pith. sign in

arxiv: 2606.12112 · v1 · pith:ZEWJFC4Knew · submitted 2026-06-10 · 💻 cs.RO

PEBRE: An Open-Hardware Compute and Perception Add-On for the Pepper Robot

Malte Kuhlmann , Ignacio Bugueno-Cordova , Emil Alms , Javier Ruiz-del-Solar , Nicol\'as Navarro-Guerrero This is my paper

Pith reviewed 2026-06-27 09:20 UTC · model grok-4.3

classification 💻 cs.RO
keywords PEBREPepper robotopen hardwarerobot add-onperception enhancementcomputational upgradehumanoid robotmodular design
0
0 comments X

The pith

The PEBRE open-hardware add-on considerably improves the Pepper robot's perception abilities and computational power.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces PEBRE, a modular open-hardware attachment designed to extend the Pepper robot by adding a Jetson Orin Nano processor along with upgraded cameras and microphones. It demonstrates through experiments that these components deliver measurable gains in how quickly the robot can process data and how accurately it can perceive its surroundings. The design is released as open hardware and open source to let other researchers attach modern parts without replacing the entire robot. This keeps an established research platform usable for new software experiments even after its original manufacturer support ends.

Core claim

PEBRE is an open-hardware and open-source modular add-on that integrates a Jetson Orin Nano, Logitech BRIO, Intel RealSense D435i, Samson UB1, and RØDE VideoMicro II with the Pepper robot; experiments confirm that the added hardware considerably improves the robot's perception abilities and computational power while enabling faster software development and easier integration of external components.

What carries the argument

The PEBRE add-on module, which physically and electrically attaches external compute, vision, and audio components to the Pepper robot to bypass its original hardware limits.

If this is right

  • The Pepper robot remains usable for research beyond its original expected lifespan.
  • Researchers can integrate newer sensors and processors more quickly without custom redesigns for each project.
  • The open design lets the community share and improve attachments for perception tasks.
  • Software experiments on Pepper can run at higher frame rates and with richer sensor data than before.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar modular add-ons could be built for other discontinued robot platforms to reuse existing research code.
  • The approach separates the base robot body from its sensing and thinking hardware, which may simplify testing of new algorithms.
  • Open release of the mechanical and electrical interface lowers the barrier for groups that lack access to high-end robot hardware.

Load-bearing premise

The measured gains in perception and speed come from the added hardware rather than unrelated changes in software or test setup.

What would settle it

A side-by-side test on identical Pepper units and tasks showing no difference in perception accuracy or processing speed when the PEBRE module is attached versus when it is removed.

read the original abstract

This paper presents the design, development, and experimental verification of PEBRE, an open-hardware add-on for fast software development on the Pepper Robot. Our project enhances Pepper's computational and perception capabilities by integrating external components such as a Jetson Orin Nano, Logitech BRIO, Intel RealSense D435i, Samson UB1, and R{\O}DE VideoMicro II. Our results show that the new hardware considerably improved Pepper's perception abilities and computational power. This development contributes to the community by implementing an open hardware and open-source modular add-on to the Pepper robot and keeping this relevant research platform functional beyond its expected lifespan. With PEBRE, we aim to facilitate faster software development and more efficient integration of external components, ultimately enhancing the capabilities of the Pepper robot.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The manuscript describes the design, development, and experimental verification of PEBRE, an open-hardware add-on for the Pepper robot. It integrates a Jetson Orin Nano for computation along with Logitech BRIO, Intel RealSense D435i, Samson UB1, and RØDE VideoMicro II hardware for enhanced perception. The authors claim that this module considerably improves Pepper's perception abilities and computational power while contributing open hardware and open-source resources to extend the platform's relevance.

Significance. If substantiated, the work would provide a practical, modular open-hardware solution for modernizing the Pepper robot, supporting continued research in human-robot interaction and education. The open-hardware and open-source emphasis is a clear strength that enables reproducibility and community extension.

major comments (1)
  1. [Abstract] Abstract: the claim that 'the new hardware considerably improved Pepper's perception abilities and computational power' and that 'experimental verification' occurred is presented without any quantitative metrics (FPS, latency, accuracy, error bars), baselines, or controlled test protocols. This is load-bearing for the central contribution and prevents evaluation of whether observed changes are attributable to the added hardware.
minor comments (1)
  1. The microphone model is written as 'R{\O}DE VideoMicro II'; verify spelling and provide a reference or datasheet link if the component is central to the perception claims.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their detailed review and constructive feedback on our manuscript. We address the single major comment below and agree that strengthening the abstract will improve clarity and allow readers to better evaluate the contribution.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that 'the new hardware considerably improved Pepper's perception abilities and computational power' and that 'experimental verification' occurred is presented without any quantitative metrics (FPS, latency, accuracy, error bars), baselines, or controlled test protocols. This is load-bearing for the central contribution and prevents evaluation of whether observed changes are attributable to the added hardware.

    Authors: We agree that the abstract, as currently written, does not include the quantitative metrics, baselines, or protocol details needed to substantiate the central claims. While the body of the manuscript (evaluation section) reports these results with controlled comparisons to the stock Pepper platform, the abstract should be self-contained. We will revise the abstract to include key metrics (e.g., perception FPS, end-to-end latency, accuracy on benchmark tasks) and a brief reference to the test protocols and baselines used. revision: yes

Circularity Check

0 steps flagged

No circularity: hardware design report with no derivations or fitted predictions

full rationale

The manuscript is a hardware integration report describing component selection, assembly, and experimental verification of an add-on module. No equations, parameter fits, predictions, or first-principles derivations appear anywhere in the text. The central claim of improved perception and compute is presented as an empirical outcome of the added hardware rather than a quantity derived from or defined in terms of itself. No self-citations function as load-bearing uniqueness theorems, and no ansatzes or renamings of known results are invoked. The derivation chain is therefore self-contained and contains no steps that reduce to their own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is an engineering hardware-integration paper with no mathematical model, free parameters, axioms, or postulated entities.

pith-pipeline@v0.9.1-grok · 5684 in / 1084 out tokens · 31036 ms · 2026-06-27T09:20:48.167095+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

26 extracted references · 6 canonical work pages

  1. [1]

    10 Years of Human-NAO Interaction Research: A Scoping Review

    Amirova A, Rakhymbayeva N, Yadollahi E, Sandygulova A, Johal W. 10 Years of Human-NAO Interaction Research: A Scoping Review. Frontiers in Robotics and AI. 2021 Nov;8(744526):358. https://doi.org/10.3389/ frobt.2021.744526

    arXiv 2021

  2. [2]

    iCub: The Not-yet-Finished Story of Building a Robot Child

    Natale L, Bartolozzi C, Pucci D, Wykowska A, Metta G. iCub: The Not-yet-Finished Story of Building a Robot Child. Science Robotics. 2017 Dec;2(13):eaaq1026. https://doi.org/10. 1126/scirobotics.aaq1026

    2017

  3. [3]

    Quori: A Community-Informed Design of a Socially Interactive Humanoid Robot

    Specian A, Mead R, Kim S, Matarić M, Yim M. Quori: A Community-Informed Design of a Socially Interactive Humanoid Robot. IEEE Transactions on Robotics. 2022 Jun;38(3):1755–1772. https://doi.org/ 10.1109/TRO.2021.3111718

    work page doi:10.1109/tro.2021.3111718 2022

  4. [4]

    Robot Making Human-Like Actions for Social Interaction: Going Deeper into ROS2 Multiprocessing and Kinemat- ics

    Sharma V, Mupenda DN, Thorvik L, Mishra D. Robot Making Human-Like Actions for Social Interaction: Going Deeper into ROS2 Multiprocessing and Kinemat- ics. In: Human-Computer Interaction. Cham: 18 Springer Nature Switzerland; 2025. p. 346– 359

    2025

  5. [5]

    A Mass-Produced Sociable Humanoid Robot: Pepper: The First Machine of Its Kind

    Pandey AK, Gelin R. A Mass-Produced Sociable Humanoid Robot: Pepper: The First Machine of Its Kind. IEEE Robotics Automa- tion Magazine. 2018 Sep;25(3):40–48. https: //doi.org/10.1109/MRA.2018.2833157

    work page doi:10.1109/mra.2018.2833157 2018

  6. [6]

    EXCLUSIVE SoftBank Shrinks Robotics Business, Stops Pepper Production- Sources

    Nussey S. EXCLUSIVE SoftBank Shrinks Robotics Business, Stops Pepper Production- Sources. Reuters. 2021 Jun

    2021

  7. [7]

    The Pepper Humanoid Robot in Front Desk Application

    Gardecki A, Podpora M, Beniak R, Klin B. The Pepper Humanoid Robot in Front Desk Application. In: Progress in Applied Electri- cal Engineering (PAEE). Koscielisko, Poland

  8. [8]

    ToBI - Team of Bielefeld Enhancing the Robot Capabilities of the Social Standard Platform Pepper

    Lier F, Kummert J, Renner P, Wachsmuth S. ToBI - Team of Bielefeld Enhancing the Robot Capabilities of the Social Standard Platform Pepper. In: RoboCup 2018: Robot World Cup XXII. vol. 11374 of LNCS. Springer International Publishing; 2019. p. 524–535

    2018

  9. [9]

    Analyzing Emotions and Engagement During Cognitive Stimulation Group Training with the Pepper Robot

    De Carolis B, Macchiarulo N, Palestra G, Pino O. Analyzing Emotions and Engagement During Cognitive Stimulation Group Training with the Pepper Robot. IEEE Transactions on Affective Computing. 2025;p. 1–14. https: //doi.org/10.1109/TAFFC.2025.3609855

    work page doi:10.1109/taffc.2025.3609855 2025

  10. [10]

    Improving Autonomy and Natural Interaction of Pep- per Robot via Large Language Models

    Rojas L, Romero JA, Manrique R. Improving Autonomy and Natural Interaction of Pep- per Robot via Large Language Models. SN Computer Science. 2026 Mar;7(4):328. https: //doi.org/10.1007/s42979-026-04892-y

    work page doi:10.1007/s42979-026-04892-y 2026

  11. [11]

    A Framework for Low-Latency, LLM-Driven Multimodal Interaction on the Pepper Robot

    Studerus E, Zhong VJ, Vonschallen S. A Framework for Low-Latency, LLM-Driven Multimodal Interaction on the Pepper Robot. In: ACM/IEEE International Conference on Human-Robot Interaction (HRI). Edinburgh, Scotland, UK; 2026. p. 1298–1302

    2026

  12. [12]

    Near Real-Time Object Recognition for Pepper Based on Deep Neural Networks Running on a Backpack

    Reyes E, Gómez C, Norambuena E, Ruiz-del- Solar J. Near Real-Time Object Recognition for Pepper Based on Deep Neural Networks Running on a Backpack. In: RoboCup 2018: Robot World Cup XXII. vol. 11374 of LNCS. Montreal,QC,Canada:SpringerInternational Publishing; 2019. p. 287–298

    2018

  13. [13]

    Adapted Pepper

    Caniot M, Bonnet V, Busy M, Labaye T, Besombes M, Courtois S, et al. Adapted Pepper. arXiv; 2020. arXiv:2009.03648

    arXiv 2020

  14. [14]

    Dimensions - A Collaborative Approach to Enhancing Research Discovery

    Adams J, Draux H, Jones P, Osipov I, Porter S, Szomszor M. Dimensions - A Collaborative Approach to Enhancing Research Discovery. Digital Science; 2018

    2018

  15. [15]

    Improved Navigation for Social Robots Through OS and Sensor Augmenta- tion [MSc]

    Bleth A. Improved Navigation for Social Robots Through OS and Sensor Augmenta- tion [MSc]. University of Minnesota; 2023

    2023

  16. [16]

    Upgrad- ing Pepper Robot’s Social Interaction with Advanced Hardware and Perception Enhancements

    Magri P, Amirian J, Chetouani M. Upgrad- ing Pepper Robot’s Social Interaction with Advanced Hardware and Perception Enhancements. In:SocialRobotics.vol.15561– 15563 of LNAI. Odense, Denmark: Springer Nature; 2025. p. 344–350

    2025

  17. [17]

    The NAO Backpack: An Open-Hardware Add-on for Fast Soft- ware Development with the NAO Robot

    Mattamala M, Olave G, González C, Has- bún N, Ruiz-del-Solar J. The NAO Backpack: An Open-Hardware Add-on for Fast Soft- ware Development with the NAO Robot. In: RoboCup 2017: Robot World Cup XXI. vol. 11175 of LNCS. Nagoya, Japan: Springer International Publishing; 2018. p. 302–311

    2017

  18. [18]

    Open Robot Hardware: Progress, Benefits, Chal- lenges, and Best Practices

    Patel VV, Liarokapis MV, Dollar AM. Open Robot Hardware: Progress, Benefits, Chal- lenges, and Best Practices. IEEE Robotics & Automation Magazine. 2023 Sep;30(3):123–

    2023

  19. [19]

    https://doi.org/10.1109/MRA.2022. 3225725

    work page doi:10.1109/mra.2022 2022

  20. [20]

    Version 4.0.5, Accessed: 2025-03-27

    Aldebaran.: Libqi. Version 4.0.5, Accessed: 2025-03-27. https://github.com/aldebaran/ libqi

    2025

  21. [21]

    Version 2.55.1 Accessed: 2025-03-27

    IntelRealSense.: Librealsense. Version 2.55.1 Accessed: 2025-03-27. https://github.com/ IntelRealSense/librealsense

    2025

  22. [22]

    Dlib-Ml: A Machine Learn- ing Toolkit

    King DE. Dlib-Ml: A Machine Learn- ing Toolkit. Journal of Machine Learning Research. 2009;10(60):1755–1758. 19

    2009

  23. [23]

    Orb-slam3: An accurate open-source library for visual, visual–inertial, and multimap slam.IEEE Transactions on Robotics2021; 37(6): 1874–1890

    Campos C, Elvira R, Rodríguez JJG, M Mon- tiel JM, D Tardós J. ORB-SLAM3: An Accu- rate Open-Source Library for Visual, Visual– Inertial, and Multimap SLAM. IEEE Transac- tions on Robotics. 2021 Dec;37(6):1874–1890. https://doi.org/10.1109/TRO.2021.3075644

    work page doi:10.1109/tro.2021.3075644 2021

  24. [24]

    https://mm.digikey.com/ Volume0/opasdata/d220001/medias/ docus/5380/Jetson_Orin_Nano_Series_ DS-11105-001_v1.1.pdf

    NVIDIA.: Jetson Orin Nano Series Mod- ule Datasheet. https://mm.digikey.com/ Volume0/opasdata/d220001/medias/ docus/5380/Jetson_Orin_Nano_Series_ DS-11105-001_v1.1.pdf

  25. [25]

    https://www.aldebaran.com/en/ support/pepper/technical-support

    Robotics S.: Pepper Technical Specifi- cations. https://www.aldebaran.com/en/ support/pepper/technical-support

  26. [26]

    https: //cdrdv2-public.intel.com/329645/ atom-e3800-m-d-i-soc-thermal-design-guide

    Intel.: Intel Atom Processor E3800 Series Thermal Design Guide. https: //cdrdv2-public.intel.com/329645/ atom-e3800-m-d-i-soc-thermal-design-guide. pdf. 20