CPEG: A Convex Predictor-corrector Entry Guidance Algorithm

Abstract

As scientific and crewed payloads have more demanding goals, precise atmospheric entry guidance is playing an increasing role in mission success. State-of-the-art entry guidance algorithms are structured in a predictor-corrector framework, where a simulation is used to predict a trajectory, and corrections are then made to the control inputs. These guidance methods are simple and effective, but current algorithms assume low lift-to-drag entry vehicles, are limited to only bank-angle control, and have a limited ability to guarantee the safety of the vehicle. We propose a new predictor-corrector entry guidance method that formulates the correction step as a convex optimization problem. This allows for more flexibility in specifying the vehicle's dynamics and control inputs, and the ability to explicitly handle safety constraints such as heating, pressure, and acceleration limits. We test the new algorithm in Mars entry scenarios similar to the Mars Science Laboratory with both bank-angle control and bank-angle plus angle-of-attack control, demonstrating both its performance and ability to generalize to future vehicle capabilities.