Trajectory Optimization and Guidance Design by Convex Programming

The field of aerospace guidance and control has recently been evolving from focusing on traditional laws and controllers to numerical algorithms with the aim of achieving onboard applications for autonomous vehicle systems. However, it is very difficult to perform complex guidance and control missions with highly nonlinear dynamic systems and many constraints onboard. In recent years, an emerging trend has occurred in the field of Computational Guidance and Control (CG&C). By taking advantage of convex optimization and highly efficient interior point methods, CG&C allows complicated guidance and control problems to be solved in real time and offers great potential for onboard applications. With the significant increase in computational efficiency, convex-optimization-based CG&C is expected to become a fundamental technology for system autonomy and autonomous operations. In this dissertation, successive convex approaches are proposed to solve optimal control programs associated with aerospace guidance and control, and the emphasis is placed on potential onboard applications. First, both fuel-optimal and time-optimal low-thrust orbit transfer problems are investigated by a successive second-order cone programming method. Then, this convex method is extended and improved to solve hypersonic entry trajectory optimization problems by taking advantage of line-search and trust-region techniques. Finally, the successive convex approach is modified to the design of autonomous entry guidance algorithms. Simulation results indicate that the proposed methodologies are capable of generating accurate solutions for low-thrust orbit transfer problems and hypersonic entry problems with fast computational speed. The proposed methods have great potential for onboard applications