Adaptive Control of a High Agility Model Airplane in the Presence of Severe Structural Damage and Failures

Baur, Stephan; Gibson, Travis E.; Annaswamy, Anuradha M.; Höcht, Leonhard; Bierling, Thomas; Holzapfel, Florian

doi:10.1007/978-3-642-19817-5_16

Cited by 2 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Detection of and recovery from failure has been studied extensively in aircraft [15]. [5] proposed linear equations of motion for an aircraft suffering from wing damage and actuator damage and implements a model reference adaptive controller to compensate for these failures. The researchers demonstrated they could accurately track a reference.…”

Section: Related Workmentioning

confidence: 99%

“…To the best of our knowledge, no current architecture accounts for both a continuous distribution of damage and demonstrates the computational speed necessary to regain control of an unstable system, e.g., a damaged aircraft. While some approached have modeled failure dynamics from first principles, e.g., [5], [6], these approaches are non-adaptive and restricted to a narrow set of point cases. For example, [] We move beyond the limitations of these prior works by developing active model learning techniques to safely acquire information about the altered plant dynamics to recover from failure.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

When Your Robot Breaks: Active Learning During Plant Failure

Schrum

Gombolay

2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Detecting and adapting to catastrophic failures in robotic systems requires a robot to learn its new dynamics quickly and safely to best accomplish its goals. To address this challenging problem, we propose probabilistically-safe, online learning techniques to infer the altered dynamics of a robot at the moment a failure (e.g., physical damage) occurs. We combine model predictive control and active learning within a chance-constrained optimization framework to safely and efficiently learn the new plant model of the robot. We leverage a neural network for function approximation in learning the latent dynamics of the robot under failure conditions. Our framework generalizes to various damage conditions while being computationally light-weight to advance real-time deployment. We empirically validate within a virtual environment that we can regain control of a severely damaged aircraft in seconds and require only 0.1 seconds to find safe, informationrich trajectories, outperforming state-of-the-art approaches.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

When Your Robot Breaks: Active Learning During Plant Failure

Schrum

Gombolay

2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

Architecture design of four redundant flight control system

谢,

周,

曾

et al. 2024

First Aerospace Frontiers Conference (AFC 2024)

View full text Add to dashboard Cite

Redundant flight control systems typically have high flexibility and customizability, suitable for different types of highvalue drones. This article proposes a new architecture for a four degree redundant flight control system, which is based on the traditional four degree redundant flight control system architecture. By adding a CANFD bus to achieve signal interaction between flight control computers, the resulting voting values can be used as a reference for signal arbitration in the next cycle. This design can increase the reliability of the redundancy system and solve the problem of inability to distinguish faults when the signal voting is 2:2 in traditional designs. Finally, using the proposed method, a redundant flight control system was developed for a certain experimental drone, and its effectiveness was successfully verified through ground semi physical experimental testing.

show abstract

Adaptive Control of a High Agility Model Airplane in the Presence of Severe Structural Damage and Failures

Cited by 2 publications

References 12 publications

When Your Robot Breaks: Active Learning During Plant Failure

When Your Robot Breaks: Active Learning During Plant Failure

Architecture design of four redundant flight control system

Contact Info

Product

Resources

About