Pith. sign in

REVIEW 4 major objections 6 minor 2 cited by

ARMOR: Egocentric Perception for Humanoid Robot Collision Avoidance and Motion Planning

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read ARMOR claims that distributed arm-mounted time-of-flight sensors, paired with a transformer imitation policy, cut humanoid collisions by 63.7% over head/external cameras and run 26x faster than sampling-based planning.

desk verdict ARMOR has a genuinely useful systems idea—distributed arm-mounted ToF sensors for humanoid collision avoidance—but the headline cuRobo comparison has an undefined denominator and needs a careful rewrite. read the letter →

arxiv 2412.00396 v1 pith:IZ4K4XJB submitted 2024-11-30 cs.RO cs.LG

Daehwa Kim , Mario Srouji , Chen Chen , Jian Zhang This is my paper · ORCID
classification cs.ROcs.LG
keywords egocentricperceptiontime-of-flightsensorshumanoidrobotcollisionavoidanceimitationlearningtransformerpolicymotionplanningdepthsensorconstellation
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 the standard way of equipping a humanoid robot with vision — a dense camera on the head or mounted outside the body — is the wrong trade for collision avoidance. It introduces ARMOR, a constellation of forty small, low-cost time-of-flight depth sensors distributed across the robot's arms, each returning a coarse 8-by-8 depth image in its own frame, and pairs this perception layer with a transformer policy trained by imitation on human motion data. In simulation, feeding ARMOR's arm-mounted point clouds to the same policy cuts collisions by 63.7% and raises task success by 78.7% relative to feeding it four dense head-height cameras. The paper also reports that the learned policy beats a sampling-based motion planner with 31.6% fewer collisions, 16.9% higher success, and about 26 times less computation. The claim, if it holds, is that sparse, distributed, low-resolution perception placed where collisions actually happen can outperform high-resolution perception placed where the robot can see its hands.

What carries the argument

The carrying object is ARMOR's distributed sensor constellation: forty time-of-flight zone-array lidars mounted twenty per arm, each producing an 8x8 depth image with a 63-degree field of view and streaming at 15 Hz. The software side is ARMOR-Policy, a transformer encoder-decoder built on an action-chunking architecture. It ingests the forty ego-frame depth images through monochannel ResNet encoders, together with current and goal joint positions and a latent style variable z that encodes trajectory diversity, and outputs action sequences for the 14 arm degrees of freedom. At inference the policy samples multiple candidate trajectories in parallel by varying z and selects the one that minimizes the robot-to-point-cloud distance evaluated with a signed distance function.

What would settle it

Run the same ARMOR and four-camera perceptual setups on a test set where obstacles are placed freely at various heights rather than generated around the demonstration trajectories, and compare collision counts and success rates; if ARMOR's advantage shrinks or reverses, the tight-trajectory data generation was the source of the gain.

Watch

Extended reading notes

Core claim

The central claim is that egocentric, distributed, low-resolution depth sensing on a humanoid's arms is a better perceptual substrate for collision avoidance than centralized high-resolution cameras, and that a transformer trained to imitate human motions can exploit it. In the paper's dense-obstacle simulation benchmark built from retargeted human arm motions, ARMOR-Policy reduces collisions by 63.7% and improves success rate by 78.7% over the same policy running on a four-camera exocentric setup that includes a head-mounted camera. Compared with a sampling-based motion planning expert using the same ARMOR perception, the neural policy shows 31.6% fewer collisions, 16.9% higher success, and about 26 times lower computation time. The system was also deployed on a real humanoid with 28 ToF lidars, updating its trajectory at 15 Hz.

Load-bearing premise

The benchmark generates obstacles tightly around the recorded human arm motions, so the arm-mounted sensors almost always see the obstacles directly while head-height cameras are often occluded.

Editorial extensions

If this is right

  • Humanoid robots could rely on body-mounted sensors instead of external cameras for collision avoidance, supporting mobile, untethered operation in dense spaces.
  • The roughly 26x reduction in planning latency enables reactive avoidance at the sensor's 15 Hz update rate, which is fast enough for dynamic environments.
  • Because the policy consumes generic ego-frame depth images, the same trained policy may transfer to different arm geometries or sensor layouts without retraining.
  • The low cost and low profile of the sensors make whole-body coverage economically feasible, effectively giving the robot a coarse artificial skin.
  • Sampling many latent trajectories at inference and picking the one with largest clearance adds a safety margin that would be expensive to obtain with a single deterministic plan.

Reading between the lines

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

  • A natural boundary of the claim is open environments: when obstacles are visible from a head-mounted camera and far from the arms, the exocentric setup may catch up, so the reported advantage should be expected to shrink outside cluttered, arm-near scenes.
  • An ablation that varies the number and mounting locations of the ToF sensors would reveal where the gains come from and how few sensors are still sufficient.
  • The same distributed egocentric sensing could apply to other manipulator morphologies, such as mobile manipulators or legged robots with arms, where head-mounted cameras share the same occlusion blind spots.
  • Because the input images are only 8x8, the policy could likely be distilled into a much smaller network that runs on embedded hardware, enabling closed-loop avoidance onboard without a workstation GPU.
Share X Bluesky LinkedIn Reddit HN

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. This paper introduces ARMOR, a distributed egocentric perception system for humanoid robots based on small ToF lidar sensors mounted on the arms, together with ARMOR-Policy, a transformer-based imitation learning policy for collision avoidance trained on AMASS-derived human motion data. The central claim is that ARMOR perception is superior to head-mounted and externally mounted depth cameras, and that the learned policy is both safer and faster than the sampling-based planner cuRobo. The paper also reports a real-robot deployment on the GR1 humanoid.

Significance. If substantiated, the contribution is practically significant: a low-cost, low-power distributed sensor constellation that reduces occlusion for humanoid arm planning, combined with a data-driven policy, could be a compelling alternative to external cameras or computationally heavy sampling planners. The use of AMASS motion data to generate diverse expert demonstrations is a reasonable design choice, and the real-hardware deployment is a useful proof of concept. However, the evaluation protocol has several issues—the composition of the test set, an undefined comparison denominator for the cuRobo baseline, and the absence of error bars—that prevent the reported improvement margins from being taken at face value. With a corrected evaluation, the system would be a meaningful contribution to the field.

major comments (4)
  1. [§IV.B, Table I(b)] The text states that cuRobo failed to find a solution on 64% of the evaluation data, and that 'we only use the sequences where cuRobo was able to find a solution when comparing the collision avoidance results... for fairness.' This filtering is described only for the collision results, but Table I(b) reports a 16.9% higher success rate for ACT-Depth relative to cuRobo. If success is evaluated on the cuRobo-solved subset, cuRobo's success rate should be near 100% by construction, leaving little room for a 16.9% improvement. If success is evaluated on the full 22,280-sequence test set, then the collision and success metrics use different denominators, and the reported improvement is not well-defined. Please report absolute collision counts and success rates for both policies on the same test set, or clearly state the denominator for each metric.
  2. [§IV.A] The 22,280 test sequences are described as a filtered subset of the 66,840 validation instances ('validated them on another 66,840 instances... This yields 22,280 motion sequences for testing'). If the same instances were used for validation (e.g., early stopping) and final evaluation, the reported success rates are optimistic and do not reflect generalization. The paper should either use a strictly disjoint test set or explicitly state that the validation set was never used for model selection. Additionally, the filtering criterion 'where a solution exists' needs a precise definition: a solution for which planner/policy, and what happens to the excluded sequences in each metric?
  3. [§III.D, §IV.A] In the data-generation scheme, obstacles are placed tightly around the re-targeted AMASS trajectory (Section III.D, Figure 5). Since the test sequences come from the same generator, obstacles are by construction close to the arm-mounted sensors, while head-height exocentric cameras may be occluded by the robot's body or by clutter. This protocol may systematically favor ARMOR over exocentric perception, independent of the sensor hardware. A more informative evaluation would include held-out environments with obstacles not derived from the expert trajectory, and would report whether the ARMOR advantage persists when obstacles are placed away from the arms or when occlusions are reduced.
  4. [§III.D] The paper states that in emergency-stop demonstrations 'the last goal position [is] a random location lying inside of an obstacle, which always leads the arm to collide,' yet these trajectories are included in the demonstration data used to train an imitation learning policy for collision avoidance. The manuscript does not explain how these colliding demonstrations are used: are they treated as positive examples, negative examples, or filtered out? If they are used as positive demonstrations, the policy would be trained to imitate colliding motions; if they are filtered out, the description should say so. Please clarify the role of emergency-stop data in the training set.
minor comments (6)
  1. [Abstract, §IV.C] The abstract and Section IV.C state that the source code, hardware description, and CAD files will be released, but no link is provided in the submitted manuscript. Please include the actual repository/URL or a clear statement of availability in the final version.
  2. [Figure 6 caption] The caption contains a typo: 'VL35L5CX' should be 'VL53L5CX'.
  3. [§I, §IV.A] The paper uses '86 hours' and '86.6 hours' interchangeably for the training data; please standardize the number.
  4. [Table I] The table reports only percentage improvements without absolute values; adding absolute counts (e.g., number of collisions, number of successes out of N) would make the results interpretable.
  5. [§IV.B] The description of point-cloud pruning for cuRobo is ambiguous when using ARMOR perception: 40 sensors at 8x8 resolution produce only 2,560 points, so the 10,000-point cap and the 3.7M points likely refer to the exocentric camera setup. Please clarify which perception configuration the pruning applies to.
  6. [§IV.A] The paper says 'we also simulate the sensor noise' but does not give the noise model used for the VL53L5CX or the D435; specify the noise parameters for reproducibility.

Circularity Check

0 steps flagged · score 2.0 of 10

No load-bearing circularity: the central comparisons are against external baselines (cuRobo, exocentric D435), with only a minor non-load-bearing self-citation in related work.

full rationale

The claimed derivation chain is mostly empirical benchmarking rather than a deduction. ARMOR's headline results compare (a) the same policy under ARMOR vs. exocentric D435 perception and (b) ACT-Depth vs. the external cuRobo planner. Neither comparison defines its metric in terms of the quantity being predicted. Training and evaluation do share the AMASS obstacle-generation pipeline of Section III.D, but the test set is a held-out split from the same generator; this is a distribution/generalization concern, not a circular reduction. The only author-overlap citation is SAFER [27] in Related Work, which is descriptive and not load-bearing. Eq. (1) selects the sampled trajectory with the largest SDF margin, and Table I(c) reports that this reduces collisions; this is an ablation of the method's own objective, not a fitted parameter renamed as a prediction. The cuRobo success-metric denominator ambiguity in Section IV.B is a reporting/correctness issue, not evidence that the IL-vs-sampling comparison reduces to its inputs. No step of the paper equates a fitted input with a predicted output by construction.

Assumptions & free parameters 3 free parameters · 4 assumptions · 1 invented entities

The central comparison depends on several hand-chosen design and data-generation choices: sensor count and placement, obstacle margins, and inference sample count. The most load-bearing assumptions are that AMASS retargeting is a valid expert and that the simulated ToF model matches reality; neither is independently validated here.

free parameters (3)
  • Sensor count and placement (40 simulation, 28 real) = 40 simulated, 28 deployed on GR1
    Chosen by hand with no ablation; placement determines visibility and the paper does not report sensitivity to this choice.
  • Obstacle clearance margin around expert trajectories = Not specified ('tight' obstacles)
    Controls task difficulty and the fairness of the exocentric comparison; no quantitative value or ablation is given.
  • Number of sampled trajectories at inference (N) = Not specified
    Inference-time optimization in Section III.C samples N candidates; N affects collision rate and the 240 ms latency but is not reported.
assumptions (4)
  • domain assumption AMASS human motion, after retargeting, constitutes a valid expert policy for humanoid collision-free motion planning.
    Used in Section III.D to generate all training and evaluation trajectories; if human motion does not transfer to GR1 dynamics, the policy learns from invalid demonstrations.
  • domain assumption The simulated ToF sensor model (including noise) faithfully represents the real VL53L5CX behavior.
    Experiments in Section IV use simulated point clouds and 'simulate the sensor noise'; no calibration or sim-to-real validation of the sensor model is reported.
  • domain assumption Obstacles generated tightly around the expert trajectory produce a distribution representative of real dense environments.
    This is the core benchmark premise in Section III.D; it makes the test distribution depend on the same pipeline that generated training data.
  • ad hoc to paper Emergency-stop data (goals inside obstacles, leading to collisions) is useful as expert demonstration data.
    Section III.D introduces a strategy where the goal lies inside an obstacle and the arm always collides; the paper does not explain how colliding trajectories are used as expert supervision.
invented entities (1)
  • ARMOR distributed ToF sensor constellation independent evidence
    purpose: Egocentric, occlusion-resistant depth perception for humanoid arms
    The hardware exists and was mounted on a real GR1 robot, but the simulation model used in the main experiments is not validated against real sensor data in this version.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ARMOR: Egocentric Perception for Humanoid Robot Collision Avoidance and Motion Planning." pith.science (2026). https://pith.science/paper/IZ4K4XJB

@misc{pith2026241200396,
  author       = {Pith},
  title        = {Pith review of: ARMOR: Egocentric Perception for Humanoid Robot Collision Avoidance and Motion Planning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IZ4K4XJB}},
  note         = {Machine review of arXiv:2412.00396}
}
read the original abstract

Humanoid robots have significant gaps in their sensing and perception, making it hard to perform motion planning in dense environments. To address this, we introduce ARMOR, a novel egocentric perception system that integrates both hardware and software, specifically incorporating wearable-like depth sensors for humanoid robots. Our distributed perception approach enhances the robot's spatial awareness, and facilitates more agile motion planning. We also train a transformer-based imitation learning (IL) policy in simulation to perform dynamic collision avoidance, by leveraging around 86 hours worth of human realistic motions from the AMASS dataset. We show that our ARMOR perception is superior against a setup with multiple dense head-mounted, and externally mounted depth cameras, with a 63.7% reduction in collisions, and 78.7% improvement on success rate. We also compare our IL policy against a sampling-based motion planning expert cuRobo, showing 31.6% less collisions, 16.9% higher success rate, and 26x reduction in computational latency. Lastly, we deploy our ARMOR perception on our real-world GR1 humanoid from Fourier Intelligence. We are going to update the link to the source code, HW description, and 3D CAD files in the arXiv version of this text.

Figures

Figures reproduced from arXiv: 2412.00396 by the authors.

Figure 1
Figure 1. ARMOR presents a novel egocentric wearable perception hardware and software system for humanoid robots (left). [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 3
Figure 3. ARMOR’s egocentric perception hardware in simu [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 5
Figure 5. Three data generation strategies. In collision-avoidance motion, a 1-second sequence of human motion in the AMASS [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗
Figures from the paper (2 more)
Figure 6
Figure 6. Figure 6: Experiment setup. The yellow geometries indi [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 7
Figure 7. Figure 7: An example of how our ARMOR perception resolves [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Humanoid Occupancy: Enabling A Generalized Multimodal Occupancy Perception System on Humanoid Robots

    cs.RO 2025-07 conditional novelty 5.0 of 10

    A humanoid-specific multimodal occupancy perception system with a new dataset, sensor layout, and a fusion network that claims state-of-the-art results on its own benchmark.

  2. Learning Fast, Tool aware Collision Avoidance for Collaborative Robots

    cs.RO 2025-08 conditional novelty 4.0 of 10

    A real-time, tool-aware collision avoidance system for cobots that blends a learned perception-safety critic with classical IK, achieving low collision rates in dynamic partially-observed environments.

Reference graph

Works this paper leans on

34 extracted references · 27 canonical work pages · cited by 2 Pith papers

  1. [1]

    A multi-modal sensor array for safe human- robot interaction and mapping

    Colette Abah, Andrew Orekhov, Garrison Johnston, Peng Yin, Howie Choset, and Nabil Simaan. A multi-modal sensor array for safe human- robot interaction and mapping. In 2019 International Conference on Robotics and Automation (ICRA) , pages 3768–3774, 05 2019

    work page 2019

  2. [2]

    Efficient bimanual manipulation using learned task schemas

    Rohan Chitnis, Shubham Tulsiani, Saurabh Gupta, and Abhinav Gupta. Efficient bimanual manipulation using learned task schemas. In 2020 IEEE International Conference on Robotics and Automation (ICRA) , pages 1149–1155, 2020

    work page 2020

  3. [3]

    Neural mp: A generalist neural motion planner

    Murtaza Dalal, Jiahui Yang, Russell Mendonca, Youssef Khaky, Rus- lan Salakhutdinov, and Deepak Pathak. Neural mp: A generalist neural motion planner. September 2024

    work page 2024

  4. [4]

    Bert: Pre-training of deep bidirectional transformers for language understanding, 2019

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre-training of deep bidirectional transformers for language understanding, 2019

    2019

  5. [5]

    Motion policy networks

    Adam Fishman, Adithyavairavan Murali, Clemens Eppner, Bryan Peele, Byron Boots, and Dieter Fox. Motion policy networks. In Karen Liu, Dana Kulic, and Jeff Ichnowski, editors, Proceedings of The 6th Conference on Robot Learning , volume 205 of Proceedings of Machine Learning Research , pages 967–977. PMLR, 14–18 Dec 2023

    work page 2023

  6. [6]

    A depth space approach to human-robot collision avoidance

    Fabrizio Flacco, Torsten Kr ¨oger, Alessandro De Luca, and Oussama Khatib. A depth space approach to human-robot collision avoidance. pages 338–345, Saint Paul, MN, USA, 2012. IEEE

    work page 2012

  7. [7]

    Humanplus: Humanoid shadowing and imitation from humans

    Zipeng Fu, Qingqing Zhao, Qi Wu, Gordon Wetzstein, and Chelsea Finn. Humanplus: Humanoid shadowing and imitation from humans. In Conference on Robot Learning (CoRL) , 2024

    work page 2024

  8. [8]

    From cyskin to proxyskin: Design, implementation and testing of a multi-modal robotic skin for human–robot interaction

    Francesco Giovinazzo, Francesco Grella, Marco Sartore, Manuela Adami, Riccardo Galletti, and Giorgio Cannata. From cyskin to proxyskin: Design, implementation and testing of a multi-modal robotic skin for human–robot interaction. Sensors, 24(4), 2024

    work page 2024

Show all 34 references
  1. [9]

    Stabilize to act: Learning to coordinate for bimanual manipulation

    Jennifer Grannen, Yilin Wu, Brandon Vu, and Dorsa Sadigh. Stabilize to act: Learning to coordinate for bimanual manipulation. In Jie Tan, Marc Toussaint, and Kourosh Darvish, editors, Proceedings of The 7th Conference on Robot Learning , volume 229 of Proceedings of Machine Le...

    2023

  2. [10]

    Baku: An efficient transformer for multi-task policy learning, 2024

    Siddhant Haldar, Zhuoran Peng, and Lerrel Pinto. Baku: An efficient transformer for multi-task policy learning, 2024

    2024

  3. [11]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016

    2016

  4. [12]

    Sparsh: Self-supervised touch representations for vision-based tactile sensing, 2024

    Carolina Higuera, Akash Sharma, Chaithanya Krishna Bodduluri, Taosha Fan, Patrick Lancaster, Mrinal Kalakrishnan, Michael Kaess, Byron Boots, Mike Lambeta, Tingfan Wu, and Mustafa Mukadam. Sparsh: Self-supervised touch representations for vision-based tactile sensing, 2024

    2024

  5. [13]

    Diffusion- seeder: Seeding motion optimization with diffusion for rapid motion planning

    Huang Huang, Balakumar Sundaralingam, Arsalan Mousavian, Adithyavairavan Murali, Ken Goldberg, and Dieter Fox. Diffusion- seeder: Seeding motion optimization with diffusion for rapid motion planning. October 2024

    2024

  6. [14]

    Digitizing touch with an artificial multimodal fingertip, 2024

    Mike Lambeta, Tingfan Wu, Ali Sengul, Victoria Rose Most, Nolan Black, Kevin Sawyer, Romeo Mercado, Haozhi Qi, Alexander Sohn, Byron Taylor, Norb Tydingco, Gregg Kammerer, Dave Stroud, Jake Khatha, Kurt Jenkins, Kyle Most, Neal Stein, Ricardo Chavira, Thomas Craven-Bartle, Eri...

    2024

  7. [15]

    Okami: Teaching humanoid robots manipu- lation skills through single video imitation

    Jinhan Li, Yifeng Zhu, Yuqi Xie, Zhenyu Jiang, Mingyo Seo, Georgios Pavlakos, and Yuke Zhu. Okami: Teaching humanoid robots manipu- lation skills through single video imitation. October 2024

    2024

  8. [16]

    Deltar: Depth estimation from a light-weight tof sensor and rgb image, 2022

    Yijin Li, Xinyang Liu, Wenqi Dong, Han Zhou, Hujun Bao, Guofeng Zhang, Yinda Zhang, and Zhaopeng Cui. Deltar: Depth estimation from a light-weight tof sensor and rgb image, 2022

    2022

  9. [17]

    Safe reinforcement learning on the constraint manifold: Theory and appli- cations

    Puze Liu, Haitham Bou-Ammar, Jan Peters, and Davide Tateo. Safe reinforcement learning on the constraint manifold: Theory and appli- cations. April 2024

    2024

  10. [18]

    Safe reinforcement learning of dynamic high-dimensional robotic tasks: Navigation, manipulation, interaction

    Puze Liu, Kuo Zhang, Davide Tateo, Snehal Jauhri, Zhiyuan Hu, Jan Peters, and Georgia Chalvatzaki. Safe reinforcement learning of dynamic high-dimensional robotic tasks: Navigation, manipulation, interaction. September 2022

    2022

  11. [19]

    Multi-Modal Neural Radiance Field for Monocular Dense SLAM with a Light-Weight ToF Sensor

    Xinyang Liu, Yijin Li, Yanbin Teng, Hujun Bao, Guofeng Zhang, Yinda Zhang, and Zhaopeng Cui. Multi-Modal Neural Radiance Field for Monocular Dense SLAM with a Light-Weight ToF Sensor. Technical report, August 2023. arXiv:2308.14383 [cs] type: article

    2023 arXiv

  12. [20]

    Troje, Gerard Pons- Moll, and Michael J

    Naureen Mahmood, Nima Ghorbani, Nikolaus F. Troje, Gerard Pons- Moll, and Michael J. Black. AMASS: Archive of motion capture as surface shapes. In International Conference on Computer Vision , pages 5442–5451, October 2019

    2019

  13. [21]

    Towards 3D Vision with Low-Cost Single-Photon Cameras

    Fangzhou Mu, Carter Sifferman, Sacha Jungerman, Yiquan Li, Mark Han, Michael Gleicher, Mohit Gupta, and Yin Li. Towards 3D Vision with Low-Cost Single-Photon Cameras. Technical report, March 2024. arXiv:2403.17801 [cs, eess] type: article

    2024 arXiv

  14. [22]

    Learning precise, contact-rich manipulation through uncalibrated tactile skins, 2024

    Venkatesh Pattabiraman, Yifeng Cao, Siddhant Haldar, Lerrel Pinto, and Raunaq Bhirangi. Learning precise, contact-rich manipulation through uncalibrated tactile skins, 2024

    2024

  15. [23]

    Improving language understanding by generative pre-training

    Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. Improving language understanding by generative pre-training. 2018

    2018

  16. [24]

    Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. Exploring the limits of transfer learning with a unified text-to-text transformer. CoRR, abs/1910.10683, 2019

    1910 arXiv

  17. [25]

    Molnar, Ryan Sinnet, and Aaron D

    Andrew Singletary, William Guffey, Tamas G. Molnar, Ryan Sinnet, and Aaron D. Ames. Safety-Critical Manipulation for Collision-Free Food Preparation. Technical report, May 2022. arXiv:2205.01026 [cs, eess] type: article

    2022 arXiv

  18. [26]

    Qwiic mini tof imager - vl53l5cx, 2024

    SparkFun. Qwiic mini tof imager - vl53l5cx, 2024. Accessed: 2024- 10-21

    2024

  19. [27]

    SAFER: Safe Collision Avoidance using Focused and Efficient Trajectory Search with Reinforcement Learning

    Mario Srouji, Hugues Thomas, Hubert Tsai, Ali Farhadi, and Jian Zhang. SAFER: Safe Collision Avoidance using Focused and Efficient Trajectory Search with Reinforcement Learning. Technical report, June 2023. arXiv:2209.11789 [cs] type: article

    2023 arXiv

  20. [28]

    Xiao esp32s3, 2024

    Seeed Studio. Xiao esp32s3, 2024. Accessed: 2024-10-21

    2024

  21. [29]

    Curobo: Parallelized collision-free robot motion generation

    Balakumar Sundaralingam, Siva Kumar Sastry Hari, Adam Fishman, Caelan Garrett, Karl Van Wyk, Valts Blukis, Alexander Millane, Helen Oleynikova, Ankur Handa, Fabio Ramos, Nathan Ratliff, and Dieter Fox. Curobo: Parallelized collision-free robot motion generation. In 2023 IEEE I...

    2023

  22. [30]

    Safety- Critical Control and Planning for Obstacle Avoidance between Poly- topes with Control Barrier Functions

    Akshay Thirugnanam, Jun Zeng, and Koushil Sreenath. Safety- Critical Control and Planning for Obstacle Avoidance between Poly- topes with Control Barrier Functions. Technical report, May 2022. arXiv:2109.12313 [cs, eess, math] type: article

    2022 arXiv

  23. [31]

    Efficient Motion Planning for Ma- nipulators with Control Barrier Function-Induced Neural Controller

    Mingxin Yu, Chenning Yu, M.-Mahdi Naddaf-Sh, Devesh Upadhyay, Sicun Gao, and Chuchu Fan. Efficient Motion Planning for Ma- nipulators with Control Barrier Function-Induced Neural Controller. Technical report, April 2024. arXiv:2404.01184 [cs] version: 1 type: article

    2024 arXiv

  24. [32]

    Generalizable humanoid manipulation with improved 3d diffusion policies

    Yanjie Ze, Zixuan Chen, Wenhao Wang, Tianyi Chen, Xialin He, Ying Yuan, Xue Bin Peng, and Jiajun Wu. Generalizable humanoid manipulation with improved 3d diffusion policies. arXiv preprint arXiv:2410.10803, 2024

    2024 arXiv

  25. [33]

    Generalizable humanoid manipulation with improved 3d diffusion policies

    Yanjie Ze, Zixuan Chen, Wenhao Wang, Tianyi Chen, Xialin He, Ying Yuan, Xue Bin Peng, and Jiajun Wu. Generalizable humanoid manipulation with improved 3d diffusion policies. October 2024

    2024

  26. [34]

    Zhao, Vikash Kumar, Sergey Levine, and Chelsea Finn

    Tony Z. Zhao, Vikash Kumar, Sergey Levine, and Chelsea Finn. Learning fine-grained bimanual manipulation with low-cost hardware, 2023

    2023

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.