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

Mattamala, Matías; Olave, Gonzalo; González, Clayder; Hasbún, Nicolás; Ruiz‐del‐Solar, Javier

doi:10.1007/978-3-030-00308-1_25

Cited by 6 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other authors [32] have proposed the implementation of a backpack for NAO that contains an ODROID XU4 running object classification algorithm. Nevertheless, when it ran ORB-SLAM2, it achieved 12 fps, which is below the benchmark for real-time applications.…”

Section: Nao Backpack Designmentioning

confidence: 99%

“…The battery sizing was based on the power required by the Ultra96 FPGA, the power consumed when executing a CNN and the autonomy in relation to the time the humanoid robot will be active. Regarding the last factor, the reference point in this study is [32], which reports that, despite the 60 minutes offered by the battery included in NAO, in the Robo Cup Soccer competition, the autonomy of the robot is only 30 minutes. Therefore, the operating time of the Ultra96 FPGA was estimated at 50 minutes of activity.…”

Section: Sizing the Battery Powering The Embedded Systemmentioning

confidence: 99%

See 1 more Smart Citation

Performance Evaluation of Convolutional Networks on Heterogeneous Architectures for Applications in Autonomous Robotics

Guajo

Alzate-Anzola

Castano-Londono

et al. 2022

TecnoL.

View full text Add to dashboard Cite

Humanoid robots find application in human-robot interaction tasks. However, despite their capabilities, their sequential computing system limits the execution of computationally expensive algorithms such as convolutional neural networks, which have demonstrated good performance in recognition tasks. As an alternative to sequential computing units, Field-Programmable Gate Arrays and Graphics Processing Units have a high degree of parallelism and low power consumption. This study aims to improve the visual perception of a humanoid robot called NAO using these embedded systems running a convolutional neural network. The methodology adopted here is based on image acquisition and transmission using simulation software: Webots and Choreographe. In each embedded system, an object recognition stage is performed using commercial convolutional neural network acceleration frameworks. Xilinx® Ultra96™, Intel® Cyclone® V-SoC and NVIDIA® Jetson™ TX2 cards were used, and Tinier-YOLO, AlexNet, Inception-V1 and Inception V3 transfer-learning networks were executed. Real-time metrics were obtained when Inception V1, Inception V3 transfer-learning and AlexNet were run on the Ultra96 and Jetson TX2 cards, with frame rates between 28 and 30 frames per second. The results demonstrated that the use of these embedded systems and convolutional neural networks can provide humanoid robots such as NAO with greater visual recognition in tasks that require high accuracy and autonomy.

show abstract

Section: Nao Backpack Designmentioning

confidence: 99%

Section: Sizing the Battery Powering The Embedded Systemmentioning

confidence: 99%

Performance Evaluation of Convolutional Networks on Heterogeneous Architectures for Applications in Autonomous Robotics

Guajo

Alzate-Anzola

Castano-Londono

et al. 2022

TecnoL.

View full text Add to dashboard Cite

show abstract

“…Humanoids are well adapted for human-designed mobile manipulation activities such as walking, reaching various types of terrain, moving in complex environments such as environments with stairs and/or narrow passages, navigating in cluttered environments without colliding with any barriers, etc. Such capabilities would make humanoid robots perfect assistants to humans,such as disaster management or housekeeping [2]- [5].…”

Section: Introductionmentioning

confidence: 99%

NAO Humanoid Robot Obstacle Avoidance Using Monocular Camera

Gouda¹,

Chikha²

2020

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

In this paper, we present an experimental approach that allows a humanoid robot to effectively plan and execute whole body motions, like climbing obstacles and straight stairs up or down, besides jumping over obstacles using only on-board sensing. Reliable and accurate motion sequence for humanoid employed in complex indoor environments is a necessity for high-level robotics tasks. Using the robot's own kinematics will construct complex dynamic motions. A series of actions to prevent the object from being performed on the basis of the identified object from the database of the robot, obtained using the robot's own monocular camera. As shown in real world experiments beside simulation using NAO H25 humanoid, the robot can effectively perform whole body movements in cluttered, multilevel environments containing items of various shapes and dimensions.

show abstract

“…Environments where service and social robots operate/live are highly dynamic, in part because of the people living in those environments constantly interact with each other and carry out daily life activities. Therefore, service and social robots require perception systems that are highly robust, but at the same time are able to operate in real-time 3 . Object detection and recognition are some of the vision tasks that require at least near real-time operation.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, the main goal of this paper is to propose a solution to this problem that considers two components: First, the selection of YOLO (You only look Once) [2], a DNN with real-time detection and recognition of objects and persons, as the most suitable DNN to this task; and second, the development and implementation of an add-on for Pepper, a backpack that permit the attachment of a single board computer onto Pepper, particularly a Nvidia Jetson TK1, which can run YOLO at about 5 FPS when processing images of 320x320 pixels. The use of an external enhancing device, attached to the robot without modifying its structure, comes as an inspiration from similar projects developed for the NAO robot [3], where a fully replicable backpack was built.…”

Section: Introductionmentioning

confidence: 99%

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

Reyes

Gómez

Norambuena

et al. 2019

RoboCup 2018: Robot World Cup XXII

Self Cite

View full text Add to dashboard Cite

The main goal of the paper is to provide Pepper with a near real-time object recognition system based on deep neural networks. The proposed system is based on YOLO (You Only Look Once), a deep neural network that is able to detect and recognize objects robustly and at a high speed. In addition, considering that YOLO cannot be run in the Pepper's internal computer in near real-time, we propose to use a Backpack for Pepper, which holds a Jetson TK1 card and a battery. By using this card, Pepper is able to robustly detect and recognize objects in images of 320x320 pixels at about 5 frames per second.

show abstract

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

Cited by 6 publications

References 10 publications

Performance Evaluation of Convolutional Networks on Heterogeneous Architectures for Applications in Autonomous Robotics

Performance Evaluation of Convolutional Networks on Heterogeneous Architectures for Applications in Autonomous Robotics

NAO Humanoid Robot Obstacle Avoidance Using Monocular Camera

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

Contact Info

Product

Resources

About