Pseudospectral Optimal Control for Flight Trajectory Optimization

This Phase II effort aims to deliver robust, reliable software that instantiates state-of-the-art feedback pseudospectral optimal control algorithms for flight trajectory optimization. The computation and real-time implementation of controls in nonlinear systems remains one of the great challenges for applying optimal control theory in demanding aerospace and industrial systems. From proportional guidance in missiles to PID controllers for UAV flight controls to linear integrators in optical tracking, linear controls dominate much of current implementation. Output feedback is of course one important consideration: optimal controls determined from Pontryagins principle are generally open-loop. Computation is a second difficulty: use of Pontryagins principle, dynamic programming, or direct optimization methods using conventional computational designs in high dimensional nonlinear systems has been considered largely unrealistic. Our Phase I proof-of-concept study has demonstrated that pseudospectral optimal control can be implemented with feedback to meet the demanding challenges of flight trajectory optimization challenges. Evolving proof-of-concept algorithms and codes to fieldable software applications that can perform in real-time hardware requires careful testing and implementation of code modules, integration of units into an operable package, and lab and field testing in scalable hardware-in-the-loop environments. Our Phase II objectives and work plan lay out the details of this development effort.