Multiple sliding surface guidance for planetary landing: Tuning and optimization via reinforcement learning

Daniel Wibben, Brian Gaudet, Roberto Furfaro, Jules Simo

Research output: Contribution to conferencePaper

1 Citation (Scopus)

Abstract

The problem of achieving pinpoint landing accuracy in future space missions to extra-terrestrial bodies such as the Moon or Mars presents many challenges, including the requirements of higher accuracy and more flexibility. These new challenges may require the development of novel and more advanced guidance algorithms. Conventional guidance schemes, which generally require a
combination of off-line trajectory generation and real-time, trajectory tracking algorithms, have worked well in the past but may not satisfy the more stringent and difficult landing requirements imposed by future mission architectures to bring landers very near to specified locations. In this paper, a novel non-linear guidance algorithm for planetary landing is proposed and analyzed. Based on
Higher-Order Sliding Control (HOSC) theory, the Multiple Sliding Surface Guidance (MSSG) algorithms has been specifically designed to take advantage of the ability of the system to reach the sliding surface in a finite time. The high control activity seen in typical sliding controllers is avoided in this formulation, resulting in a guidance law that is both globally stable and robust against unknown, but bounded perturbations. The proposed MSSG does not require any off-line trajectory generation and therefore it is flexible enough to target a large variety of point on the planet’s surface without the need for calculation of multiple reference trajectories. However, after initial analysis, it has been seen that the performance of MSSG is very sensitive to the choice in guidance gains. MSSG generated trajectories have been compared to an optimal solution to begin an investigation of the relationship between the optimality and performance of MSSG and the selection of the guidance parameters. A full study has been performed to investigate and tune the parameters of MSSG utilizing reinforcement learning in order to truly optimize the performance of the MSSG algorithm. Results show that the MSSG algorithm can indeed generate trajectories that come very close to the optimal solution in terms of fuel usage. A full comparison of the trajectories is included, as well as a further study examining the capability of the MSSG algorithm under perturbed conditions using the optimized set of parameters.
LanguageEnglish
PagesPaper AAS 13-328, p56
Publication statusPublished - 10 Feb 2013
Event23rd AAS/AIAA Space Flight Mechanics Conference - Kauai, Hawaii, United States
Duration: 10 Feb 201314 Feb 2013

Conference

Conference23rd AAS/AIAA Space Flight Mechanics Conference
CountryUnited States
CityKauai, Hawaii
Period10/02/1314/02/13

Fingerprint

Electronic guidance systems
Reinforcement learning
Landing
Tuning
Trajectories
Moon
Planets
Control theory

Keywords

  • sliding mode control
  • planetary landing
  • tuning control applications
  • reinforcement learning

Cite this

Wibben, D., Gaudet, B., Furfaro, R., & Simo, J. (2013). Multiple sliding surface guidance for planetary landing: Tuning and optimization via reinforcement learning. Paper AAS 13-328, p56. Paper presented at 23rd AAS/AIAA Space Flight Mechanics Conference, Kauai, Hawaii, United States.
Wibben, Daniel ; Gaudet, Brian ; Furfaro, Roberto ; Simo, Jules. / Multiple sliding surface guidance for planetary landing : Tuning and optimization via reinforcement learning. Paper presented at 23rd AAS/AIAA Space Flight Mechanics Conference, Kauai, Hawaii, United States.
@conference{12b8149944d24527b1abcf2866d44723,
title = "Multiple sliding surface guidance for planetary landing: Tuning and optimization via reinforcement learning",
abstract = "The problem of achieving pinpoint landing accuracy in future space missions to extra-terrestrial bodies such as the Moon or Mars presents many challenges, including the requirements of higher accuracy and more flexibility. These new challenges may require the development of novel and more advanced guidance algorithms. Conventional guidance schemes, which generally require acombination of off-line trajectory generation and real-time, trajectory tracking algorithms, have worked well in the past but may not satisfy the more stringent and difficult landing requirements imposed by future mission architectures to bring landers very near to specified locations. In this paper, a novel non-linear guidance algorithm for planetary landing is proposed and analyzed. Based onHigher-Order Sliding Control (HOSC) theory, the Multiple Sliding Surface Guidance (MSSG) algorithms has been specifically designed to take advantage of the ability of the system to reach the sliding surface in a finite time. The high control activity seen in typical sliding controllers is avoided in this formulation, resulting in a guidance law that is both globally stable and robust against unknown, but bounded perturbations. The proposed MSSG does not require any off-line trajectory generation and therefore it is flexible enough to target a large variety of point on the planet’s surface without the need for calculation of multiple reference trajectories. However, after initial analysis, it has been seen that the performance of MSSG is very sensitive to the choice in guidance gains. MSSG generated trajectories have been compared to an optimal solution to begin an investigation of the relationship between the optimality and performance of MSSG and the selection of the guidance parameters. A full study has been performed to investigate and tune the parameters of MSSG utilizing reinforcement learning in order to truly optimize the performance of the MSSG algorithm. Results show that the MSSG algorithm can indeed generate trajectories that come very close to the optimal solution in terms of fuel usage. A full comparison of the trajectories is included, as well as a further study examining the capability of the MSSG algorithm under perturbed conditions using the optimized set of parameters.",
keywords = "sliding mode control, planetary landing, tuning control applications, reinforcement learning",
author = "Daniel Wibben and Brian Gaudet and Roberto Furfaro and Jules Simo",
year = "2013",
month = "2",
day = "10",
language = "English",
pages = "Paper AAS 13--328, p56",
note = "23rd AAS/AIAA Space Flight Mechanics Conference ; Conference date: 10-02-2013 Through 14-02-2013",

}

Wibben, D, Gaudet, B, Furfaro, R & Simo, J 2013, 'Multiple sliding surface guidance for planetary landing: Tuning and optimization via reinforcement learning' Paper presented at 23rd AAS/AIAA Space Flight Mechanics Conference, Kauai, Hawaii, United States, 10/02/13 - 14/02/13, pp. Paper AAS 13-328, p56.

Multiple sliding surface guidance for planetary landing : Tuning and optimization via reinforcement learning. / Wibben, Daniel; Gaudet, Brian; Furfaro, Roberto; Simo, Jules.

2013. Paper AAS 13-328, p56 Paper presented at 23rd AAS/AIAA Space Flight Mechanics Conference, Kauai, Hawaii, United States.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Multiple sliding surface guidance for planetary landing

T2 - Tuning and optimization via reinforcement learning

AU - Wibben, Daniel

AU - Gaudet, Brian

AU - Furfaro, Roberto

AU - Simo, Jules

PY - 2013/2/10

Y1 - 2013/2/10

N2 - The problem of achieving pinpoint landing accuracy in future space missions to extra-terrestrial bodies such as the Moon or Mars presents many challenges, including the requirements of higher accuracy and more flexibility. These new challenges may require the development of novel and more advanced guidance algorithms. Conventional guidance schemes, which generally require acombination of off-line trajectory generation and real-time, trajectory tracking algorithms, have worked well in the past but may not satisfy the more stringent and difficult landing requirements imposed by future mission architectures to bring landers very near to specified locations. In this paper, a novel non-linear guidance algorithm for planetary landing is proposed and analyzed. Based onHigher-Order Sliding Control (HOSC) theory, the Multiple Sliding Surface Guidance (MSSG) algorithms has been specifically designed to take advantage of the ability of the system to reach the sliding surface in a finite time. The high control activity seen in typical sliding controllers is avoided in this formulation, resulting in a guidance law that is both globally stable and robust against unknown, but bounded perturbations. The proposed MSSG does not require any off-line trajectory generation and therefore it is flexible enough to target a large variety of point on the planet’s surface without the need for calculation of multiple reference trajectories. However, after initial analysis, it has been seen that the performance of MSSG is very sensitive to the choice in guidance gains. MSSG generated trajectories have been compared to an optimal solution to begin an investigation of the relationship between the optimality and performance of MSSG and the selection of the guidance parameters. A full study has been performed to investigate and tune the parameters of MSSG utilizing reinforcement learning in order to truly optimize the performance of the MSSG algorithm. Results show that the MSSG algorithm can indeed generate trajectories that come very close to the optimal solution in terms of fuel usage. A full comparison of the trajectories is included, as well as a further study examining the capability of the MSSG algorithm under perturbed conditions using the optimized set of parameters.

AB - The problem of achieving pinpoint landing accuracy in future space missions to extra-terrestrial bodies such as the Moon or Mars presents many challenges, including the requirements of higher accuracy and more flexibility. These new challenges may require the development of novel and more advanced guidance algorithms. Conventional guidance schemes, which generally require acombination of off-line trajectory generation and real-time, trajectory tracking algorithms, have worked well in the past but may not satisfy the more stringent and difficult landing requirements imposed by future mission architectures to bring landers very near to specified locations. In this paper, a novel non-linear guidance algorithm for planetary landing is proposed and analyzed. Based onHigher-Order Sliding Control (HOSC) theory, the Multiple Sliding Surface Guidance (MSSG) algorithms has been specifically designed to take advantage of the ability of the system to reach the sliding surface in a finite time. The high control activity seen in typical sliding controllers is avoided in this formulation, resulting in a guidance law that is both globally stable and robust against unknown, but bounded perturbations. The proposed MSSG does not require any off-line trajectory generation and therefore it is flexible enough to target a large variety of point on the planet’s surface without the need for calculation of multiple reference trajectories. However, after initial analysis, it has been seen that the performance of MSSG is very sensitive to the choice in guidance gains. MSSG generated trajectories have been compared to an optimal solution to begin an investigation of the relationship between the optimality and performance of MSSG and the selection of the guidance parameters. A full study has been performed to investigate and tune the parameters of MSSG utilizing reinforcement learning in order to truly optimize the performance of the MSSG algorithm. Results show that the MSSG algorithm can indeed generate trajectories that come very close to the optimal solution in terms of fuel usage. A full comparison of the trajectories is included, as well as a further study examining the capability of the MSSG algorithm under perturbed conditions using the optimized set of parameters.

KW - sliding mode control

KW - planetary landing

KW - tuning control applications

KW - reinforcement learning

UR - http://www.space-flight.org/docs/2013_winter/FinalProgram.pdf

M3 - Paper

SP - Paper AAS 13-328, p56

ER -

Wibben D, Gaudet B, Furfaro R, Simo J. Multiple sliding surface guidance for planetary landing: Tuning and optimization via reinforcement learning. 2013. Paper presented at 23rd AAS/AIAA Space Flight Mechanics Conference, Kauai, Hawaii, United States.