Model-Predictive Spiral and Spin Upset Recovery Control for the Generic Transport Model Simulation⋆

Cunis, Torbjørn; Liao‐McPherson, Dominic; Kolmanovsky, Ilya; Burlion, Laurent

doi:10.1109/ccta41146.2020.9206158

Cited by 3 publications

(2 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method adds two regularization terms to the loss, a temporal smoothness term (14) and a spatial smoothness term (15). As the authors demonstrate, this method is more successful in generating controllers that yield smooth control signals compared to the action reward-term approach.…”

Section: B Reward and Objective Designmentioning

confidence: 99%

“…Many challenging control problems arise during advanced operation of fixed-wing unmanned aerial vehicles (UAVs), such as aerobatic maneuvering [9], perching [14], deep-stall landing [29], recovery from loss of control [15], flying in strong wind fields [27], or performing VTOL transitions between hover and forward flight [4]. Fixed-wing UAVs, as illustrated in Fig.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments

Bøhn¹,

Coates²,

Reinhardt³

et al. 2021

Preprint

View full text Add to dashboard Cite

Attitude control of fixed-wing unmanned aerial vehicles (UAVs) is a difficult control problem in part due to uncertain nonlinear dynamics, actuator constraints, and coupled longitudinal and lateral motions. Current state-of-the-art autopilots are based on linear control and are thus limited in their effectiveness and performance. Deep reinforcement learning (DRL) is a machine learning method to automatically discover optimal control laws through interaction with the controlled system, that can handle complex nonlinear dynamics. We show in this paper that DRL can successfully learn to perform attitude control of a fixedwing UAV operating directly on the original nonlinear dynamics, requiring as little as three minutes of flight data. We initially train our model in a simulation environment and then deploy the learned controller on the UAV in flight tests, demonstrating comparable performance to the state-of-the-art ArduPlane proportional-integral-derivative (PID) attitude controller with no further online learning required. To better understand the operation of the learned controller we present an analysis of its behaviour, including a comparison to the existing well-tuned PID controller.

show abstract

Section: B Reward and Objective Designmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments

Bøhn¹,

Coates²,

Reinhardt³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

Verification of a Nonlinear Upset Recovery Control Law using Scenario Optimisation^⋆

Holzinger,

Cunis

2024

2024 IEEE Conference on Control Technology and Applications (CCTA)

View full text Add to dashboard Cite

Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments

Bøhn

Coates

Reinhardt

et al. 2024

IEEE Trans. Neural Netw. Learning Syst.

View full text Add to dashboard Cite

Model-Predictive Spiral and Spin Upset Recovery Control for the Generic Transport Model Simulation⋆

Cited by 3 publications

References 28 publications

Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments

Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments

Verification of a Nonlinear Upset Recovery Control Law using Scenario Optimisation^⋆

Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments

Contact Info

Product

Resources

About

Model-Predictive Spiral and Spin Upset Recovery Control for the Generic Transport Model Simulation⋆

Cited by 3 publications

References 28 publications

Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments

Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments

Verification of a Nonlinear Upset Recovery Control Law using Scenario Optimisation⋆

Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments

Contact Info

Product

Resources

About

Verification of a Nonlinear Upset Recovery Control Law using Scenario Optimisation^⋆