Pith. sign in

REVIEW 4 major objections 6 minor 57 references

Bipedalism for Quadrupedal Robots: Versatile Loco-Manipulation through Risk-Adaptive Reinforcement Learning

T0 review · 4 major / 6 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 quadruped robot with a risk-adaptive reinforcement-learning policy walks on its hind legs and uses the freed front legs to push carts, probe obstacles, and carry payloads, in simulation and on a real Unitree Go2.

desk verdict Useful real-world bipedal loco-manipulation demo and a plausible training study, but the central uncertainty-adaptive claim is undercut by a missing ablation that would separate the CV signal from a time-decaying risk schedule. read the letter →

arxiv 2507.20382 v1 pith:QNFD5ZQG submitted 2025-07-27 cs.RO cs.LG

Yuyou Zhang , Radu Corcodel , Ding Zhao This is my paper · ORCID
classification cs.ROcs.LG
keywords bipedallocomotionquadrupedalrobotloco-manipulationdistributionalreinforcementlearningrisk-adaptivepolicycoefficientofvariationsim-to-realtransferUnitreeGo2
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

A quadruped robot can walk on its hind legs and use its freed front legs for real manipulation if the reinforcement-learning agent adjusts its own risk preference during training instead of committing to a fixed risk level. The proposed method, risk-adaptive distributional PPO, estimates the distribution of future returns with quantile regression and sets the risk parameter from the coefficient of variation of that distribution, becoming conservative when uncertainty is high and optimistic when it is low. In simulation the adaptive policy beats fixed-risk baselines in velocity tracking and total reward, and the same single policy transfers to a real Unitree Go2 robot for cart pushing, obstacle probing, and payload carrying. The paper argues that this uncertainty-driven risk adaptation is what makes the inherently unstable bipedal gait stable enough for loco-manipulation.

What carries the argument

The load-bearing mechanism is the risk-adaptive distortion of the critic's return distribution. The critic predicts $N$ quantiles $\theta_i(x)$ of the return distribution, and the value used for advantage estimation is computed through Wang's distortion function $g_\alpha(\tau)=\Phi(\Phi^{-1}(\tau)+\alpha)$ instead of the plain mean. The distortion parameter is updated at each training step by $\alpha_t=(\alpha_0-\alpha_T)e^{-t/T}\mathrm{CV}_t+\alpha_T$, where $\mathrm{CV}_t$ is the batch-averaged coefficient of variation of the predicted return distribution. This couples risk preference to the critic's uncertainty: high uncertainty biases the value estimate toward worst-case outcomes, while low uncertainty allows optimism, which the authors argue encourages exploration without sacrificing stability.

What would settle it

Train the same distributional PPO with Eq. (8) replaced by a time-only schedule from $\alpha_0=0$ to $\alpha_T=-0.2$, keeping all other hyperparameters and seeds identical; if this ablation matches or exceeds the reported velocity-tracking reward and success rate, the coefficient-of-variation term is not the cause of the improvement. Varying $\alpha_T$ across $-0.1$, $-0.2$, and $-0.4$ and checking whether the margin over fixed-risk baselines persists would also show whether the result depends on one hand-picked endpoint.

Watch

Extended reading notes

Core claim

The central claim is that adapting the risk level during training to the critic's estimated return uncertainty, rather than presetting it, produces a bipedal gait policy that is both performant and transferable. The agent learns a distribution over returns with quantile regression, applies a distortion risk measure $\rho_{g_\alpha}$ to obtain a risk-adjusted value, and sets the distortion parameter by a law that makes the policy more conservative when the batch-averaged coefficient of variation of the predicted return distribution is high and more optimistic as training progresses. The authors report that this risk-adaptive distributional PPO consistently outperforms fixed-risk distributional PPO variants and standard PPO in velocity tracking and total reward, and that the trained policy transfers to real hardware for cart pushing, contact-aware obstacle probing, and payload carrying without task-specific retraining.

Load-bearing premise

The load-bearing premise is that the average spread of the critic's predicted return, divided by its mean, is a trustworthy signal for how cautious or optimistic the policy should be, and that the hand-chosen endpoints $\alpha_0=0$ and $\alpha_T=-0.2$ work across tasks; if the signal is noisy or the constants are task-specific, the reported advantage could be an artifact of tuning rather than of uncertainty adaptation.

Editorial extensions

If this is right

  • Because risk preference is internalized during training, deployment needs no manual risk-level tuning at test time.
  • A fixed risk-seeking policy fails catastrophically, while the adaptive policy keeps exploration alive and avoids those failures.
  • The policy retains higher success rates and lower tracking error than baselines under external forces, and generalizes to out-of-distribution velocity commands.
  • Three real-world loco-manipulation tasks—cart pushing, contact-aware obstacle probing, and payload carrying—are achieved by a single bipedal locomotion policy without task-specific retraining.

Reading between the lines

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

  • A direct ablation study that removes the coefficient-of-variation term from Eq. (8), leaving only the time-based schedule, would clarify whether the uncertainty adaptation itself rather than the schedule drives the reported gains.
  • The same uncertainty-adaptive risk mechanism could transfer to other unstable loco-manipulation settings, such as carrying a payload while climbing stairs or standing on one support leg, where the critic's return uncertainty may track destabilization.
  • A natural extension is to adapt the endpoint values $\alpha_0$ and $\alpha_T$ themselves, or to use a per-state coefficient of variation instead of a batch average, which could remove the need to hand-pick these constants for each new task.
  • Because the real-world demonstrations are qualitative, a quantitative benchmark of cart-pushing force, obstacle-probing success rate, and payload-carrying stability would allow direct comparison with arm-equipped or single-leg-manipulation baselines.
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.

Referee Report

4 major / 6 minor

Summary. The paper proposes risk-adaptive distributional PPO (DPPO) for quadrupedal robots walking bipedally, with the goal of freeing the front legs for loco-manipulation. The risk parameter alpha in Wang's distortion is updated online according to Eq. (8) as a function of the batch-averaged coefficient of variation (CV) of the critic's predicted return distribution and an exponential time-decay term. The method is evaluated in Isaac Gym against fixed-risk DPPO (neutral, averse, seeking) and standard PPO, with reported improvements in training reward, velocity-tracking error, success rate, and robustness to external forces. The policy is also deployed on a Unitree Go2 to demonstrate cart pushing, obstacle probing, and payload carrying.

Significance. If the risk-adaptive mechanism is genuinely responsible for the reported gains, the paper makes a useful contribution to legged-locomotion RL by replacing manual risk-level selection with an automatically adjusted risk preference, and the real-world bipedal loco-manipulation demos are interesting. The paper includes four baselines, a reward-function ablation, and out-of-distribution velocity evaluation, which are positive elements. However, the central novelty claim hinges on the CV signal in Eq. (8), and the paper currently lacks the control experiment needed to separate that signal from a pure time-decay schedule. The external-force result in Table V is also statistically weak. These issues are fixable with additional experiments, so the paper is not beyond repair.

major comments (4)
  1. [Section V-A and Eq. (8)] The paper does not ablate the coefficient-of-variation term CV_t against a pure time-dependent schedule. In Eq. (8), alpha_t = (alpha0-alphaT)e^{-t/T} CV_t + alphaT, and since CV_t generally decreases during training as the critic improves, alpha_t is strongly correlated with time. The comparison against fixed-risk baselines (alpha = 0, 0.2, -0.2) therefore cannot establish that the uncertainty signal, rather than the exponential annealing, causes the reported advantage. A time-only baseline alpha_t = (alpha0-alphaT)e^{-t/T} + alphaT, or a version with CV_t replaced by a constant or a scrambled signal, is necessary to support the claim that risk adaptiveness is the source of improvement. This is load-bearing because the adaptive risk law is the paper's primary contribution.
  2. [Table V] The external-force success-rate numbers do not support the stated conclusion. DPPOadaptive achieves 0.601 ± 0.418 and DPPOaverse achieves 0.327 ± 0.268; these intervals overlap heavily, so the claim that 'our method achieves the highest success rate, nearly doubling the second-best' is not statistically supported. The paper should report the number of seeds, per-seed values, and a proper confidence interval or statistical test. As written, this table undermines the robustness claim rather than supporting it.
  3. [Section V-A (hyperparameter selection)] The endpoints alpha0 = 0 and alphaT = -0.2 are hand-chosen, and no sensitivity analysis is provided. Because the fixed-risk seeking baseline is set to alpha = -0.2, matching the final adaptive value, the reported advantage could be an artifact of a well-tuned annealing schedule rather than of adaptation to uncertainty. A sweep over (alpha0, alphaT), or at least a comparison with several time-only schedules, is needed to rule out tuning effects.
  4. [Section V-D] The real-world evaluation is entirely qualitative. No task success rates, cart masses, payload weights, disturbance magnitudes, or quantitative tracking errors are reported for cart pushing, obstacle probing, or payload carrying. Since the paper lists 'robust real-world applications' as a main contribution, the current video-based evidence is anecdotal. Quantitative measurements from a small number of repeated trials would make the real-world claim commensurate with the simulation claims.
minor comments (6)
  1. [Equation (2)] The text says 'Generalizable Advantage Estimation'; the standard term is 'Generalized Advantage Estimation', and the ratio expression contains a typo ('at|ott').
  2. [Figure 7 caption] The caption says 'Coefficient of Variance'; the correct term is 'Coefficient of Variation'.
  3. [Section IV-C and Figure 7] Eq. (8) uses the batch-averaged CV (denoted CV_t), but Figure 7 plots a per-timestep CV during evaluation. The relationship between these two quantities and the way alpha is applied at deployment should be clarified.
  4. [Table II and Section V-A] The statement that 'all methods share the same hyperparameters if applicable' is vague because the adaptive method has additional hyperparameters (alpha0, alphaT, and time constant T); the paper should state which hyperparameters are shared and which are method-specific.
  5. [Tables IV and V] Several numbers contain stray spaces (e.g., '0 .337', '64 .11%'), and the 'Success Rate Drop' column should explicitly define the reference success rate used to compute the drop.
  6. [References] Reference [12] appears to have an incomplete title; please verify the bibliographic entry.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the risk-adaptation rule is an explicit algorithmic law and the headline performance claims are benchmarked against independent baselines and real-world trials.

full rationale

The central derivation chain is Eq. (8), alpha_t = (alpha0 - alphaT) exp(-t/T) CV_t + alphaT, which is an explicit control law, not a hidden reuse of the result it is supposed to explain. The claim that alpha responds to uncertainty is definitionally true from Eq. (8), but the paper's actual claims of superiority ('Our method consistently outperforms the baselines in both velocity tracking and total reward', Section V-A) are evaluated against fixed-risk DPPO and PPO in simulation, plus real-world cart pushing, obstacle probing, and payload carrying. These are external benchmarks, not outputs of Eq. (8). The use of CV from the same critic distribution that computes advantages is a self-referential algorithmic loop, but no theorem or prediction is reduced to an input by this loop; it is a training mechanism. The paper contains one self-citation [23] (co-author D. Zhao) in the related-work discussion of claws; it is not load-bearing for any of the paper's central claims. The explicit admission 'we did not fully explore the potential of our method by tuning the initial and final risk levels' and the absence of an ablation replacing CV with a time-only schedule are verification gaps that concern attribution of the improvement, not circularity of the derivation. No step in the paper equates a prediction with a fitted input or imports a conclusion via a self-citation chain.

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

The central claim rests on hand-chosen risk schedule constants (alpha0, alphaT), reward composition constants, an unvalidated assumption that the critic's CV is a reliable uncertainty signal, and an unstated domain randomization setup for sim-to-real transfer. No new physical entities are introduced.

free parameters (4)
  • alpha0 = 0
    Initial Wang distortion parameter. Chosen by hand in Section V-A to maintain stability at the start of training and avoid early catastrophic failures.
  • alphaT = -0.2
    Final Wang distortion parameter. Chosen by hand in Section V-A to limit excessive optimism later; baselines use alpha = ±0.2, matching this final value.
  • Reward composition coefficient c = 0.02
    Used in total reward r+ × e^{c r-}. Stated in Section IV-A, but no sensitivity analysis is provided.
  • Reward scaling factors sigma and sigma_yaw = not stated
    Scale the exponential tracking and upright balance rewards; values are not given in the paper, presumably inherited from [54].
assumptions (5)
  • domain assumption The transition in the POMDP is deterministic, as stated in Section IV-C.
    Used to separate aleatory from epistemic uncertainty and motivate the CV adaptation. In practice the training uses domain randomization, which injects stochastic transitions, making the assumption approximate at best.
  • domain assumption The quantile critic accurately approximates the true return distribution.
    The CV of the critic's predicted return distribution is used to set risk; if the critic is poorly calibrated, the adaptive risk signal is unreliable. The paper provides no calibration analysis.
  • domain assumption Isaac Gym simulation with the unspecified domain randomization setting is faithful enough for zero-shot real-world transfer.
    The real-world demonstrations in Section V-D rely on this. Domain randomization ranges are not reported.
  • standard math Wang's distortion function with standard normal CDF is a valid distortion risk measure.
    Standard result from [56]; used without proof.
  • domain assumption The hand-designed reward functions in Table I induce the intended bipedal behavior and are not trivially gameable.
    Reward design is central to RL training; the ablations only remove whole terms and do not test for reward hacking.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bipedalism for Quadrupedal Robots: Versatile Loco-Manipulation through Risk-Adaptive Reinforcement Learning." pith.science (2026). https://pith.science/paper/QNFD5ZQG

@misc{pith2026250720382,
  author       = {Pith},
  title        = {Pith review of: Bipedalism for Quadrupedal Robots: Versatile Loco-Manipulation through Risk-Adaptive Reinforcement Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QNFD5ZQG}},
  note         = {Machine review of arXiv:2507.20382}
}
read the original abstract

Loco-manipulation of quadrupedal robots has broadened robotic applications, but using legs as manipulators often compromises locomotion, while mounting arms complicates the system. To mitigate this issue, we introduce bipedalism for quadrupedal robots, thus freeing the front legs for versatile interactions with the environment. We propose a risk-adaptive distributional Reinforcement Learning (RL) framework designed for quadrupedal robots walking on their hind legs, balancing worst-case conservativeness with optimal performance in this inherently unstable task. During training, the adaptive risk preference is dynamically adjusted based on the uncertainty of the return, measured by the coefficient of variation of the estimated return distribution. Extensive experiments in simulation show our method's superior performance over baselines. Real-world deployment on a Unitree Go2 robot further demonstrates the versatility of our policy, enabling tasks like cart pushing, obstacle probing, and payload transport, while showcasing robustness against challenging dynamics and external disturbances.

Figures

Figures reproduced from arXiv: 2507.20382 by the authors.

Figure 1
Figure 1. Risk-adaptive distributional RL framework overview for bipedal [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The robot accelerates (a), stays neutral (b), and decelerates (c) [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Learning curves of proposed method (DPPO adaptive) against baselines listed in Table II. The rewards are averaged over three seeds, and the shaded region represents the standard error. Our method consistently outperforms the baselines in both velocity tracking and total reward. Risk-neutral DPPO and PPO perform similarly and both achieve a lower total reward compared to DPPO with adaptive risk. Risk-seeking DPPO fai… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Learning curves of our method and reward function ablations. The [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 6
Figure 6. Figure 6: Snapshots of bipedal loco-manipulation in the real world. From top to bottom, the images showcase the quadrupedal to bipedal transition, bipedal [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Uncertainties represented by the Coefficient of Variance (CV) [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 32 canonical work pages

  1. [54]

    Walk these ways: Tuning robot control for generalization with multiplicity of behavior,

    G. B. Margolis and P. Agrawal, “Walk these ways: Tuning robot control for generalization with multiplicity of behavior,” in Conference on Robot Learning . PMLR, 2023, pp. 22–31

    2023

  2. [1]

    Learning to walk in minutes using massively parallel deep reinforcement learning,

    N. Rudin, D. Hoeller, P. Reist, and M. Hutter, “Learning to walk in minutes using massively parallel deep reinforcement learning,” in Conference on Robot Learning . PMLR, 2022, pp. 91–100

    2022

  3. [2]

    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

  4. [3]

    Anymal parkour: Learning agile navigation for quadrupedal robots,

    D. Hoeller, N. Rudin, D. Sako, and M. Hutter, “Anymal parkour: Learning agile navigation for quadrupedal robots,” Science Robotics , vol. 9, no. 88, p. eadi7566, 2024

    2024

  5. [4]

    Hybrid internal model: Learning agile legged locomotion with simulated robot response,

    J. Long, Z. Wang, Q. Li, L. Cao, J. Gao, and J. Pang, “Hybrid internal model: Learning agile legged locomotion with simulated robot response,” in The Twelfth International Conference on Learning Representations, 2024

    2024

  6. [5]

    Soloparkour: Constrained reinforcement learning for visual locomotion from privileged experience,

    E. Chane-Sane, J. Amigo, T. Flayols, L. Righetti, and N. Mansard, “Soloparkour: Constrained reinforcement learning for visual locomotion from privileged experience,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https://openreview.net/forum?id=DSdAEsEGhE

    work page 2024

  7. [6]

    SLR: Learning quadruped locomotion without privileged information,

    S. Chen, Z. Wan, S. Yan, C. Zhang, W. Zhang, Q. Li, D. Zhang, and F. U. D. Farrukh, “SLR: Learning quadruped locomotion without privileged information,” in 8th Annual Conference on Robot Learning, 2024. [Online]. Available: https://openreview.net/forum?id= RMkdcKK7jq

    work page 2024

  8. [7]

    Learning h-infinity locomotion control,

    J. Long, W. Yu, Q. Li, Z. Wang, D. Lin, and J. Pang, “Learning h-infinity locomotion control,” in 8th Annual Conference on Robot Learning, 2024. [Online]. Available: https://openreview.net/forum?id= uMZ2jnZUDX

    work page 2024

Show all 57 references
  1. [8]

    Gaitor: Learning a unified representation across gaits for real-world quadruped locomotion,

    A. L. Mitchell, W. Merkt, A. Papatheodorou, I. Havoutis, and I. Posner, “Gaitor: Learning a unified representation across gaits for real-world quadruped locomotion,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https://openreview.net/ forum?id=ySI0tBYxpz

    2024

  2. [9]

    TOP-nav: Legged navigation integrating terrain, obstacle and proprioception estimation,

    J. Ren, Y . Liu, Y . Dai, J. Long, and G. Wang, “TOP-nav: Legged navigation integrating terrain, obstacle and proprioception estimation,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https://openreview.net/forum?id=O05tIQt2d5

    2024

  3. [10]

    Robotkeyframing: Learning locomotion with high-level objectives via mixture of dense and sparse rewards,

    F. Zargarbashi, J. Cheng, D. Kang, R. Sumner, and S. Coros, “Robotkeyframing: Learning locomotion with high-level objectives via mixture of dense and sparse rewards,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https: //openreview.net/forum?id=wcbrhPnOei

    2024

  4. [11]

    Generalized animal imitator: Agile locomotion with versatile motion prior,

    R. Yang, Z. Chen, J. Ma, C. Zheng, Y . Chen, Q. Nguyen, and X. Wang, “Generalized animal imitator: Agile locomotion with versatile motion prior,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https://openreview.net/forum?id=9XV3dBqcfe )RUZDUG %DFNZDUG...

    2024

  5. [12]

    UMI-on-legs: Making manipulation policies mobile with a manipulation-centric whole-body controller,

    H. Ha, Y . Gao, Z. Fu, J. Tan, and S. Song, “UMI-on-legs: Making manipulation policies mobile with a manipulation-centric whole-body controller,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https://openreview.net/forum?id=3i7j8ZPnbm

    2024

  6. [13]

    Guided reinforcement learning for robust multi-contact loco-manipulation,

    J. P. Sleiman, M. Mittal, and M. Hutter, “Guided reinforcement learning for robust multi-contact loco-manipulation,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https: //openreview.net/forum?id=9aZ4ehSTRc

    2024

  7. [14]

    Continuously improving mobile manipulation with autonomous real-world RL,

    R. Mendonca, E. Panov, B. Bucher, J. Wang, and D. Pathak, “Continuously improving mobile manipulation with autonomous real-world RL,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https://openreview.net/forum?id=46SluHKoE9

    2024

  8. [15]

    Learning to open and traverse doors with a legged manipulator,

    M. Zhang, Y . Ma, T. Miki, and M. Hutter, “Learning to open and traverse doors with a legged manipulator,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https://openreview.net/forum?id=V oC3wF6fbh

    2024

  9. [16]

    A versatile planner for learning dexterous and whole-body manipulation,

    J. Bruedigam, A. A. Abbas, M. Sorokin, K. Fang, B. Hung, M. Guru, S. G. Sosnowski, J. Wang, S. Hirche, and S. L. Cleac’h, “A versatile planner for learning dexterous and whole-body manipulation,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https://o...

    2024

  10. [17]

    Hierarchical reinforcement learning for precise soccer shooting skills using a quadrupedal robot,

    Y . Ji, Z. Li, Y . Sun, X. B. Peng, S. Levine, G. Berseth, and K. Sreenath, “Hierarchical reinforcement learning for precise soccer shooting skills using a quadrupedal robot,” in 2022 IEEE/RSJ International Confer- ence on Intelligent Robots and Systems (IROS) . IEEE, 2022, pp...

    2022

  11. [18]

    Dribblebot: Dynamic legged manipulation in the wild,

    Y . Ji, G. B. Margolis, and P. Agrawal, “Dribblebot: Dynamic legged manipulation in the wild,” in 2023 IEEE International Conference on Robotics and Automation (ICRA) . IEEE, 2023, pp. 5155–5162

    2023

  12. [19]

    Event- based agile object catching with a quadrupedal robot,

    B. Forrai, T. Miki, D. Gehrig, M. Hutter, and D. Scaramuzza, “Event- based agile object catching with a quadrupedal robot,” in 2023 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2023, pp. 12 177–12 183

    2023

  13. [20]

    Creating a dynamic quadrupedal robotic goalkeeper with reinforcement learning,

    X. Huang, Z. Li, Y . Xiang, Y . Ni, Y . Chi, Y . Li, L. Yang, X. B. Peng, and K. Sreenath, “Creating a dynamic quadrupedal robotic goalkeeper with reinforcement learning,” in 2023 IEEE/RSJ International Con- ference on Intelligent Robots and Systems (IROS) . IEEE, 2023, pp. 2715–2722

    2023

  14. [21]

    Versatile multicontact planning and control for legged loco-manipulation,

    J.-P. Sleiman, F. Farshidian, and M. Hutter, “Versatile multicontact planning and control for legged loco-manipulation,” Science Robotics, vol. 8, no. 81, p. eadg5014, 2023

    2023

  15. [22]

    Deep whole-body control: learning a unified policy for manipulation and locomotion,

    Z. Fu, X. Cheng, and D. Pathak, “Deep whole-body control: learning a unified policy for manipulation and locomotion,” in Conference on Robot Learning. PMLR, 2023, pp. 138–149

    2023

  16. [23]

    Locoman: Advancing versatile quadrupedal dexterity with lightweight loco-manipulators,

    C. Lin, X. Liu, Y . Yang, Y . Niu, W. Yu, T. Zhang, J. Tan, B. Boots, and D. Zhao, “Locoman: Advancing versatile quadrupedal dexterity with lightweight loco-manipulators,” arXiv preprint arXiv:2403.18197, 2024

    2024 arXiv

  17. [24]

    The evolution of human bipedality: ecology and func- tional morphology,

    K. D. Hunt, “The evolution of human bipedality: ecology and func- tional morphology,” Journal of human evolution , vol. 26, no. 3, pp. 183–202, 1994

    1994

  18. [25]

    Models, feedback control, and open problems of 3d bipedal robotic walking,

    J. W. Grizzle, C. Chevallereau, R. W. Sinnet, and A. D. Ames, “Models, feedback control, and open problems of 3d bipedal robotic walking,” Automatica, vol. 50, no. 8, pp. 1955–1988, 2014

    1955

  19. [26]

    Risk-aware robotics: Tail risk measures in planning, control, and verification,

    P. Akella, A. Dixit, M. Ahmadi, L. Lindemann, M. P. Chapman, G. J. Pappas, A. D. Ames, and J. W. Burdick, “Risk-aware robotics: Tail risk measures in planning, control, and verification,” arXiv preprint arXiv:2403.18972, 2024

    2024 arXiv

  20. [27]

    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

  21. [28]

    Efficient exploration via epistemic-risk-seeking pol- icy optimization,

    B. O’Donoghue, “Efficient exploration via epistemic-risk-seeking pol- icy optimization,” in Proceedings of the 40th International Conference on Machine Learning , ser. ICML’23. JMLR.org, 2023

    2023

  22. [29]

    Alma-articulated locomotion and ma- nipulation for a torque-controllable robot,

    C. D. Bellicoso, K. Kr ¨amer, M. St ¨auble, D. Sako, F. Jenelten, M. Bjelonic, and M. Hutter, “Alma-articulated locomotion and ma- nipulation for a torque-controllable robot,” in 2019 International conference on robotics and automation (ICRA) . IEEE, 2019, pp. 8477–8483

    2019

  23. [30]

    A unified mpc framework for whole-body dynamic locomotion and manipula- tion,

    J.-P. Sleiman, F. Farshidian, M. V . Minniti, and M. Hutter, “A unified mpc framework for whole-body dynamic locomotion and manipula- tion,” IEEE Robotics and Automation Letters , vol. 6, no. 3, pp. 4688– 4695, 2021

    2021

  24. [31]

    Visual whole-body control for legged loco-manipulation,

    M. Liu, Z. Chen, X. Cheng, Y . Ji, R.-Z. Qiu, R. Yang, and X. Wang, “Visual whole-body control for legged loco-manipulation,” in 8th Annual Conference on Robot Learning , 2024. [Online]. Available: https://openreview.net/forum?id=cT2N3p1AcE

    2024

  25. [32]

    Asc: Adaptive skill coordination for robotic mobile manipulation,

    N. Yokoyama, A. Clegg, J. Truong, E. Undersander, T.-Y . Yang, S. Ar- naud, S. Ha, D. Batra, and A. Rai, “Asc: Adaptive skill coordination for robotic mobile manipulation,” IEEE Robotics and Automation Letters , vol. 9, no. 1, pp. 779–786, 2023

    2023

  26. [33]

    Learning force control for legged manipulation,

    T. Portela, G. B. Margolis, Y . Ji, and P. Agrawal, “Learning force control for legged manipulation,” arXiv preprint arXiv:2405.01402 , 2024

    2024 arXiv

  27. [34]

    Combining learning-based locomotion policy with model-based manipulation for legged mobile manipulators,

    Y . Ma, F. Farshidian, T. Miki, J. Lee, and M. Hutter, “Combining learning-based locomotion policy with model-based manipulation for legged mobile manipulators,” IEEE Robotics and Automation Letters , vol. 7, no. 2, pp. 2377–2384, 2022

    2022

  28. [35]

    Gamma: Graspability-aware mobile manipulation policy learning based on online grasping pose fusion,

    J. Zhang, N. Gireesh, J. Wang, X. Fang, C. Xu, W. Chen, L. Dai, and H. Wang, “Gamma: Graspability-aware mobile manipulation policy learning based on online grasping pose fusion,” in 2024 IEEE Inter- national Conference on Robotics and Automation (ICRA) . IEEE, 2024, pp. 1399–1405

    2024

  29. [36]

    Learning visual quadrupedal loco-manipulation from demonstrations,

    Z. He, K. Lei, Y . Ze, K. Sreenath, Z. Li, and H. Xu, “Learning visual quadrupedal loco-manipulation from demonstrations,” arXiv preprint arXiv:2403.20328, 2024

    2024 arXiv

  30. [37]

    Hilma-res: A general hierarchical framework via residual rl for combining quadrupedal locomotion and manipulation,

    X. Huang, Q. Liao, Y . Ni, Z. Li, L. Smith, S. Levine, X. B. Peng, and K. Sreenath, “Hilma-res: A general hierarchical framework via residual rl for combining quadrupedal locomotion and manipulation,” arXiv preprint arXiv:2407.06584 , 2024

    2024 arXiv

  31. [38]

    Pedipulate: Enabling manipulation skills using a quadruped robot’s leg,

    P. Arm, M. Mittal, H. Kolvenbach, and M. Hutter, “Pedipulate: Enabling manipulation skills using a quadruped robot’s leg,” in 41st IEEE Conference on Robotics and Automation (ICRA 2024) , 2024

    2024

  32. [39]

    Visual manipulation with legs,

    X. He, C. Yuan, W. Zhou, R. Yang, D. Held, and X. Wang, “Visual manipulation with legs,” in 8th Annual Conference on Robot Learning, 2024. [Online]. Available: https://openreview.net/forum?id= E4K3yLQQ7s

    2024

  33. [40]

    Long- horizon locomotion and manipulation on a quadrupedal robot with large language models,

    Y . Ouyang, J. Li, Y . Li, Z. Li, C. Yu, K. Sreenath, and Y . Wu, “Long- horizon locomotion and manipulation on a quadrupedal robot with large language models,” arXiv preprint arXiv:2404.05291 , 2024

    2024 arXiv

  34. [41]

    Legs as manipulator: Pushing quadrupedal agility beyond locomotion,

    X. Cheng, A. Kumar, and D. Pathak, “Legs as manipulator: Pushing quadrupedal agility beyond locomotion,” in 2023 IEEE International Conference on Robotics and Automation (ICRA) . IEEE, 2023, pp. 5106–5112

    2023

  35. [42]

    Learning agile bipedal motions on a quadrupedal robot,

    Y . Li, J. Li, W. Fu, and Y . Wu, “Learning agile bipedal motions on a quadrupedal robot,” arXiv preprint arXiv:2311.05818 , 2023

    2023 arXiv

  36. [43]

    Leveraging symmetry in rl-based legged locomotion control,

    Z. Su, X. Huang, D. Ordo ˜nez-Apraez, Y . Li, Z. Li, Q. Liao, G. Turrisi, M. Pontil, C. Semini, Y . Wu et al., “Leveraging symmetry in rl-based legged locomotion control,” arXiv preprint arXiv:2403.17320 , 2024

    2024 arXiv

  37. [44]

    Adaptive risk- tendency: Nano drone navigation in cluttered environments with distributional reinforcement learning,

    C. Liu, E.-J. van Kampen, and G. C. De Croon, “Adaptive risk- tendency: Nano drone navigation in cluttered environments with distributional reinforcement learning,” in 2023 IEEE International Conference on Robotics and Automation (ICRA) . IEEE, 2023, pp. 7198–7204

    2023

  38. [45]

    Learning risk-aware quadrupedal locomotion using distributional reinforcement learning,

    L. Schneider, J. Frey, T. Miki, and M. Hutter, “Learning risk-aware quadrupedal locomotion using distributional reinforcement learning,” in 2024 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2024, pp. 11 451–11 458

    2024

  39. [46]

    Learning locomotion for quadruped robots via distributional ensemble actor-critic,

    S. Li, Y . Pang, P. Bai, J. Li, Z. Liu, S. Hu, L. Wang, and G. Wang, “Learning locomotion for quadruped robots via distributional ensemble actor-critic,” IEEE Robotics and Automation Letters , 2024

    2024

  40. [47]

    Robust quadrupedal locomotion via risk-averse policy learning,

    J. Shi, C. Bai, H. He, L. Han, D. Wang, B. Zhao, M. Zhao, X. Li, and X. Li, “Robust quadrupedal locomotion via risk-averse policy learning,” in 2024 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2024, pp. 11 459–11 466

    2024

  41. [48]

    Learning risk-aware costmaps for traversability in challenging environments,

    D. D. Fan, A.-A. Agha-Mohammadi, and E. A. Theodorou, “Learning risk-aware costmaps for traversability in challenging environments,” IEEE robotics and automation letters , vol. 7, no. 1, pp. 279–286, 2021

    2021

  42. [49]

    A distributional per- spective on reinforcement learning,

    M. G. Bellemare, W. Dabney, and R. Munos, “A distributional per- spective on reinforcement learning,” in International conference on machine learning. PMLR, 2017, pp. 449–458

    2017

  43. [50]

    Distributional reinforcement learning with quantile regression,

    W. Dabney, M. Rowland, M. Bellemare, and R. Munos, “Distributional reinforcement learning with quantile regression,” in Proceedings of the AAAI conference on artificial intelligence , vol. 32, no. 1, 2018

    2018

  44. [51]

    Implicit quantile networks for distributional reinforcement learning,

    W. Dabney, G. Ostrovski, D. Silver, and R. Munos, “Implicit quantile networks for distributional reinforcement learning,” in International conference on machine learning . PMLR, 2018, pp. 1096–1105

    2018

  45. [52]

    Efficient risk-averse reinforcement learning,

    I. Greenberg, Y . Chow, M. Ghavamzadeh, and S. Mannor, “Efficient risk-averse reinforcement learning,” Advances in Neural Information Processing Systems, vol. 35, pp. 32 639–32 652, 2022

    2022

  46. [53]

    Estimating risk and uncertainty in deep reinforcement learning,

    W. R. Clements, B. Van Delft, B.-M. Robaglia, R. B. Slaoui, and S. Toth, “Estimating risk and uncertainty in deep reinforcement learning,” arXiv preprint arXiv:1905.09638 , 2019

    1905 arXiv

  47. [55]

    High- dimensional continuous control using generalized advantage estima- tion,

    J. Schulman, P. Moritz, S. Levine, M. Jordan, and P. Abbeel, “High- dimensional continuous control using generalized advantage estima- tion,” arXiv preprint arXiv:1506.02438 , 2015

    2015 arXiv

  48. [56]

    A class of distortion operators for pricing financial and insurance risks,

    S. S. Wang, “A class of distortion operators for pricing financial and insurance risks,” Journal of risk and insurance , pp. 15–36, 2000

    2000

  49. [57]

    Isaac gym: High performance gpu-based physics simulation for robot learning,

    V . Makoviychuk, L. Wawrzyniak, Y . Guo, M. Lu, K. Storey, M. Mack- lin, D. Hoeller, N. Rudin, A. Allshire, A. Handa et al. , “Isaac gym: High performance gpu-based physics simulation for robot learning,” arXiv preprint arXiv:2108.10470 , 2021

    2021 arXiv

Pith tools

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