SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2008.03596 · 2020

TriFinger: An Open-Source Robot for Learning Dexterity

arXiv · PDF · Open in the Atlas

Code that ran

9 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 1 failure of our instrument; 1 no recorded run), measured 2026-11-21; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2008.03596.json · badge: /api/ran/2008.03596.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
open-dynamic-robot-initiative/robot_fingers canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
open-dynamic-robot-initiative/trifinger_simulation community (classifier label)
classifier basis not recorded on the link
8 ran · 1 no recorded run
FunctionStatusWhere it lives
calib_data_to_matrix Ran
ran on a synthesized input
open-dynamic-robot-initiative/trifinger_simulation/trifinger_simulation/camera.py
code served (permissive licence) · get_code("642ef04048b73593")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (3, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Construct a valid camera calibration data dictionary
# The function expects a dict with 'data', 'rows', and 'cols' keys
# It will flatten 'data' and reshape it to (rows, cols)

data_array = np.array([
    1.0, 2.0, 3.0, 4.0,
    5.0, 6.0, 7.0, 8.0,
    9.0, 10.0, 11.0, 12.0
])

ARGS = {
    'data': {
        'data': data_array,
        'rows': 3,
        'cols': 4
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (3, 4) — shape (3, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
check_finger_type Ran
ran on a synthesized input
open-dynamic-robot-initiative/trifinger_simulation/trifinger_simulation/finger_types_data.py
code served (permissive licence) · get_code("0a2b36887be2bc93")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import typing

class FingerTypesDataFormat(typing.NamedTuple):
    """Describes the format for the finger type data,
    comprising of the corresponding urdf, and the
    number of fingers.
    """
    urdf_file: str
    number_of_fingers: int
    initial_joint_positions: typing.Sequence[float]

finger_types_data = {'fingerone': FingerTypesDataFormat(urdf_file='finger.urdf', number_of_fingers=1, initial_joint_positions=[0, 0, 0]), 'trifingerone': FingerTypesDataFormat(urdf_file='trifinger.urdf', number_of_fingers=3, initial_joint_positions=[0, -0.9, -1.7] * 3), 'fingeredu': FingerTypesDataFormat(urdf_file='edu/fingeredu.urdf', number_of_fingers=1, initial_joint_positions=[0, 0.9, -1.7]), 'trifingeredu': FingerTypesDataFormat(urdf_file='edu/trifingeredu.urdf', number_of_fingers=3, initial_joint_positions=[0, 0.9, -1.7] * 3), 'fingerpro': FingerTypesDataFormat(urdf_file='pro/fingerpro.urdf', number_of_fingers=1, initial_joint_positions=[0, 0.9, -1.7]), 'trifingerpro': FingerTypesDataFormat(urdf_file='pro/trifingerpro.urdf', number_of_fingers=3, initial_joint_positions=[0, 0.9, -1.7] * 3)}

def check_finger_type(name: str) -> str:
    """
    Check if *name* is a valid finger type.

    Args:
        name: Name of the finger type.

    Returns:
        The name if it is valid.

    Raises:
        ValueError: If *name* is not a valid finger type.
    """
    if name not in finger_types_data.keys():
        raise ValueError("Invalid finger type '%s'.  Valid types are %s" % (name, finger_types_data.keys()))
    else:
        return name

ARGS = {
    "name": "fingerone"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
find_robot_config_file Ran
ran; our contract draft was wrong, not the code
open-dynamic-robot-initiative/robot_fingers/scripts/trifinger_backend.py
code served (permissive licence) · get_code("918348f04f63c3af")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned PosixPath. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import pathlib
import tempfile
import os

# Create a temporary directory structure with a config file
temp_dir = tempfile.mkdtemp()
config_dir_path = pathlib.Path(temp_dir)

# Create one of the expected config files so the function can find it
config_file = config_dir_path / "trifinger.yml"
config_file.touch()

ARGS = {
    "config_dir": config_dir_path,
    "filenames": ('trifinger.yml', 'trifingerpro.yml')
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PosixPath — a PosixPath output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PosixPath output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_finger_urdf Ran
ran on a synthesized input
open-dynamic-robot-initiative/trifinger_simulation/trifinger_simulation/finger_types_data.py
code served (permissive licence) · get_code("9a7fcdb941137f0d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import typing

class FingerTypesDataFormat(typing.NamedTuple):
    """Describes the format for the finger type data,
    comprising of the corresponding urdf, and the
    number of fingers.
    """
    urdf_file: str
    number_of_fingers: int
    initial_joint_positions: typing.Sequence[float]

finger_types_data = {'fingerone': FingerTypesDataFormat(urdf_file='finger.urdf', number_of_fingers=1, initial_joint_positions=[0, 0, 0]), 'trifingerone': FingerTypesDataFormat(urdf_file='trifinger.urdf', number_of_fingers=3, initial_joint_positions=[0, -0.9, -1.7] * 3), 'fingeredu': FingerTypesDataFormat(urdf_file='edu/fingeredu.urdf', number_of_fingers=1, initial_joint_positions=[0, 0.9, -1.7]), 'trifingeredu': FingerTypesDataFormat(urdf_file='edu/trifingeredu.urdf', number_of_fingers=3, initial_joint_positions=[0, 0.9, -1.7] * 3), 'fingerpro': FingerTypesDataFormat(urdf_file='pro/fingerpro.urdf', number_of_fingers=1, initial_joint_positions=[0, 0.9, -1.7]), 'trifingerpro': FingerTypesDataFormat(urdf_file='pro/trifingerpro.urdf', number_of_fingers=3, initial_joint_positions=[0, 0.9, -1.7] * 3)}

def check_finger_type(name: str) -> str:
    """
    Check if *name* is a valid finger type.

    Args:
        name: Name of the finger type.

    Returns:
        The name if it is valid.

    Raises:
        ValueError: If *name* is not a valid finger type.
    """
    if name not in finger_types_data.keys():
        raise ValueError("Invalid finger type '%s'.  Valid types are %s" % (name, finger_types_data.keys()))
    else:
        return name

def get_finger_urdf(name: str) -> str:
    """
    Get the name of the URDF-file with the model of the specified finger type.

    Args:
        name: Name of the finger type.

    Returns:
        The name of the URDF file in the ``robot_properties_fingers`` package.

    Raises:
        ValueError: If *name* is not a valid finger type.
    """
    check_finger_type(name)
    return finger_types_data[name].urdf_file

ARGS = {
    "name": "fingerone"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_number_of_fingers Ran
ran on a synthesized input
open-dynamic-robot-initiative/trifinger_simulation/trifinger_simulation/finger_types_data.py
code served (permissive licence) · get_code("00f1719c13bc8c50")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned int, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import typing

class FingerTypesDataFormat(typing.NamedTuple):
    """Describes the format for the finger type data,
    comprising of the corresponding urdf, and the
    number of fingers.
    """
    urdf_file: str
    number_of_fingers: int
    initial_joint_positions: typing.Sequence[float]

finger_types_data = {
    'fingerone': FingerTypesDataFormat(
        urdf_file='finger.urdf',
        number_of_fingers=1,
        initial_joint_positions=[0, 0, 0]
    ),
    'trifingerone': FingerTypesDataFormat(
        urdf_file='trifinger.urdf',
        number_of_fingers=3,
        initial_joint_positions=[0, -0.9, -1.7] * 3
    ),
    'fingeredu': FingerTypesDataFormat(
        urdf_file='edu/fingeredu.urdf',
        number_of_fingers=1,
        initial_joint_positions=[0, 0.9, -1.7]
    ),
    'trifingeredu': FingerTypesDataFormat(
        urdf_file='edu/trifingeredu.urdf',
        number_of_fingers=3,
        initial_joint_positions=[0, 0.9, -1.7] * 3
    ),
    'fingerpro': FingerTypesDataFormat(
        urdf_file='pro/fingerpro.urdf',
        number_of_fingers=1,
        initial_joint_positions=[0, 0.9, -1.7]
    ),
    'trifingerpro': FingerTypesDataFormat(
        urdf_file='pro/trifingerpro.urdf',
        number_of_fingers=3,
        initial_joint_positions=[0, 0.9, -1.7] * 3
    )
}

def check_finger_type(name: str) -> str:
    """
    Check if *name* is a valid finger type.

    Args:
        name: Name of the finger type.

    Returns:
        The name if it is valid.

    Raises:
        ValueError: If *name* is not a valid finger type.
    """
    if name not in finger_types_data.keys():
        raise ValueError("Invalid finger type '%s'.  Valid types are %s" % (name, finger_types_data.keys()))
    else:
        return name

def get_number_of_fingers(name: str) -> int:
    """
    Get the number of fingers of the specified finger type

    Args:
        name: Name of the finger type.

    Returns:
        Number of fingers.

    Raises:
        ValueError: If *name* is not a valid finger type.
    """
    check_finger_type(name)
    return finger_types_data[name].number_of_fingers

ARGS = {
    'name': 'fingerone'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
int_to_rgba Ran
ran on a synthesized input
open-dynamic-robot-initiative/trifinger_simulation/trifinger_simulation/sim_finger.py
code served (permissive licence) · get_code("93edcb9355562a7b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import typing

def int_to_rgba(color: int, alpha: int=255) -> typing.Tuple[float, float, float, float]:
    """Convert an 24-bit integer to an rgba tuple.

    Converts color given as a single 24-bit integer (e.g. a hex value 0xFF0011)
    to an rgba tuple where each value is in the range [0, 1].

    Args:
        color: An RGB colour given as single number (e.g. 0xFF0011).
        alpha: Optional alpha value in the interval [0, 0xFF].

    Returns:
        tuple: The colour as a tuple (r, g, b, a) where each element is in
        [0.0, 1.0].
    """

    def trim_and_convert(x):
        return (x & 255) / 255
    return (trim_and_convert(color >> 16), trim_and_convert(color >> 8), trim_and_convert(color), trim_and_convert(alpha))


ARGS = {
    "color": 0xFF0011,
    "alpha": 255
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_camera_parameters Ran
ran on a synthesized input
open-dynamic-robot-initiative/trifinger_simulation/trifinger_simulation/camera.py
code served (permissive licence) · get_code("2be4c7e29899e0c7")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import pathlib
import tempfile
import numpy as np
import yaml

# Create temporary directory for camera config files
tmpdir = tempfile.TemporaryDirectory()
config_dir = pathlib.Path(tmpdir.name)

# Create sample camera YAML files for camera IDs 60, 180, 300
camera_configs = {
    60: {
        'camera_name': 'camera_60',
        'image_width': 640,
        'image_height': 480,
        'camera_matrix': {
            'rows': 3,
            'cols': 3,
            'data': [520.0, 0.0, 320.0, 0.0, 520.0, 240.0, 0.0, 0.0, 1.0]
        },
        'distortion_coefficients': {
            'rows': 1,
            'cols': 5,
            'data': [0.1, -0.2, 0.0, 0.0, 0.05]
        },
        'tf_world_to_camera': {
            'rows': 4,
            'cols': 4,
            'data': [1.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 1.0]
        }
    },
    180: {
        'camera_name': 'camera_180',
        'image_width': 800,
        'image_height': 600,
        'camera_matrix': {
            'rows': 3,
            'cols': 3,
            'data': [550.0, 0.0, 400.0, 0.0, 550.0, 300.0, 0.0, 0.0, 1.0]
        },
        'distortion_coefficients': {
            'rows': 1,
            'cols': 5,
            'data': [0.15, -0.25, 0.01, 0.02, 0.03]
        },
        'tf_world_to_camera': {
            'rows': 4,
            'cols': 4,
            'data': [0.9, 0.1, 0.0, 1.0, -0.1, 0.9, 0.0, 2.0, 0.0, 0.0, 1.0, 3.0, 0.0, 0.0, 0.0, 1.0]
        }
    },
    300: {
        'camera_name': 'camera_300',
        'image_width': 1024,
        'image_height': 768,
        'camera_matrix': {
            'rows': 3,
            'cols': 3,
            'data': [600.0, 0.0, 512.0, 0.0, 600.0, 384.0, 0.0, 0.0, 1.0]
        },
        'distortion_coefficients': {
            'rows': 1,
            'cols': 5,
            'data': [0.2, -0.3, 0.005, 0.01, 0.02]
        },
        'tf_world_to_camera': {
            'rows': 4,
            'cols': 4,
            'data': [0.8, 0.2, 0.0, 2.0, -0.2, 0.8, 0.0, 3.0, 0.0, 0.0, 1.0, 5.0, 0.0, 0.0, 0.0, 1.0]
        }
    }
}

# Write YAML files to temporary directory
for camera_id, config in camera_configs.items():
    filename = config_dir / f'camera{camera_id}.yml'
    with open(filename, 'w') as f:
        yaml.dump(config, f, default_flow_style=False, sort_keys=False)

ARGS = {
    'config_dir': config_dir,
    'filename_pattern': 'camera{id}.yml'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
random_joint_positions Ran
ran on a synthesized input
open-dynamic-robot-initiative/trifinger_simulation/trifinger_simulation/sample.py
code served (permissive licence) · get_code("5b0703ca7f5ab015")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import random
import math

# Set seed for reproducibility
random.seed(42)

# Define ARGS dictionary with all parameters
ARGS = {
    "number_of_fingers": 3,
    "lower_bounds": [-math.radians(30), -math.radians(60), -math.radians(100)],
    "upper_bounds": [math.radians(30), math.radians(60), math.radians(2)]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
random_position_in_arena Ran
ran on a synthesized input
open-dynamic-robot-initiative/trifinger_simulation/trifinger_simulation/sample.py
code served (permissive licence) · get_code("162398f4bb6b9e79")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import random
import math

# Set seeds for reproducibility
random.seed(42)
np.random.seed(42)

# Define ARGS dict with all parameters for random_position_in_arena
ARGS = {
    "height_limits": (0.05, 0.15),
    "angle_limits": (-2 * math.pi, 2 * math.pi),
    "radius_limits": (0.0, 0.15)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
feasible_random_joint_positions_for_reaching Not yet run
unverified (harvested, no recorded run)
open-dynamic-robot-initiative/trifinger_simulation/trifinger_simulation/sample.py
code served (permissive licence) · get_code("90e3c1bcfa611626")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 10 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Dexterous object manipulation remains an open problem in robotics, despite the rapid progress in machine learning during the past decade. We argue that a hindrance is the high cost of experimentation on real systems, in terms of both time and money. We address this problem by proposing an open-source robotic platform which can safely operate without human supervision. The hardware is inexpensive (about \SI{5000}[\$]{}) yet highly dynamic, robust, and capable of complex interaction with external objects. The software operates at 1-kilohertz and performs safety checks to prevent the hardware from breaking. The easy-to-use front-end (in C++ and Python) is suitable for real-time control as well as deep reinforcement learning. In addition, the software framework is largely robot-agnostic and can hence be used independently of the hardware proposed herein. Finally, we illustrate the potential of the proposed platform through a number of experiments, including real-time optimal control, deep reinforcement learning from scratch, throwing, and writing.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2008.03596")
get_code_for_paper("2008.03596")
have("2008.03596")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2008.03596.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2008.03596.svg)](https://syntology.ai/paper/2008.03596)

Connect an agent — have() is free.