Reinforcement Learning-based Fault-Tolerant Control for Quadrotor with Online Transformer Adaptation

Dohyun Kim; Hyochoong Bang; Jayden Dongwoo Lee; Jungho Bae

REVIEW 4 major objections 5 minor 16 references

Reinforcement Learning-based Fault-Tolerant Control for Quadrotor with Online Transformer Adaptation

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A transformer-based adaptation module lets a hybrid RL-PID quadrotor controller ride out loss-of-effectiveness faults, with 95% hovering success and 0.129 m position RMSE in simulation and transfer to untrained quadrotor configurations.

desk verdict A credible but incremental simulation study showing a transformer-based RMA-style adapter works for quadrotor actuator faults; the core numbers hold, but fault-range ambiguity and a suspicious 50x latent-inference gap need fixing before the headline claims stand. read the letter →

arxiv 2505.08223 v1 pith:QVRHYMJS submitted 2025-05-13 cs.RO cs.AI

Dohyun Kim , Jayden Dongwoo Lee , Hyochoong Bang , Jungho Bae This is my paper · ORCID

classification cs.ROcs.AI

keywords reinforcementlearningfault-tolerantcontrolquadrotortransformeronlineadaptationlatentrepresentationteacher-studenttrainingloss-of-effectiveness

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that a hybrid controller—a nominal PID plus a reinforcement-learning policy whose output is capped at 20% of the nominal command—can be made fault-tolerant to loss-of-effectiveness actuator faults by conditioning the policy on a latent vector inferred online from the last second of state-action data. The authors' central claim is that a transformer-based adaptation module can recover this latent description of the drone and its fault accurately enough that the same policy, trained once in simulation, keeps a quadrotor hovering under up to 30% thrust loss on one motor and transfers to untrained quadrotor configurations without retraining. In their PyBullet experiments the proposed method reaches a 95% success rate and 0.129 m positional RMSE under faults, compared with 86% and 0.153 m for a CNN-based RMA-style adaptation baseline and 53% success for plain PID. The reason to care is practical: if true, this removes the need to know the fault model or the drone's exact dynamics in advance, which is the main obstacle to field use of RL-based fault-tolerant control.

What carries the argument

The central object is the latent vector z_t, a low-dimensional representation of the environment and fault parameters (mass, inertia, arm length, thrust and torque coefficients, and the per-motor efficiency losses) that the policy conditions on. In phase one, a privileged encoder μ is trained jointly with the policy to map the true parameters ξ_t into z_t; in phase two, a transformer-based adaptation module h is trained by supervised regression to produce ẑ_t from a one-second window (30 steps) of state-action history, with μ frozen as teacher. The transformer uses self-attention over the sequence to filter noise and vibration and to extract the temporal signature of the fault; this inference replaces the privileged encoder at deployment, so the policy needs only its own history, not a model of the fault.

What would settle it

Run the same two-phase training on a quadrotor with a different nominal controller or with a shorter history window, then compare hovering success under 30% loss-of-effectiveness; if success drops to the PID level or latent RMSE grows sharply as the window shrinks, the mechanism is the window size and latent fidelity, not the transformer per se. A sharper test is to feed the policy the ground-truth privileged code z_t under a 30% fault: if closed-loop success does not approach the no-fault oracle level (100%, 0.063 m), then the bottleneck is the policy or the controller, not the adaptation module.

Watch

Extended reading notes

Core claim

The paper's central discovery is that teacher-student latent adaptation, previously demonstrated for legged robots and near-hover quadcopter control, can be applied to actuator fault tolerance, and that a transformer improves the student's inference enough to change closed-loop outcomes. The privileged encoder μ compresses the true dynamics and fault parameters into a latent code z_t during training; at deployment the adaptation module h regresses an estimate ẑ_t from the last 30 state-action pairs at 30 Hz, and the frozen policy replaces the privileged code with this estimate. The authors report that the transformer-based module infers the latent code with RMSE 0.1113, versus 5.5798 for the 1-D CNN RMA-style module, and that the corresponding hovering success on the training-configuration quadrotor is 95% versus 86%, with position RMSE 0.129 m versus 0.153 m. The same frozen controller also succeeds on two untrained quadrotor variants (91% and 96%), which the authors present as evidence that the latent code, not the specific vehicle, carries the adaptation.

Load-bearing premise

The controller works only if the internal description of the drone and its fault, learned during training, can be rebuilt from the last second of sensor readings and commands; if that reconstruction misses the fault, the policy cannot adapt and the claimed transfer to other drones fails.

Editorial extensions

If this is right

The same frozen policy can transfer to untrained quadrotor sizes and masses without retraining, with success rates of 91% and 96% on the two unseen variants.
Because latent inference stays accurate at fault magnitudes up to the tested limit, the controller remains stable beyond the narrow training fault range in the robustness experiments.
At deployment the privileged encoder is replaced entirely by the adaptation module, so real-time knowledge of mass, inertia, or motor health is not needed.
Since the policy command is capped at 20% of the nominal command, the nominal PID still provides the bulk of control effort, making the learned policy a bounded correction rather than a full replacement.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The paper evaluates only single-motor loss-of-effectiveness with efficiency between 0.8 and 0.9; if the latent mechanism is as general as claimed, the same teacher-student scheme should extend to multi-motor faults and partial rotor drag, but that extension is not demonstrated.
The large gap in latent-inference RMSE (0.1113 vs 5.5798) suggests the CNN baseline is not merely worse but effectively fails to regress the privileged code; a reader could test whether closed-loop success tracks latent accuracy by corrupting ẑ_t with varying noise.
A natural deployment test is to induce a real rotor efficiency loss on a hovering quadrotor and see whether the 30 Hz, one-second history window still yields stable hovering when the nominal PID is not tuned for the damaged vehicle.
Because the policy correction is capped at 20% of the nominal command, the fault tolerance demonstrated here is bounded by what the nominal PID can still stabilize; on platforms with weaker nominal controllers the same latent code may not rescue performance.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Read the paper. The honest summary is: it takes the RMA teacher-student recipe that Zhang et al. already applied to quadrotors, moves it to actuator loss-of-effectiveness faults, and swaps the CNN estimator for a transformer. That is a legitimate new application, not a new conceptual framework. The experiments cover three drone configurations, including two unseen ones, and the success/RMSE numbers in Table II are internally consistent and match the trajectory plots. The no-fault oracle, PID baseline, and RMA baseline are all used appropriately. Credit where due.

The soft spots are concentrated in two places. First, the fault-magnitude reporting is genuinely inconsistent. Section II-A says lambda_i in [0.8, 0.9], which is 10–20% effectiveness loss and up to 36% thrust loss. Table I lists "Fault Range [0.1, 0.2]" in the same column, and Table II says "under 30% LoE OFF2". I cannot tell whether the test condition is 30% LoE meaning lambda = 0.7 (outside training range) or 30% thrust loss meaning lambda ≈ 0.836 (inside the range). This matters for the "robust beyond training distribution" claim. The authors need to define LoE unambiguously and report exact lambda values for each experiment.

Second, the latent-inference comparison reports transformer RMSE 0.1113 versus CNN 5.5798. A factor of 50 against a frozen policy's training data is not credible on its face. Either the CNN was undertrained, its hyperparameters were not comparable, or the latent space is dominated by one or two coordinates the CNN gets systematically wrong. The paper does not tell us about the CNN training budget or the latent-coordinate distribution. As written, the "outperforming existing adaptation methods" claim is weakened because the comparison may be against a broken baseline.

The stress-test note about the 20% command cap is a fair question but not fatal. The cap limits the policy's authority but does not by itself explain the gap between the two adaptation modules. Missing code and data are real limitations, but not disqualifying.

My overall take: the direction is right, and the conclusion that the transformer adapter performs better in this setting is plausible, but the evidence as presented is not tight enough to be a headline result. With the fault-range definitions cleaned up, a fairly tuned CNN baseline, and some sensitivity analysis over fault magnitude, this becomes a solid conference paper. As is, conditional accept.

I would send this to a competent referee rather than desk reject. It is an incremental but honest engineering contribution, and a good referee can help the authors fix the reporting issues.

Referee Report

4 major / 5 minor

Summary. The paper proposes a hybrid RL-based fault-tolerant controller for quadrotor hovering under loss-of-effectiveness (LoE) actuator faults. The controller combines a nominal cascade PID with an RL policy that receives a latent code z_t; in Phase 1 the privileged encoder μ and the policy are trained jointly with PPO, and in Phase 2 a transformer-based adaptation module h is trained by supervised regression to predict z_t from a window of state-action history. At deployment h replaces μ. Experiments in gym-pybullet-drones compare the proposed method (RL-ours), a CNN-based RMA-style adaptation baseline (RL-RMA), and pure PID on the nominal Crazyflie and two untrained variants (UAV1, UAV2). The reported results are a 95% success rate and 0.129 m position RMSE for RL-ours versus 86% and 0.153 m for the CNN baseline and 53% for PID under a 30% motor-2 loss; the latent-inference RMSE is 0.1113 versus 5.5798. The paper concludes that the transformer module enables robust adaptation to unseen dynamics without retraining.

Significance. The work combines a known teacher-student adaptation framework (RMA) with a transformer student and applies it to quadrotor actuator-fault tolerance. Its strengths are that the system is evaluated on the deployment task itself (success rate, RMSE, max error) in simulation, the comparison covers three quadrotor configurations including untrained variants, and the architecture is described in sufficient detail to be plausibly reproduced. If the reported advantages survive proper statistical testing and the inconsistencies noted below are resolved, this would be a useful contribution to practical online adaptation for quadrotor FTC. However, the current evidence is weakened by ambiguous fault-range definitions, a possibly undertrained CNN baseline, a mismatch in the claimed context window, and the absence of confidence intervals or significance tests; the significance of the claimed improvement is therefore not yet established.

major comments (4)

[Section II-A, Table I, Section III-B] The fault-representation notation is internally inconsistent. Equation (2) defines λ_i as motor efficiency, and Section II-A states λ_i ∈ [0.8, 0.9], leading to up to 36% thrust loss. Table I, however, lists 'Fault Range' as [0.1, 0.2] for all three quadrotors, and the results section refers both to '30% LoE OFF2' (Table II caption) and '30% thrust loss in motor 2' (Section III-B). As written, [0.1,0.2] cannot be λ_i, and '30% LoE' is not equivalent to '30% thrust loss' (the latter corresponds to λ ≈ 0.837). The authors must define what the Table I fault range denotes, state the exact λ or loss fraction used in Table II and in the robustness tests, and indicate whether the tested condition lies inside or outside the training distribution. Without this, the comparison across methods and the generalization claims are not well posed.
[Section III-B, Latent Inference] The reported latent-inference RMSE of 0.1113 ± 0.0036 for the transformer versus 5.5798 ± 0.0009 for the CNN is implausibly large if both modules are trained to minimize the same MSE objective on the same frozen policy's data. A factor-of-50 gap suggests either the CNN baseline was not trained to convergence, the two modules received different input histories or latent dimensions, or the latent space has coordinates with very different scales. Please report per-coordinate RMSE, training curves, and the exact dataset size and optimization budget for both modules. Without this information, the claim that the transformer 'outperforms existing adaptation methods' is not supported, because the CNN baseline may be a strawman rather than a well-tuned RMA alternative.
[Section II-C and Section III-A] The input context window is inconsistently specified. Section II-C states that the adaptation module h infers ẑ_t from '1 second state-action trajectories,' while Section II-B gives a control frequency of 30 Hz. Section III-A, however, specifies a transformer sequence length of 64, which at 30 Hz corresponds to about 2.13 seconds of history, not 1 second. Please clarify the sampling rate of the adaptation module's input and the exact context length used for both the transformer and the CNN baseline. If the effective context is longer than 1 second, the 'online' adaptation claim and the fairness of comparing the two modules' information access need to be revisited.
[Table II and Section III-B] Table II reports point estimates and standard deviations but no confidence intervals, no number of trials for each metric, and no repeated training seeds. The success-rate differences (e.g., 95% vs 86% on CF, 91% vs 85% on UAV1, 96% vs 87% on UAV2) are each based on 100 episodes, and without a confidence interval or significance test the claimed advantage cannot be distinguished from sampling noise. Please provide bootstrap confidence intervals for success rate, RMSE, and max error, and ideally repeat the two-phase training with multiple seeds to show that the transformer advantage is not seed-dependent.

minor comments (5)

[Table I] The 'Fault Range' row should be explicitly defined (e.g., as the allowed λ_i interval, or as the loss fraction 1−λ_i, or as the achieved thrust loss 1−λ_i^2). The current label is ambiguous given the notation in Eq. (2).
[Section III-B] The success definition 'maintaining a positional error within 0.173 m for at least one second' should specify the episode length, whether the one-second window must be contiguous, and whether success is evaluated at any point during the episode or at the end.
[Table II] The row labeled 'No Fault (oracle)' is not an oracle in the usual sense; it is a no-fault run of the same hybrid controller. Please rename it 'No Fault' or explicitly explain why it constitutes an upper bound.
[Section III-A] The PPO and ADAM hyperparameters (learning rates, batch sizes, training steps, dataset size for Phase 2) are not reported; adding them would substantially improve reproducibility.
[Related Work and Section III-A] Since references [8] and [9] are RMA-style quadrotor controllers, they should be included as baselines (or at least discussed quantitatively) rather than only the legged-robot RMA [7]; the phrase 'existing adaptation methods' in the abstract is broader than the single CNN baseline evaluated.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the main success and RMSE claims are direct simulation measurements, and the latent-inference RMSE is a training objective reported as an internal check, not as a derived prediction.

full rationale

The paper's central claims are empirical: the 95% success rate and 0.129 m RMSE are measured directly in PyBullet simulations under fault conditions (Section III-B, Table II), and the comparison against RL-RMA and PID is an experimental benchmark rather than a derivation. The teacher-student adaptation pipeline is standard: the privileged encoder is trained jointly with the policy in Phase 1, and the adaptation module is trained in Phase 2 to minimize the same mean-squared-error objective that is later reported in the latent-inference subsection. Reporting the training objective as 'inference accuracy' is an internal consistency check, not a circular proof of the deployment result, because the deployment success does not reduce to that error and is independently measured. No load-bearing self-citation or imported uniqueness theorem appears; the cited RMA, transformer, and simulator references are external prior work. The apparent mismatch between the stated fault range (lambda_i in [0.8,0.9]), Table I's 'Fault Range [0.1,0.2]', and the '30% LoE' caption is an experimental-design or reporting inconsistency, not a circularity: it does not make any claimed output equal to an input by construction. Likewise, the success threshold matching the reward's sparse bonus is a standard RL evaluation practice, not a self-definitional prediction. Overall, the derivation chain is self-contained and experimentally verified within the paper's stated simulation setup.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the teacher-student latent adaptation premise rather than on derived physics, and all reported numbers depend on hand-set reward weights, success thresholds, fault range, and the 20% command cap, none of which are ablated.

free parameters (5)

Reward coefficients k_pos, k_att, k_vel, k_rate, k_smooth = 0.5, 0.1, 0.001, 0.5, 0.1
Hand-set scaling factors in Eq. 3 with no sensitivity analysis reported, so the central training signal depends on these values.
Success thresholds epsilon_goal and epsilon_near = 0.173 m and 0.520 m
Hand-set in Eq. 4, and the headline success metric used in Table II depends directly on these thresholds.
Policy command cap = 20% of nominal command
Section II-B limits the RL correction; the method's ability to compensate a 36% per-motor thrust loss depends on this cap being large enough.
Transformer hyperparameters = model dimension 128, 2 heads, 2 layers, latent dimension 8, sequence length 64
Architecture choices in Section III-A with no ablation beyond replacing the transformer with a CNN baseline.
Fault range = [0.1, 0.2] with one motor affected
Sampling range in Table I defines task difficulty and is reported alongside inconsistent terminology about '30% LoE' and 'up to 36% thrust loss'.

assumptions (4)

domain assumption The privileged latent vector z_t is a sufficient statistic for the policy's control under varied environment parameters.
Phase 1 jointly trains the policy and the privileged encoder in Section II-B; if z_t discards fault-relevant information, the student adaptation module cannot recover it.
domain assumption The gym-pybullet-drones simulator reproduces loss-of-effectiveness fault behavior faithfully enough for the conclusions.
All results are simulated in PyBullet; no flight tests validate the fault model or the control outputs.
domain assumption The 1-second history window contains enough information to identify the latent code.
The adaptation module input is derived from the previous state-action history in Section II-C and Fig. 1, but identifiability is not proven.
standard math Standard RL and supervised learning convergence assumptions for PPO and ADAM training hold.
Training uses Stable Baselines3 PPO and ADAM, with no convergence certificates, which is typical for empirical RL papers.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Reinforcement Learning-based Fault-Tolerant Control for Quadrotor with Online Transformer Adaptation." pith.science (2026). https://pith.science/paper/QVRHYMJS

@misc{pith2026250508223,
  author       = {Pith},
  title        = {Pith review of: Reinforcement Learning-based Fault-Tolerant Control for Quadrotor with Online Transformer Adaptation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QVRHYMJS}},
  note         = {Machine review of arXiv:2505.08223}
}

read the original abstract

Multirotors play a significant role in diverse field robotics applications but remain highly susceptible to actuator failures, leading to rapid instability and compromised mission reliability. While various fault-tolerant control (FTC) strategies using reinforcement learning (RL) have been widely explored, most previous approaches require prior knowledge of the multirotor model or struggle to adapt to new configurations. To address these limitations, we propose a novel hybrid RL-based FTC framework integrated with a transformer-based online adaptation module. Our framework leverages a transformer architecture to infer latent representations in real time, enabling adaptation to previously unseen system models without retraining. We evaluate our method in a PyBullet simulation under loss-of-effectiveness actuator faults, achieving a 95% success rate and a positional root mean square error (RMSE) of 0.129 m, outperforming existing adaptation methods with 86% success and an RMSE of 0.153 m. Further evaluations on quadrotors with varying configurations confirm the robustness of our framework across untrained dynamics. These results demonstrate the potential of our framework to enhance the adaptability and reliability of multirotors, enabling efficient fault management in dynamic and uncertain environments. Website is available at http://00dhkim.me/paper/rl-ftc

Figures

Figures reproduced from arXiv: 2505.08223 by the authors.

**Figure 2.** State comparison between our method and the baselines for each [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** State comparison between our method and the baselines for the [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Position RMSE and success/failure results between our method and [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 8 canonical work pages

[1]

Fault-tolerant predictive control with deep-reinforcement-learning-based torque dis- tribution for four in-wheel motor drive electric vehicles,

H. Deng, Y . Zhao, A.-T. Nguyen, and C. Huang, “Fault-tolerant predictive control with deep-reinforcement-learning-based torque dis- tribution for four in-wheel motor drive electric vehicles,” IEEE/ASME Transactions on Mechatronics , vol. 28, no. 2, pp. 668–680, 2023

work page 2023
[2]

Fault-tolerant trajectory tracking control based on ddpg algorithm for underwater vehicle with propeller faults,

J. Qin, M. Zhong, W. Gai, and Z. Ding, “Fault-tolerant trajectory tracking control based on ddpg algorithm for underwater vehicle with propeller faults,” International Journal of Control, Automation and Systems, vol. 22, no. 4, pp. 1418–1429, 2024

work page 2024
[3]

Transformer- based fault-tolerant control for fixed-wing uavs using knowledge dis- tillation and in-context adaptation,

F. Giral, I. G ´omez, R. Vinuesa, and S. Le-Clainche, “Transformer- based fault-tolerant control for fixed-wing uavs using knowledge dis- tillation and in-context adaptation,” arXiv preprint arXiv:2411.02975 , 2024

arXiv 2024
[4]

Learn-to-recover: Retrofitting uavs with reinforcement learning-assisted flight control under cyber- physical attacks,

F. Fei, Z. Tu, D. Xu, and X. Deng, “Learn-to-recover: Retrofitting uavs with reinforcement learning-assisted flight control under cyber- physical attacks,” in 2020 IEEE International Conference on Robotics and Automation (ICRA) , 2020, pp. 7358–7364

work page 2020
[5]

A novel hybrid approach for fault-tolerant control of uavs based on robust reinforce- ment learning,

Y . Soh `ege, M. Qui ˜nones-Grueiro, and G. Provan, “A novel hybrid approach for fault-tolerant control of uavs based on robust reinforce- ment learning,” in 2021 IEEE International Conference on Robotics and Automation (ICRA) , 2021, pp. 10 719–10 725

work page 2021
[6]

Reinforcement learning- based fault-tolerant control for quadrotor uavs under actuator fault,

X. Liu, Z. Yuan, Z. Gao, and W. Zhang, “Reinforcement learning- based fault-tolerant control for quadrotor uavs under actuator fault,” IEEE Transactions on Industrial Informatics , 2024

work page 2024
[7]

Rma: Rapid motor adaptation for legged robots,

A. Kumar, Z. Fu, D. Pathak, and J. Malik, “Rma: Rapid motor adaptation for legged robots,” arXiv preprint arXiv:2107.04034 , 2021

arXiv 2021
[8]

Learning a single near-hover position controller for vastly different quadcopters,

D. Zhang, A. Loquercio, X. Wu, A. Kumar, J. Malik, and M. W. Mueller, “Learning a single near-hover position controller for vastly different quadcopters,” in 2023 IEEE International Conference on Robotics and Automation (ICRA) , 2023, pp. 1263–1269

work page 2023

Show all 16 references

[9]

A learning-based quadcopter controller with extreme adap- tation,

D. Zhang, A. Loquercio, J. Tang, T.-H. Wang, J. Malik, and M. W. Mueller, “A learning-based quadcopter controller with extreme adap- tation,” arXiv preprint arXiv:2409.12949 , 2024

2024 arXiv
[10]

Reforma: Robust reinforcement learning via adaptive adversary for drones flying under disturbances,

H.-L. Hsu, H. Meng, S. Luo, J. Dong, V . Tarokh, and M. Pajic, “Reforma: Robust reinforcement learning via adaptive adversary for drones flying under disturbances,” in 2024 IEEE International Con- ference on Robotics and Automation (ICRA) , 2024, pp. 5169–5175

2024
[11]

Pybullet, a python module for physics sim- ulation for games, robotics and machine learning,

E. Coumans and Y . Bai, “Pybullet, a python module for physics sim- ulation for games, robotics and machine learning,” http://pybullet.org, 2016–2021

2016
[12]

Learning to fly—a gym environment with pybullet physics for reinforcement learning of multi-agent quadcopter control,

J. Panerati, H. Zheng, S. Zhou, J. Xu, A. Prorok, and A. P. Schoellig, “Learning to fly—a gym environment with pybullet physics for reinforcement learning of multi-agent quadcopter control,” in 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 20...

2021
[13]

Attention is all you need,

A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” Advances in neural information processing systems , vol. 30, 2017

2017
[14]

Proximal policy optimization algorithms,

J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov, “Proximal policy optimization algorithms,” arXiv preprint arXiv:1707.06347, 2017

2017 arXiv
[15]

Adam: A method for stochastic optimiza- tion,

D. P. Kingma and J. Ba, “Adam: A method for stochastic optimiza- tion,” arXiv preprint arXiv:1412.6980 , 2014

2014 arXiv
[16]

Stable-baselines3: Reliable reinforcement learning im- plementations,

A. Raffin, A. Hill, A. Gleave, A. Kanervisto, M. Ernestus, and N. Dormann, “Stable-baselines3: Reliable reinforcement learning im- plementations,” Journal of machine learning research , vol. 22, no. 268, pp. 1–8, 2021

2021

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Fault-tolerant predictive control with deep-reinforcement-learning-based torque dis- tribution for four in-wheel motor drive electric vehicles,

H. Deng, Y . Zhao, A.-T. Nguyen, and C. Huang, “Fault-tolerant predictive control with deep-reinforcement-learning-based torque dis- tribution for four in-wheel motor drive electric vehicles,” IEEE/ASME Transactions on Mechatronics , vol. 28, no. 2, pp. 668–680, 2023

work page 2023

[2] [2]

Fault-tolerant trajectory tracking control based on ddpg algorithm for underwater vehicle with propeller faults,

J. Qin, M. Zhong, W. Gai, and Z. Ding, “Fault-tolerant trajectory tracking control based on ddpg algorithm for underwater vehicle with propeller faults,” International Journal of Control, Automation and Systems, vol. 22, no. 4, pp. 1418–1429, 2024

work page 2024

[3] [3]

Transformer- based fault-tolerant control for fixed-wing uavs using knowledge dis- tillation and in-context adaptation,

F. Giral, I. G ´omez, R. Vinuesa, and S. Le-Clainche, “Transformer- based fault-tolerant control for fixed-wing uavs using knowledge dis- tillation and in-context adaptation,” arXiv preprint arXiv:2411.02975 , 2024

arXiv 2024

[4] [4]

Learn-to-recover: Retrofitting uavs with reinforcement learning-assisted flight control under cyber- physical attacks,

F. Fei, Z. Tu, D. Xu, and X. Deng, “Learn-to-recover: Retrofitting uavs with reinforcement learning-assisted flight control under cyber- physical attacks,” in 2020 IEEE International Conference on Robotics and Automation (ICRA) , 2020, pp. 7358–7364

work page 2020

[5] [5]

A novel hybrid approach for fault-tolerant control of uavs based on robust reinforce- ment learning,

Y . Soh `ege, M. Qui ˜nones-Grueiro, and G. Provan, “A novel hybrid approach for fault-tolerant control of uavs based on robust reinforce- ment learning,” in 2021 IEEE International Conference on Robotics and Automation (ICRA) , 2021, pp. 10 719–10 725

work page 2021

[6] [6]

Reinforcement learning- based fault-tolerant control for quadrotor uavs under actuator fault,

X. Liu, Z. Yuan, Z. Gao, and W. Zhang, “Reinforcement learning- based fault-tolerant control for quadrotor uavs under actuator fault,” IEEE Transactions on Industrial Informatics , 2024

work page 2024

[7] [7]

Rma: Rapid motor adaptation for legged robots,

A. Kumar, Z. Fu, D. Pathak, and J. Malik, “Rma: Rapid motor adaptation for legged robots,” arXiv preprint arXiv:2107.04034 , 2021

arXiv 2021

[8] [8]

Learning a single near-hover position controller for vastly different quadcopters,

D. Zhang, A. Loquercio, X. Wu, A. Kumar, J. Malik, and M. W. Mueller, “Learning a single near-hover position controller for vastly different quadcopters,” in 2023 IEEE International Conference on Robotics and Automation (ICRA) , 2023, pp. 1263–1269

work page 2023