Optimal Autonomous Guidance of Rocket Propelled Projectiles
Abstract
The objective of the proposed research is to develop preliminary analytical and computational methods for optimal onboard guidance of Rocket Propelled Projectiles (RPPs). The proposed effort will develop a real-time convex optimization based method that will fully exploit the flight envelope of the projectile by finding the optimal trajectories achieving the mission goals while satisfying mission and operational constraints. The proposed guidance methods are based on the convexification of the underlying optimal control problems. Convexification is a mathematical technique that enables the conversion of the complex trajectory optimization problems with many mission constraints into numerically tractable convex optimization problems, which can then be solved reliably and rapidly by existing Interior Point Method (IPM) algorithms. IPMs guarantee the computation of solutions in milliseconds allowing real-time onboard implementability. Indeed our research group has developed one of the first such algorithm for optimal guidance of planetary landing spacecraft and demonstrated it onboard a NASA test rocket, which proved a dramatic advancement in autonomous rocket technology. The proposed research builds on the same analytical and numerical foundation and it will expand the onboard active guidance capabilities of the RPPs. An important characteristic of the proposed optimization based guidance methodology is that it enables a systematic analysis of tradeoffs between key physical and operational parameters and the performance objectives. Such tradeoffs can include the amount of onboard fuel versus flight distance and accuracy, or onboard sensor quality versus delivery precision. The proposed methods and algorithms will quantify these tradeoffs systematically and accurately. Furthermore they will allow these computations to be done rapidly, which will also enable rigorous RPP design optimization. Specifically the proposed research is a preliminary investigation that aims to establish the necessary analytical framework and the preliminary algorithmic capabilities for optimal guidance of RPPs by progressing in three complementary research thrusts. · Thrust I – Formulation of optimal guidance problems for RPPs: What are the essential dynamics of the RPPs? What types of actuators are used for guidance? What are the actuator constraints? What are the available sensors and their measurement type, quality, and frequency? What are the relevant performance metrics, and how can they be formulated? What are possible mission and operational constraints and how can they be formulated? · Thrust II – Development of preliminary convex optimization based guidance algorithms: How can we formulate the optimal guidance problems, with all relevant constraints and performance objectives, as convex optimization problems? How can we minimize the problem complexity without sacrificing performance? Are these solutions methods real-time onboard implementable? · Thrust III – Development of preliminary algorithms to compute flight performance envelopes: How can we characterize the boundaries of the flight envelope as a function of physical parameters of an RPP, and mission and operational constraints? Can these flight envelopes be used for systematic design optimization for advanced RPP systems? This proposal will establish the fundamental analytical and numerical framework to develop optimal RPP guidance algorithms and will provide the initial evidence for potential improvements in optimal guidance of autonomous RPPs. As a part of this study, we will also provide a roadmap for the next stage of the research to push these algorithms into practice and to enable their implementations on the next generation RPPs of the Naval services.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 12, 2016
- Source ID
- N000141512410
Entities
People
- Behçet Açıkmeşe
Organizations
- Office of Naval Research
- United States Navy
- University of Texas at Austin