system. Results show that the two new methodologies can provide improved attitude control performance over the heritage control system. The methods are also demonstrated for translational and attitude control of two small spacecraft in a deep space environment, where they provide improved position and attitude control performance as compared to a traditional control method. This work demonstrates, through simulation and hardware testing, that the two neural network methods presented can offer improved translational and attitude control performance of spaceflight systems where the dynamic environment may be unknown or poorly understood and/or modeled”--Abstract, page iv.

Name thesis:degree_name

Ph. D. in Aerospace Engineering

Grantor

Missouri University of Science and Technology

Author dc:creator

• Galchenko, Pavel

• Control
• Formation Flight
• Gimbal Systems
• Neural Networks
• Spaceflight Systems
• Aerospace Engineering
• Artificial Intelligence and Robotics

Repository record dc:identifier

https://scholarsmine.mst.edu/doctoral_dissertations/3148

OAI identifier oai:identifier

oai:scholarsmine.mst.edu:doctoral_dissertations-4153