OSCR

Artificial plateau neurons with in-situ spike-malleability for rhythmic quadrupedal locomotion.

Code ↔ Paper

2 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 2 matches
  1. [1] § Methods › Quadruped robot actuations ↔ simulation/go2_config.py, lines 3–55 · score 0.57 · hip joints, calf joints, Go2, angle, thigh, legs
  2. [2] § Methods › Quadruped robot actuations ↔ simulation_revision/go2_config.py, lines 3–55 · score 0.57 · hip joints, calf joints, Go2, angle, thigh, legs

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 64 lines · 2.3 KB · no license · 1 match

  1. from legged_gym.envs.base.legged_robot_config import LeggedRobotCfg, LeggedRobotCfgPPO
  3. class GO2RoughCfg( LeggedRobotCfg ):
  4. class env( LeggedRobotCfg.env):
  5. num_observations = 36
  6. num_actions = 8
  8. class init_state( LeggedRobotCfg.init_state ):
  9. pos = [0.0, 0.0, 0.42] # x,y,z [m]
  10. default_joint_angles = { # = target angles [rad] when action = 0.0
  11. 'FL_hip_joint': 0.1, # [rad]
  12. 'RL_hip_joint': 0.1, # [rad]
  13. 'FR_hip_joint': -0.1 , # [rad]
  14. 'RR_hip_joint': -0.1, # [rad]
  16. 'FL_thigh_joint': 0.8, # [rad]
  17. 'RL_thigh_joint': 1., # [rad]
  18. 'FR_thigh_joint': 0.8, # [rad]
  19. 'RR_thigh_joint': 1., # [rad]
  21. 'FL_calf_joint': -1.5, # [rad]
  22. 'RL_calf_joint': -1.5, # [rad]
  23. 'FR_calf_joint': -1.5, # [rad]
  24. 'RR_calf_joint': -1.5, # [rad]
  25. }
  27. class control( LeggedRobotCfg.control ):
  28. # PD Drive parameters:
  29. control_type = 'P'
  30. stiffness = {'joint': 20.} # [N*m/rad]
  31. damping = {'joint': 0.5} # [N*m*s/rad]
  32. # action scale: target angle = actionScale * action + defaultAngle
  33. action_scale = 0.25
  34. # decimation: Number of control action updates @ sim DT per policy DT
  35. decimation = 4
  37. class asset( LeggedRobotCfg.asset ):
  38. file = '{LEGGED_GYM_ROOT_DIR}/resources/robots/go2/urdf/go2.urdf'
  39. name = "go2"
  40. foot_name = "foot"
  41. penalize_contacts_on = ["thigh", "calf"]
  42. terminate_after_contacts_on = ["base"]
  43. self_collisions = 1 # 1 to disable, 0 to enable...bitwise filter
  44. fix_base_link = True
  46. class rewards( LeggedRobotCfg.rewards ):
  47. soft_dof_pos_limit = 0.9
  48. base_height_target = 0.25
  49. class scales( LeggedRobotCfg.rewards.scales ):
  50. torques = -0.0002
  51. dof_pos_limits = -10.0
  52. class viewer (LeggedRobotCfg.viewer):
  53. ref_env = 0
  54. pos = [0, -1, 0.42] # [m]
  55. lookat = [0, 0, 0.42] # [m]
  57. class GO2RoughCfgPPO( LeggedRobotCfgPPO ):
  58. class algorithm( LeggedRobotCfgPPO.algorithm ):
  59. entropy_coef = 0.01
  60. class runner( LeggedRobotCfgPPO.runner ):
  61. run_name = ''
  62. experiment_name = 'rough_go2'

go2_config.py at commit 30856cb, no license · at the source

Overview

Authors: Hailiang Wang1, Yishu Zhang1,2, Qingao Chai3,4, Qian He1, Jiayang Hu1, Yongqing Bai1, Guanyu Liu1,2, Zongwen Li1, Jian Chai1, Xin He2, Mengze Zhao5, Guodong Xue5, Kaihui Liu5, Yu Fu6, Huajin Tang3,4, Yang Xu1, Bin Yu1
ORCID iDs: Yishu Zhang, Qingao Chai, Qian He, Xin He, Kaihui Liu, Yang Xu, Bin Yu
  1. College of Integrated Circuits, Zhejiang University, Hangzhou, Zhejiang China
  2. ZJU-Hangzhou Global Scientific and Technological Innovation Center, Hangzhou, Zhejiang China
  3. College of Computer Science and Technology, Zhejiang University, Hangzhou, China
  4. State Key Lab of Brain-Machine Intelligence, Zhejiang University, Hangzhou, China
  5. State Key Laboratory for Mesoscopic Physics, Frontiers Science Center for Nano-Optoelectronics, School of Physics, Peking University, Beijing, China
  6. Key Laboratory of Quantum State Construction and Manipulation, Department of Physics, Renmin University of China, Beijing, China
Institutions: Zhejiang University (China); Peking University (China); Renmin University of China (China)
Journal: Nature communications, volume 17, issue 1, article 5801
Dates: received 20 August 2025; accepted 30 March 2026; published online 28 April 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1038/s41467-026-72428-2 · PMID 42045188 · PMCID PMC13332192 · OpenAlex W7156183590
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: systems (subfield)
Keywords: Electronic devices, Motor control
MeSH: Locomotion*, Neurons*, Action Potentials, Animals, Gait, Intelligent Systems, Models, Neurological, Periodicity, Robotics (* major topic)
Topic: Advanced Memory and Neural Computing (Electrical and Electronic Engineering, Engineering), according to OpenAlex
Funding: National Natural Science Foundation of China (DT23F0401); Natural Science Foundation of Zhejiang Province (LDT23F04011F04)
Citations: not cited yet (Europe PMC); 51 references in the paper

Abstract

Whole-body intelligent locomotion systems face persistent challenges of redundant actuation and poor energy efficiency, limiting real-world deployment. Bio-inspired central pattern generators offer a promising framework for rhythmic control, yet hardware implementations struggle to match the efficiency and adaptability of biological systems. Here, we introduce an in-situ spike-malleable artificial plateau neuron integrating a bistable plateau gate with a transient threshold-switch. The neuron generates amplitude-programmable rhythmic spike bursts, achieving energy-efficient, antagonistic activation of extensors and flexors via a scalable circuit comprising two paired units (plateau gate and threshold-switch). The design leverages distributed encoding for coordinated muscle control, operating at ultra-low energy dissipation (141.37 pJ/spike). An expanded four-unit circuit enhances dynamic spike malleability, enabling parallel processing for multi-joint coordination. On a quadruped robot (Unitree Go2), these distributed circuits directly drive joint-level proportional derivative controllers using the Gaussian-filtered rhythmic spikes, enabling energy-efficient trotting without centralized computation. Critically, the system achieves stable on-ground locomotion and demonstrates adaptive gait transitions in real-world environments. Our approach merges ultra-compact hardware with bio-inspired architecture, advancing neuromorphic systems for energy-efficient autonomous robotics.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 2 matches between paragraphs and lines of code.

chaiqingao/go2_control

License: none: the authors keep all their rights
State: the link answers, verified on 30 September 2026
Evidence: files inventoried
Commit: 30856cbcb0a53493dc15cecef5cf6eccb43069b1, 2 February 2026
Languages: Python (10)
Size: 13 files, 10 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: NumPy (8 files), PyTorch (3 files), pandas (2 files)
Availability: 1 check, the latest on 30 September 2026: the link answers
  • 30 September 2026: the link answers
11 files
At the source: github.com/chaiqingao/go2_control

Code availability

The code supporting the configuration of the robot in this study is available at https://github.com/chaiqingao/go2_control with unrestricted access.

Reproduced under the paper's license (CC BY), from the paper cited above.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 10 scripts, each with its path and the digest of its content;
  • 2 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

Datasets cited

Data availability

The data supporting the quadruped robot application findings in this study are openly available in the Figshare database at 10.6084/m9.figshare.31746325. Detailed research data supporting the plots of this study are available from the corresponding authors upon request.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 30 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 17 authors, 2 keywords, 9 MeSH terms, 2 funders, 39 references.

Cite

This paper

Wang, H., Zhang, Y., Chai, Q., He, Q., Hu, J., Bai, Y., Liu, G., Li, Z., Chai, J., He, X., Zhao, M., Xue, G., Liu, K., Fu, Y., Tang, H., Xu, Y., & Yu, B. (2026). Artificial plateau neurons with in-situ spike-malleability for rhythmic quadrupedal locomotion. Nature communications, 17(1), 5801. https://doi.org/10.1038/s41467-026-72428-2

BibTeX

@article{wang2026artificial,
author = {Wang, Hailiang and Zhang, Yishu and Chai, Qingao and He, Qian and Hu, Jiayang and Bai, Yongqing and Liu, Guanyu and Li, Zongwen and Chai, Jian and He, Xin and Zhao, Mengze and Xue, Guodong and Liu, Kaihui and Fu, Yu and Tang, Huajin and Xu, Yang and Yu, Bin},
title = {{Artificial plateau neurons with in-situ spike-malleability for rhythmic quadrupedal locomotion}},
journal = {Nature communications},
year = {2026},
month = apr,
volume = {17},
number = {1},
pages = {5801},
publisher = {Nature Publishing Group},
issn = {2041-1723},
doi = {10.1038/s41467-026-72428-2},
url = {https://doi.org/10.1038/s41467-026-72428-2},
pmid = {42045188},
pmcid = {PMC13332192}
}

RIS

TY - JOUR
AU - Wang, Hailiang
AU - Zhang, Yishu
AU - Chai, Qingao
AU - He, Qian
AU - Hu, Jiayang
AU - Bai, Yongqing
AU - Liu, Guanyu
AU - Li, Zongwen
AU - Chai, Jian
AU - He, Xin
AU - Zhao, Mengze
AU - Xue, Guodong
AU - Liu, Kaihui
AU - Fu, Yu
AU - Tang, Huajin
AU - Xu, Yang
AU - Yu, Bin
TI - Artificial plateau neurons with in-situ spike-malleability for rhythmic quadrupedal locomotion
T2 - Nature communications
J2 - Nat Commun
PY - 2026
DA - 2026/04/28
VL - 17
IS - 1
SP - 5801
SN - 2041-1723
PB - Nature Publishing Group
DO - 10.1038/s41467-026-72428-2
UR - https://doi.org/10.1038/s41467-026-72428-2
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41467-026-72428-2",
"type": "article-journal",
"title": "Artificial plateau neurons with in-situ spike-malleability for rhythmic quadrupedal locomotion",
"container-title": "Nature communications",
"author": [
{
"family": "Wang",
"given": "Hailiang"
},
{
"family": "Zhang",
"given": "Yishu"
},
{
"family": "Chai",
"given": "Qingao"
},
{
"family": "He",
"given": "Qian"
},
{
"family": "Hu",
"given": "Jiayang"
},
{
"family": "Bai",
"given": "Yongqing"
},
{
"family": "Liu",
"given": "Guanyu"
},
{
"family": "Li",
"given": "Zongwen"
},
{
"family": "Chai",
"given": "Jian"
},
{
"family": "He",
"given": "Xin"
},
{
"family": "Zhao",
"given": "Mengze"
},
{
"family": "Xue",
"given": "Guodong"
},
{
"family": "Liu",
"given": "Kaihui"
},
{
"family": "Fu",
"given": "Yu"
},
{
"family": "Tang",
"given": "Huajin"
},
{
"family": "Xu",
"given": "Yang"
},
{
"family": "Yu",
"given": "Bin"
}
],
"container-title-short": "Nat Commun",
"volume": "17",
"issue": "1",
"page": "5801",
"DOI": "10.1038/s41467-026-72428-2",
"PMID": "42045188",
"PMCID": "PMC13332192",
"ISSN": "2041-1723",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41467-026-72428-2",
"language": "en",
"issued": {
"date-parts": [
[
2026,
4,
28
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1038/s41467-026-73825-3 [code]
Volatile self-selective memristive neuron for millisecond-latency neuromorphic object detection at the edge.
Journal: Nature communications
In common: 2 references, author Yishu Zhang
[2] doi:10.1038/s41467-026-76185-0 [code]
Spinal-inspired artificial tactile interneuron with high-order burst spiking for intelligent edge interfaces.
Journal: Nature communications
In common: PyTorch, pandas, NumPy, 3 references
[3] doi:10.1038/s41467-026-74228-0 [code]
Spatial and network principles behind neural generation of locomotion.
Journal: Nature communications
In common: NumPy, 4 references
[4] doi:10.1038/s41467-026-72900-z [code]
Resistive memory-based neural differential equation solver for score-based diffusion model.
Journal: Nature communications
In common: PyTorch, pandas, NumPy, 2 references
[5] doi:10.1038/s41467-026-73669-x [code]
Homeostatic dendritic neuron based on co-integrated volatile and non-volatile memristors for neuromorphic processing.
Journal: Nature communications
In common: PyTorch, NumPy, 2 references
[6] doi:10.1523/jneurosci.1506-25.2026 [code]
Controlling Spatio-Temporal Sequences of Neural Activity by Local Synaptic Changes.
Journal: The Journal of neuroscience : the official journal of the Society for Neuroscience
In common: pandas, NumPy, 2 references
[7] doi:10.1038/s41467-026-75979-6 [code]
Artificial neural manifolds.
Journal: Nature communications
In common: NumPy, systems, 2 references
[8] doi:10.1038/s41467-026-75347-4 [code]
Sleep reveals dynamics integrating and segregating movement and stimulus representations in V1.
Journal: Nature communications
In common: PyTorch, pandas, NumPy, systems, 1 reference
[9] doi:10.7554/elife.111876 [code]
Distinct sensorimotor encoding in tuft dendrites and somata associated with action, correction, and learning.
Journal: eLife
In common: PyTorch, pandas, NumPy, 1 reference
[10] doi:10.1016/j.isci.2026.117010 [code]
Deep learning-assisted mapping of dendritic spines using sequential 2D two-photon calcium imaging.
Journal: iScience
In common: PyTorch, pandas, NumPy, 1 reference

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.