Pith. sign in

REVIEW 4 major objections 5 minor 25 references

Spatiotemporal Agility: Time-Constrained Reinforcement Learning for Vision-Guided Dynamic Quadrupedal Interception

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a quadruped robot catches thrown balls more reliably when its reinforcement-learned policy is conditioned on the predicted landing point and time-to-impact rather than on velocity commands.

desk verdict Solid system paper with a genuinely time-conditioned policy interface, but the real-world throwing protocol doesn't back the workspace-wide success-rate claim in the abstract. read the letter →

arxiv 2608.06907 v1 pith:WP3LHJJN submitted 2026-08-07 cs.RO

Yidong Zhu , Zibo Dai , Tongning Zhang , Leixin Chang , Hua Chen This is my paper · ORCID
classification cs.RO
keywords quadrupedlocomotionballcatchingtime-constrainedinterceptionreinforcementlearningsim-to-realtransfervision-guidedcontroltrajectorypredictionposition-conditionedpolicy
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

Ball catching is a sharp test of quadruped agility: the robot must perceive a flying ball, predict where and when it will hit the ground, and get there in time. This paper claims that the right way to command that motion is not a velocity setpoint but the spatiotemporal target itself—the predicted landing position and the remaining time until impact. The authors assemble this into a complete system with off-board cameras, a Kalman filter for landing-point and time prediction, a teacher-student RL policy trained with time-gated rewards, and deployment on a Unitree Go2 with an onboard inference computer. In their simulation and real-world trials, the position-and-time-conditioned policy catches more balls than a velocity-tracking baseline, and it shows a smaller drop after sim-to-real transfer.

What carries the argument

The load-bearing mechanism is a time-gated, position-conditioned reward that couples spatial accuracy with arrival time: $r_{\mathrm{pos,time}} = M(t_{\mathrm{go}}, D)\, [1 + (\lVert p_{\mathrm{target}} - p_{\mathrm{robot}}\rVert / \sigma_p)^2]^{-1}$, where the duration mask $M(t_{\mathrm{go}}, D) = \frac{1}{D}\mathbb{I}(t_{\mathrm{go}} \le D)$ switches the reward on only when the remaining time is within the interception window. The policy's observation is the target position error $(\Delta x, \Delta y)$, orientation error $\Delta \theta$, normalized remaining time $t_t$, proprioception, and a short action history. Training uses concurrent teacher-student distillation in IsaacGym, with domain randomization, curriculum expansion of target distances, and RND exploration; the deployed student policy sees only the target commands and proprioception. The perception module supplies the predicted landing point and time through YOLO-based detection with an HSV fallback and a Kalman filter under a parabolic motion model.

What would settle it

Run both policies against one fixed set of throws sampled uniformly across the 0–2 m range with identical flight-time distribution and fixed robot start; if the target-conditioned policy does not beat the velocity-tracking baseline on that common distribution, the paper's central success-rate claim fails in the stricter sense.

Watch

Extended reading notes

Core claim

The paper's central claim is that conditioning a legged locomotion policy on the predicted landing position and time-to-impact—rather than on an intermediate velocity command—produces better dynamic interception for landing spots within 2 m and flight times of roughly 0.8–1.2 s. In simulation, the time-conditioned target-reaching policy reaches 86.73% catch success for targets within 0.5 m and 35.63% for 1–2 m, whereas the velocity-tracking baseline reaches 67.39% and 16.28% in those bins; the baseline is better in the middle band. In real trials on a Unitree Go2, the proposed policy's strict catch rates are 37.5%, 24.4%, and 36.8% across the three distance bins, against 6.25%, 9.62%, and 25.0% for the baseline, while its track (rim-contact) rates stay above 60%. The authors interpret these results as showing that explicit spatiotemporal conditioning, together with a time-gated reward, yields a rotation-prioritized, time-synchronized motion that degrades less under sim-to-real transfer. They are careful to describe the real comparison as feasibility-oriented, with throws placed near each method's feasible catching zone.

Load-bearing premise

The real-world comparison assumes that throwing the ball into each method's own feasible catching zone is a fair basis for comparing success rates; the paper states the trials were feasibility-oriented rather than a fully standardized workspace-wide benchmark.

Editorial extensions

If this is right

  • A quadruped can perform time-constrained interception with a roughly one-second horizon when commanded directly with predicted landing position and remaining time, instead of velocity setpoints.
  • Time-gated rewards produce a distinct motion strategy—turn first, then sprint—that outperforms lateral-sliding velocity tracking at short and long catch distances in the paper's simulation tests.
  • Explicit target-space commands transfer more cleanly to hardware: in real trials the strict catch rate of the proposed policy stays between 24% and 38% while the velocity-tracking baseline falls below 10% at short range.
  • Because the policy receives predicted landing point and time rather than live ball kinematics, brief detection dropouts are bridged by the Kalman filter without requiring the network to learn ballistic dynamics implicitly.

Reading between the lines

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

  • If time conditioning is the active ingredient, then removing the $t_t$ observation or the duration mask should measurably reduce catch rates; the paper does not run this ablation, so the causal role of the timing signal is an inference, not a demonstrated fact.
  • The same target-space interface—predicted position plus time-to-impact—could be applied to other dynamic tasks such as timed jumping, interception with a manipulator, or dodging, since the policy is not tied to ball-specific visual features.
  • The paper's own feasibility-oriented real-world protocol means the reported advantage is not yet a statement about general superiority; a uniform throw distribution across the whole workspace would be a stricter test.
  • The observed rotation-prioritized gait emerged without an explicit velocity-direction reward, suggesting that time-constrained RL can discover nontrivial motor strategies; whether this is robust across terrain and robot models remains open.
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 / 5 minor

Summary. The paper proposes a framework for quadrupedal ball catching in which off-board cameras detect the ball and the robot, a Kalman filter predicts the ball's landing position and landing time, and a reinforcement-learned locomotion policy is conditioned directly on the position error and remaining time rather than on intermediate velocity commands. The policy is trained with a teacher-student distillation scheme in IsaacGym and deployed on a Unitree Go2. The authors compare their method against a velocity-tracking baseline in simulation (MuJoCo) and in real-world trials, and report higher catch success rates for their method in the real-world comparison. The paper also introduces a time-aware reward with a duration mask and observes a rotation-prioritized interception behavior. The manuscript is candid about several limitations, including different flight-time ranges between simulation and real trials, a non-standardized real-world throwing protocol, and the absence of ablations isolating time conditioning, distillation, reward terms, and prediction error.

Significance. If the central comparative claim were conclusively established, the paper would make a useful system-level contribution: it demonstrates a closed-loop pipeline that combines perception, prediction, and time-conditioned locomotion on real hardware, and it provides evidence that direct position-and-time conditioning can be a practical alternative to velocity tracking for time-constrained interception. The paper also explicitly frames its contributions as system-level rather than claiming a fundamentally new learning algorithm, which is appropriate. The manuscript deserves credit for its honest acknowledgment of limitations, particularly in Sections 4.2 and 4.3, where the authors state that the simulation and real-world evaluations are not strictly like-for-like and that the real-world comparison is feasibility-oriented. However, the same acknowledged limitations directly affect the abstract's headline claim: the claimed higher success rate for landing points within 2 meters and flight times of 0.8-1.2 seconds is supported only by a real-world comparison in which throws were placed near each method's feasible catching zone rather than from one shared distribution.

major comments (4)
  1. [§4.3, Table 4] The real-world comparison does not use a shared throwing distribution for the two methods. The text states that throws were placed 'near their feasible catching zones' for each method and that the comparison is 'feasibility-oriented rather than a fully standardized workspace-wide benchmark.' Consequently, the higher strict catch rates reported in Table 4 could reflect easier throws for the proposed method (e.g., front-lateral landing points) rather than a genuine policy advantage. Since the abstract makes a claim about the full 2-meter workspace, the authors should either provide a standardized throwing protocol with the same distribution for both methods or explicitly restrict the claim to the feasibility-oriented setup. At minimum, per-bin trial counts and confidence intervals are needed to assess whether the observed gaps are meaningful.
  2. [§4.2 vs §4.3] The simulation and real-world evaluations use different flight-time ranges: the simulation uses 1.15-1.35 seconds, while the real-world trials use 0.8-1.2 seconds. In addition, the simulation success criterion counts a ball that bounces out of the basket as a success, whereas the real-world strict 'catch' criterion requires the ball to directly hit the basket bottom. These differences mean that the abstract's specified regime (0.8-1.2 seconds flight time) is only evaluated in the non-standardized real-world test, and statements about sim-to-real gaps or improved sim-to-real behavior may conflate different task definitions. The authors should align these definitions or clearly separate the claims made for each regime.
  3. [Table 4] Per-bin trial counts are not reported, and with only 100 total trials per method spread across three distance bins, the reported percentages may be based on very small denominators. For example, in the strict catch row for the 1-2 m bin, the gap between 25.00% and 36.84% would not be significant with typical bin sizes, and even the larger gaps in other bins lose strength if the effective sample size is small. The authors should report the number of trials per bin and, ideally, bootstrap confidence intervals or a statistical test for the success-rate differences.
  4. [§4.3] The deployed velocity-tracking baseline is not identical to the simulation baseline: the real-world experiments use Kp=5.0 for distances below 0.5 m instead of the simulation-tuned Kp=2.5, and both methods receive a shared yaw-command shaping rule. While these changes are disclosed and the yaw shaping is applied to both methods, the baseline policy used in the real-world comparison is therefore not the same policy that was tuned and evaluated in the simulation. The authors should justify that these modifications do not bias the comparison, or provide results with an unmodified baseline, so that the real-world advantage can be attributed to the proposed conditioning rather than to a retuned baseline.
minor comments (5)
  1. [Abstract] The abstract contains grammatical issues, such as 'this work presents a system-level contribution that completes real-time robotic interception system,' and would benefit from careful proofreading.
  2. [§3.2.2, Eq. (4)] The notation for the duration mask is confusing: the mask is defined as M(tgo, D) with a duration threshold D, but the reward uses D_rew without a clear definition, and the reward weight for r_pos_time in Table 1 is not included in Eq. (4). Please clarify the relationship between D, D_rew, and the reward weight.
  3. [§4.3, Figure 8] Because the real-world landing-point distributions are shown separately for the two methods and the throwing protocol intentionally differs, the figure should be accompanied by a clear statement of how the distributions were generated and how many throws fell into each distance bin.
  4. [§5] The conclusion acknowledges that the experiments do not isolate the effects of time conditioning, teacher-student distillation, reward terms, or prediction error. This is an important confound and should be stated earlier in the experimental sections, not only in the conclusion, so that readers can properly interpret the comparisons.
  5. [§1] The introductory paragraph contains citation formatting errors, for example '[ 1, 2], use perception-based methods to enhance...' which reads as an incomplete sentence. Please fix the sentence structure and citation placement.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation: the 'prediction' is an externally evaluated catch rate, not a fitted quantity; self-citations are non-load-bearing.

full rationale

The paper's central claim is empirical: the position-and-time-conditioned policy achieves higher success rates than a velocity-tracking baseline in catching a ball (Table 4). Success is defined externally by whether the ball lands in the basket, not by the training reward (Table 1), so there is no self-definitional reduction. The predicted landing point and time from the Kalman filter (Sec. 3.1.3) enter the policy observation, but the evaluation measures actual interception, so the prediction is not a fitted input renamed as an outcome. No equation equates the claimed result with an input by construction; r_pos_time is a training objective, not the success metric. The paper does rely on prior work by overlapping authors: the CTS distillation framework [22] and reward terms [24]. These are used as architectural/training priors, explicitly described as 'a practical sim-to-real training choice rather than as a claim that direct PPO is fundamentally unsuitable,' and no uniqueness theorem or forced choice is imported from those citations. The real-world comparison is weakened by the paper's own admission that throws were placed 'near their feasible catching zones' per method and that the comparison is 'feasibility-oriented rather than a fully standardized workspace-wide benchmark' (Sec. 4.3). This is an external-validity limitation, not circularity, and the abstract's workspace-wide phrasing should be read with this caveat. Overall, the derivation chain is self-contained: perception, prediction, RL training, and evaluation are distinct, and the reported advantage is a measurable result that could in principle fail.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central claim rests mainly on experimental design choices rather than mathematical derivation. The free parameters are RL reward and training hyperparameters, common in learned control systems. The most consequential assumption is the feasibility-oriented comparison in Section 4.3, which weakens the comparative claim.

free parameters (6)
  • sigma_p (position-time reward scale) = not stated
    In Eq. 4, r_pos_time uses 1/(1 + (||p_target - p_robot||/sigma_p)^2); sigma_p controls how sharply spatial error is penalized and is not reported.
  • D (duration threshold in time-gated mask) = not stated
    M(t_go, D) = (1/D) * I(t_go <= D); D sets the urgency window for the time-aware reward and is not reported.
  • Reward weight vector W (Table 1) = listed in Table 1
    Twenty-plus weights balancing task, stability, energy, and exploration terms; values are hand-chosen and not fitted, but they shape the learned catching behavior.
  • Curriculum command ranges = initially [-0.5,1.0] m and [-1.5,1.5] rad, expanding to [-5.0,7.5] m and [-pi,pi] rad
    Sampling ranges and automatic expansion schedule in Section 3.2.3 are manually designed without sensitivity analysis.
  • Observation history length H = 5
    History length of proprioceptive observations chosen in Section 3.2.1 to trade off delay and information.
  • RND exploration reward coefficient = r_e weight = 1 in Table 1
    Intrinsic reward scale is a hand-picked hyperparameter and its isolated effect is not studied.
assumptions (6)
  • domain assumption Ball flight follows a parabolic constant-gravity model with position measurement noise (Eq. 3)
    The Kalman filter in Section 3.1.3 assumes a linear constant-velocity-plus-gravity model; air drag, spin, and wind are neglected, which is reasonable for low-speed tennis balls but not validated quantitatively.
  • standard math Linear-Gaussian Kalman filter assumptions hold
    Eq. 3 uses standard Kalman filtering; process and measurement noise covariances are not reported.
  • domain assumption Simulator dynamics and domain randomization approximate real hardware
    Section 3.2 and Table 2 assume IsaacGym and MuJoCo models plus randomization spans the sim-to-real gap; no quantitative validation of these ranges is provided.
  • domain assumption Teacher-student distillation transfers from privileged to proprioceptive observations
    Section 3.2 follows [22] as a practical sim-to-real choice; no isolation of distillation effects is performed.
  • domain assumption AprilTag localization and camera calibration are accurate within the workspace
    Section 3.1.1 relies on external cameras and AprilTags for robot and ball position; calibration errors are not quantified.
  • ad hoc to paper The method-specific feasible-zone throwing design is an acceptable basis for comparison
    Section 4.3 uses per-method feasible catching zones, which the paper itself notes makes the comparison feasibility-oriented rather than a standardized benchmark.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Spatiotemporal Agility: Time-Constrained Reinforcement Learning for Vision-Guided Dynamic Quadrupedal Interception." pith.science (2026). https://pith.science/paper/WP3LHJJN

@misc{pith2026260806907,
  author       = {Pith},
  title        = {Pith review of: Spatiotemporal Agility: Time-Constrained Reinforcement Learning for Vision-Guided Dynamic Quadrupedal Interception},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WP3LHJJN}},
  note         = {Machine review of arXiv:2608.06907}
}
read the original abstract

Legged robots require robust agility to perceive and interact with complex and dynamic environments within a constrained time. However, most existing quadruped locomotion works rely on velocity-tracking policy, which struggle to reach precise targets within strict temporal constraints. Moreover, integrating real-time perception with agile locomotion for highly dynamic targets remains challenging due to sensor latency and processing delays. To concretely study and benchmark such agility in dynamic settings, we introduce a challenging ball-catching task for legged robots. This paper proposes an integrated framework that combines a vision module for landing point and time prediction with a direct position and time conditioned RL locomotion policy, instead of intermediate velocity commands. Beyond the method design, this work presents a system-level contribution that completes real-time robotic interception system that integrates multi-camera perception, online trajectory prediction, low-latency target communication, and sim-to-real locomotion control into a closed-loop deployment pipeline. By explicitly predicting the future spatial-temporal target, our approach mitigates perception latency during dynamic interception. We conducted extensive ball-catching experiments for the legged robot. Through comparative experiments against a velocity-tracking baseline, our direct target-conditioned approach achieves a higher success rate in catching balls with predicted landing spots within 2 meters and flight times between 0.8 and 1.2 seconds. This shows that the robot has successfully completed the dynamic ball-catching task under our tested setup. Furthermore, our policy exhibits a smaller performance gap after deployment, suggesting improved sim-to-real behavior in these trials.

Figures

Figures reproduced from arXiv: 2608.06907 by the authors.

Figure 1
Figure 1. System Overview. Our system comprises a visual part and an RL part. The visual part gets the RGB input of the ball and the April-tag on the back of the robot, and outputs the position of the robot and the estimated ball landing position and time. Then the RL part gets that information as policy input to send the action to the robot camera to world frame is defined as pworld = T world cam · pcam (1) To deal with the … view at source ↗
Figure 2
Figure 2. Overview of the proposed position-conditioned local navigation policy training framework. [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Experiment Setup An overhead RGB fisheye camera, calibrated for undistortion, provides a global view for localizing the robot via an AprilTag on its back. A second Intel RealSense D415 camera is positioned near the ball’s release point, with its field of view encompassing the flight path and the base coordinate system defined by the AprilTag. The whole experiment setup is illustrated in [PITH_FULL_IMAGE:figures/ful… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Sim-to-real tennis ball catching experiment [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Comparison of angular and lateral velocities. Ours executes a rapid turn with minimal [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: Comparison of Time Ratio (tactual/tpredicted) across catching distances. Time ratios < 1.0, = 1.0, and > 1.0 signify preemptive agile arrival, perfect synchronization, and late arrival, respectively 11 [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: Comparison of different motion patterns. Our method demonstrates a rotation-prioritized [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]
Figure 8
Figure 8. Figure 8: Distribution of landing points. Comparison of landing positions for our method (left) and [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 20 canonical work pages

  1. [1]

    Event- based agile object catching with a quadrupedal robot

    Benedek Forrai, Takahiro Miki, Daniel Gehrig, Marco Hutter, and Davide Scaramuzza. Event- based agile object catching with a quadrupedal robot. In2023 IEEE International Conference on Robotics and Automation (ICRA), pages 12177–12183, 2023

    work page 2023

  2. [2]

    Percep- tion for humanoid robots.Current Robotics Reports, 4:127–140, 2023

    Arindam Roychoudhury, Shahram Khorshidi, Subham Agrawal, and Maren Bennewitz. Percep- tion for humanoid robots.Current Robotics Reports, 4:127–140, 2023

    work page 2023

  3. [3]

    Learning agile locomotion on risky terrains

    Chong Zhang, Nikita Rudin, David Hoeller, and Marco Hutter. Learning agile locomotion on risky terrains. In2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 11864–11871, 2024

    work page 2024

  4. [4]

    Advanced skills by learning locomotion and local navigation end-to-end

    Nikita Rudin, David Hoeller, Marko Bjelonic, and Marco Hutter. Advanced skills by learning locomotion and local navigation end-to-end. In2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 2497–2503, 2022. 15

    work page 2022

  5. [5]

    Extreme parkour with legged robots

    Xuxin Cheng, Kexin Shi, Ananye Agarwal, and Deepak Pathak. Extreme parkour with legged robots. In2024 IEEE International Conference on Robotics and Automation (ICRA), pages 11443–11450, 2024

    work page 2024

  6. [6]

    Agile but safe: Learning collision-free high-speed legged locomotion

    Tairan He, Chong Zhang, Wenli Xiao, Guanqi He, Changliu Liu, and Guanya Shi. Agile but safe: Learning collision-free high-speed legged locomotion. arXiv:2401.17583, 2024

    arXiv 2024

  7. [7]

    Sim-to-real: Learning agile locomotion for quadruped robots

    Jie Tan, Tingnan Zhang, Erwin Coumans, Atil Iscen, Yunfei Bai, Danijar Hafner, Steven Bohez, and Vincent Vanhoucke. Sim-to-real: Learning agile locomotion for quadruped robots. In Proceedings of Robotics: Science and Systems, Pittsburgh, Pennsylvania, June 2018

    work page 2018

  8. [8]

    Reinforcement learning for versatile, dynamic, and robust bipedal locomotion control.The International Journal of Robotics Research, 44:840–888, 2025

    Zhongyu Li, Xue Bin Peng, Pieter Abbeel, Sergey Levine, Glen Berseth, and Koushil Sreenath. Reinforcement learning for versatile, dynamic, and robust bipedal locomotion control.The International Journal of Robotics Research, 44:840–888, 2025

    work page 2025

Show all 25 references
  1. [9]

    Catch it! learning to catch in flight with mobile dexterous hands

    Yuanhang Zhang, Tianhai Liang, Zhenyang Chen, Yanjie Ze, and Huazhe Xu. Catch it! learning to catch in flight with mobile dexterous hands. In2025 IEEE International Conference on Robotics and Automation (ICRA), pages 14385–14391, 2025

    2025

  2. [10]

    Learning agility and adaptive legged locomotion via curricular hindsight reinforcement learning.Scientific Reports, 14:28089, 2024

    Sicen Li, Gang Wang, Yiming Pang, Panju Bai, Shihao Hu, Zhaojin Liu, Liquan Wang, and Jiawei Li. Learning agility and adaptive legged locomotion via curricular hindsight reinforcement learning.Scientific Reports, 14:28089, 2024

    2024

  3. [11]

    Learning diverse natural behaviors for enhancing the agility of quadrupedal robots

    Huiqiao Fu, Haoyu Dong, Wentao Xu, Zhehao Zhou, Guizhou Deng, Kaiqiang Tang, Daoyi Dong, and Chunlin Chen. Learning diverse natural behaviors for enhancing the agility of quadrupedal robots. arXiv:2505.09979, 2025

    2025 arXiv

  4. [12]

    Impedance matching: Enabling an rl-based running jump in a quadruped robot

    Neil Guan, Shangqun Yu, Shifan Zhu, and Donghyun Kim. Impedance matching: Enabling an rl-based running jump in a quadruped robot. In2024 21st International Conference on Ubiquitous Robots (UR), pages 755–761, 2024

    2024

  5. [13]

    Dynamic object catching with quadruped robot front legs

    Andr´ e Schakkal, Guillaume Bellegarda, and Auke Ijspeert. Dynamic object catching with quadruped robot front legs. In2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 6848–6855, 2024

    2024

  6. [14]

    Anymal parkour: Learning agile navigation for quadrupedal robots.Science Robotics, 9:eadi7566, 2024

    David Hoeller, Nikita Rudin, Dhionis Sako, and Marco Hutter. Anymal parkour: Learning agile navigation for quadrupedal robots.Science Robotics, 9:eadi7566, 2024

    2024

  7. [15]

    Motion priors reimagined: Adapting flat-terrain skills for complex quadruped mobility

    Zewei Zhang, Chenhao Li, Takahiro Miki, and Marco Hutter. Motion priors reimagined: Adapting flat-terrain skills for complex quadruped mobility. In Joseph Lim, Shuran Song, and Hae-Won Park, editors,Proceedings of The 9th Conference on Robot Learning, volume 305 of Proceedings...

    2025

  8. [16]

    Learning whole-body loco-manipulation for omni-directional task space pose tracking with a wheeled-quadrupedal- manipulator.IEEE Robotics and Automation Letters, 10:1481–1488, 2025

    Kaiwen Jiang, Zhen Fu, Junde Guo, Wei Zhang, and Hua Chen. Learning whole-body loco-manipulation for omni-directional task space pose tracking with a wheeled-quadrupedal- manipulator.IEEE Robotics and Automation Letters, 10:1481–1488, 2025

    2025

  9. [17]

    Yolov13: Real-time object detection with hypergraph-enhanced adaptive visual perception

    Mengqi Lei, Siqi Li, Yihong Wu, Han Hu, You Zhou, Xinhu Zheng, Guiguang Ding, Shaoyi Du, Zongze Wu, and Yue Gao. Yolov13: Real-time object detection with hypergraph-enhanced adaptive visual perception. arXiv:2506.17733, 2025. 16

    2025 arXiv

  10. [18]

    Schoellig

    Ke Dong, Karime Pereida, Florian Shkurti, and Angela P. Schoellig. Catch the ball: Accurate high-speed motions for mobile manipulators via inverse dynamics learning. In2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 6718–6725, 2020

    2020

  11. [19]

    Neural motion prediction for in-flight uneven object catching

    Hongxiang Yu, Dashun Guo, Huan Yin, Anzhe Chen, Kechun Xu, Zexi Chen, Minhang Wang, Qimeng Tan, Yue Wang, and Rong Xiong. Neural motion prediction for in-flight uneven object catching. In2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 466...

    2021

  12. [20]

    Run and catch: Dynamic object-catching of quadrupedal robots

    Yangwei You, Tianlin Liu, Xiaowei Liang, Zhe Xu, Mingliang Zhou, Zhibin Li, and Shiwu Zhang. Run and catch: Dynamic object-catching of quadrupedal robots. In2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 743–750, 2023

    2023

  13. [21]

    Proximal policy optimization algorithms

    John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. Proximal policy optimization algorithms. arXiv:1707.06347, 2017

    2017 arXiv

  14. [22]

    Cts: Concurrent teacher-student reinforcement learning for legged locomotion.IEEE Robotics and Automation Letters, 9:9191– 9198, 2024

    Hongxi Wang, Haoxiang Luo, Wei Zhang, and Hua Chen. Cts: Concurrent teacher-student reinforcement learning for legged locomotion.IEEE Robotics and Automation Letters, 9:9191– 9198, 2024

    2024

  15. [23]

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

    Viktor Makoviychuk, Lukasz Wawrzyniak, Yunrong Guo, Michelle Lu, Kier Storey, Miles Macklin, David Hoeller, Nikita Rudin, Arthur Allshire, Ankur Handa, et al. Isaac gym: High performance gpu based physics simulation for robot learning. InThirty-fifth Conference on Neural Infor...

    2021

  16. [24]

    Beyond robustness: Learning unknown dynamic load adaptation for quadruped locomotion on rough terrain

    Leixin Chang, Yuxuan Nai, Hua Chen, and Liangjing Yang. Beyond robustness: Learning unknown dynamic load adaptation for quadruped locomotion on rough terrain. In2025 IEEE International Conference on Robotics and Automation (ICRA), pages 10282–10288, 2025

    2025

  17. [25]

    Exploration by random network distillation

    Yuri Burda, Harrison Edwards, Amos Storkey, and Oleg Klimov. Exploration by random network distillation. InInternational Conference on Learning Representations, 2019. 17

    2019

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.