Comparing Computing Platforms for Deep Learning on a Humanoid Robot

被引：7

作者：

Biddulph, Alexander ^{[1
]}

Houliston, Trent ^{[1
]}

Mendes, Alexandre ^{[1
]}

Chalup, Stephan K. ^{[1
]}

机构：

[1] Univ Newcastle, Sch Elect Engn & Comp, Callaghan, NSW 2308, Australia

来源：

NEURAL INFORMATION PROCESSING (ICONIP 2018), PT VII | 2018年 / 11307卷

关键词：

Deep learning; Robot vision; GPU computing; Low powered devices;

D O I：

10.1007/978-3-030-04239-4_11

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The goal of this study is to test two different computing platforms with respect to their suitability for running deep networks as part of a humanoid robot software system. One of the platforms is the CPU-centered Intel (R) NUC7i7BNH and the other is a NVIDIA (R) Jetson TX2 system that puts more emphasis on GPU processing. The experiments addressed a number of benchmarking tasks including pedestrian detection using deep neural networks. Some of the results were unexpected but demonstrate that platforms exhibit both advantages and disadvantages when taking computational performance and electrical power requirements of such a system into account.

引用

页码：120 / 131

页数：12

共 23 条

[1] Improving Performance of Dense Linear Algebra with Multi-core Architecture
Abouelfarag, Ahmed A.
Nouh, Nada Magdy
ElShenawy, Marwa
[J]. 2017 INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING & SIMULATION (HPCS), 2017, : 870 - 874
[2] Allgeuer P, 2015, IEEE-RAS INT C HUMAN, P33, DOI 10.1109/HUMANOIDS.2015.7363519
[3] [Anonymous], 2011, IMPROVING SPEED NEUR
[4] [Anonymous], 2017, LNCS LNAI, DOI [DOI 10.1007/978-3, 10.1007/978-3-319-68792-6_2]
[5] [Anonymous], 2017, Nvidia Jetson TX2 Delivers Twice the Intelligence to the Edge
[6] [Anonymous], 2017, ABS170404861 CORR
[7] Bainville E, 2009, GPU BENCHMARKS
[8] Machine learning with AIBO robots in the four-legged league of RoboCup
Chalup, Stephan K.
Murch, Craig L.
Quinlan, Michael J.
[J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART C-APPLICATIONS AND REVIEWS, 2007, 37 (03): : 297 - 310
[9] Ghidary S.S., 2017, 21 ANN ROBOCUP INT S
[10] Goodfellow I, 2016, ADAPT COMPUT MACH LE, P1

← 1 2 3 →