DIRTREL: Robust Trajectory Optimization with Ellipsoidal Disturbances and LQR Feedback

Manchester, Zachary; Kuindersma, Scott

doi:10.15607/rss.2017.xiii.057

Cited by 23 publications

(19 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have made our code available at github.com/RoboticExplorationLab/SystemVe rification.jl. Algorithm 1 along with a multiplierbased funnel solver, a Monte Carlo funnel solver and a linear dynamics ellipsoidal funnel solver [40] are implemented in the package. The optimization solver Mosek [41] is used to solve SDPs where needed.…”

Section: Simulation Resultsmentioning

confidence: 99%

Robust Entry Vehicle Guidance with Sampling-Based Invariant Funnels

Derollez¹,

Cleac’h²,

Manchester³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Managing uncertainty is a fundamental and critical issue in spacecraft entry guidance. This paper presents a novel approach for uncertainty propagation during entry, descent and landing that relies on a new sum-of-squares robust verification technique. Unlike risk-based and probabilistic approaches, our technique does not rely on any probabilistic assumptions. It uses a set-based description to bound uncertainties and disturbances like vehicle and atmospheric parameters and winds. The approach leverages a recently developed sampling-based version of sum-of-squares programming to compute regions of finite time invariance, commonly referred to as "invariant funnels". We apply this approach to a three-degree-of-freedom entry vehicle model and test it using a Mars Science Laboratory reference trajectory. We compute tight approximations of robust invariant funnels that are guaranteed to reach a goal region with increased landing accuracy while respecting realistic thermal constraints.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Robust Entry Vehicle Guidance with Sampling-Based Invariant Funnels

Derollez¹,

Cleac’h²,

Manchester³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Third, we believe a strong case exists for optimizing for robustness when generating the reference trajectory, as explored in [17]. For instance, the closer the vehicle is to 45 degrees pitch when it exits the water (see Figure 19), the more likely it is to transition into a prop-hang and then forward flight.…”

Section: Discussionmentioning

confidence: 99%

Closed-Loop Control of a Delta-Wing Unmanned Aerial-Aquatic Vehicle

Moore

2019

Preprint

View full text Add to dashboard Cite

We present a closed-loop control strategy for a delta-wing unmanned aerial aquatic-vehicle (UAAV) that enables autonomous swim, fly, and water-to-air transition. Our control system consists of a hybrid state estimator and a closedloop feedback policy which is capable of trajectory following through the water, air and transition domains. To test our estimator and control approach in hardware, we instrument the vehicle presented in [1] with a minimalistic set of commercial off-the-shelf sensors. Finally, we demonstrate a successful autonomous water-to-air transition with our prototype UAAV system and discuss the implications of these results with regards to robustness.

show abstract

“…The linear-quadratic regulator is a well-established control scheme for (locally) linear systems. Control strategies based on LQR have been proposed in a variety of forms and for several systems, including the acrobot [43]- [46] and the cartpole [47]- [50], which are also investigated in this paper. All of the LQR variants cited above are built on minimal-coordinate representations.…”

Section: Linear-quadratic Regulationmentioning

confidence: 99%

Linear-Quadratic Optimal Control in Maximal Coordinates

Brüdigam,

Manchester

2020

Preprint

Self Cite

View full text Add to dashboard Cite

The Linear-quadratic regulator (LQR) is an efficient control method for linear and linearized systems. Typically, LQR is implemented in minimal coordinates (also called generalized or "joint" coordinates). However, recent research suggests that there may be numerical and control-theoretic advantages when using higher-dimensional non-minimal state parameterizations for dynamical systems. One such parameterization is maximal coordinates, in which each link in a multi-body system is parameterized by its full SE(3) pose and joints between links are modeled with algebraic constraints. Such constraints can also account for closed kinematic loops or contact with the environment. This paper presents an extension to the standard LQR control method to handle dynamical systems with explicit physical constraints. Experiments comparing the basins of attraction and tracking performance of minimal-and maximal-coordinate LQR controllers suggest that maximal-coordinate LQR achieves greater robustness and improved tracking compared to minimalcoordinate LQR when applied to nonlinear systems.

show abstract

DIRTREL: Robust Trajectory Optimization with Ellipsoidal Disturbances and LQR Feedback

Cited by 23 publications

References 34 publications

Robust Entry Vehicle Guidance with Sampling-Based Invariant Funnels

Robust Entry Vehicle Guidance with Sampling-Based Invariant Funnels

Closed-Loop Control of a Delta-Wing Unmanned Aerial-Aquatic Vehicle

Linear-Quadratic Optimal Control in Maximal Coordinates

Contact Info

Product

Resources

About