Abstract The development of robust fuel-optimal feedback controllers for the pinpoint landing problem has application to a variety of aerospace vehicles including lunar and planetary landers. Imitation learning has been used for a variety of low-fidelity lander formulations to train a policy based on numerically generated open-loop optimal trajectories. The simplest technique, Behavioral Cloning, can suffer from distribution shift, where the distribution of training data is different than the distribution of states seen in rollout. This paper presents a new metric to measure this shift, and a comprehensive study is performed on a variety of lander formulations and policy classes. It is found that shift increases with dimensionality of and the level of coupling in equations of motion, and that the non-parametric policy yields more shift than the parametric one.
Graphical abstract Display Omitted
Highlights Imitation learning can produce optimal feedback controllers for the pinpoint landing problem. Shifts in state distributions between training and rollout can cause poor performance. Gaussian Mixture Models can fit to state distributions. Approximations of the Kullback–Leibler divergence show higher shift for some landing formulations/policies than others.
Metric to evaluate distribution shift from behavioral cloning for fuel-optimal landing policies
Acta Astronautica ; 203 ; 421-428
2022-12-09
8 pages
Article (Journal)
Electronic Resource
English
Fuel-Optimal Rocket Landing with Aerodynamic Controls
AIAA | 2018
|Fuel-Optimal Rocket Landing with Aerodynamic Controls
AIAA | 2017
|Method and system for behavioral cloning of autonomous driving policies for safe autonomous agents
European Patent Office | 2023
|