TIME-OPTIMAL REORIENTATION USING NEURAL NETWORK AND PARTICLE SWARM FORMULATION

A neural network will be developed to supplement a particle swarm algorithm to find near-minimum-time reorientation maneuvers in the presence of path constraints. The method employs a quaternion formulation of the kinematics, using B-splines to represent the quaternions. Dynamic Inversion will be used in the supervised training of the neural network.