An adaptive Neuro-Fuzzy control approach for motion control of a spacecraft maneuver

This paper proposes an adaptive Neuro-Fuzzy control approach to predict the torque required to control the attitude and rate in a spacecraft maneuver. In the real world environment, the mathematical models of many complex systems are often not accurate, due to the presence of continuous disturbances that effect their dynamic equations, in addition to errors in parameter knowledge. Consequently, methods that rely less on precise mathematical models are often preferred. One such Adaptive Machine Learning Technique is proposed for motion control in spacecraft maneuver. The controller uses an inverse learning Adaptive Neuro-Fuzzy Inference System (ANFIS) model only to train itself from certain desired trajectories and tries to mimic the same in its response. Ideally, these training trajectories are obtained by directly measuring the spacecraft manuever response for various test inputs. Once the system is fully trained, the manuever is tested on a new trajectory with uncertain plant dynamics. However, for algorithm validation, trajectories generated through simulations based on mathematical models assumed to be reasonably accurate, can also be used for the training purpose. This approach is used for design and implementation of an ANFIS controller which is shown to work satisfactorily. Further possible developments of the method are outlined.