SlideShare ist ein Scribd-Unternehmen logo
1 von 24
Real-Time Cloud Robotics
in Practical Smart City Applications
Nazli Khan Beigi, Bahar Partov, Soodeh Farokhi
C2RO: Collaborative Cloud Robotics
Montreal, Canada
Paper Contribution
• Cloud robotics as a means of
resource optimization in smart city applications.
• C2RO platform contributing in
• Parallel computing,
• Cloud and edge computing,
• Machine learning,
• Computer vision.
What is Cloud Robotics?
• The major trend in today’s robotics since its
emergence in 2010.
• Network-connected robots offloading
• Intensive and complex computation tasks
• Data sharing available in a centralized location.
• Advancements in the cloud computing and big data:
• To push the barriers in artificial intelligence applications.
C2RO Platform
• C2RO cloud robotics platform uses real-time stream processing technology
• Virtually connects, stores, and processes
• The energy-efficient and low-cost mobile robotic devices or sensors.
Cloud Robotics Challenge: Latency
End-to-end latency = processing time + network response time
Robotics applications classifications:
• Hard-real time applications, e.g., collision avoidance
• Soft real-time applications, e.g., object recognition
• Delay tolerant applications, e.g., mapping
Latency, Processing Time
In a collision avoidance application, for instance, the PT includes:
• Video capturing
• Video compression/ decompression
• Object recognition
• Algorithm Implementation
Latency, Network Response Time
• Latency in lower layers of communication, i.e., from devices to the edge of the network
• Latency in higher levels of the communication network, i.e., from edge to the cloud:
• Professional data stream network (DSN)
• To reduce the routing latency
• To provide redundant paths
• A partnership with PubNub, globally scaled DSN
• With 14 data centers and 99.999% service level agreements (SLAs).
Ultimate Solution to Latency:
Hybrid Cloud Robotics
• The real-time processing is dynamically
distributed among resources
• on-board,
• on the edge,
• on the cloud.
• Edge computing, i.e., fog, to fill the latency
and network incompetency gap in real-time
applications.
C2RO Hybrid Cloud Robotics
Computation Model Examples
• LIDAR sensor | car processing unit | cloud;
• Vacuum cleaner or toy | personal computer or home network router | cloud;
• Surveillance camera | significant robot | cloud.
Experimental Setup:
• Real time Object Recognition
• C-SLAM: Localization and Autonomous Mapping
Application of C2RO Platform in Smart Cities
• Smart cities have scattered massive number of
sensors
• Complicated infrastructure to connect and process
• Cloud robotics: an efficient computing means in
intensive data processing applications.
• A high-profile collaborative urban “People
Counting” project done in collaboration with
Philips and MIT.
• Street lights are transformed the into
multifunctional digital urban platforms.
• Cloud connected smart thermal sensors are added
People Counting System Configuration
• Raspberry Pi 3 SBC,
• Radiometric-capable LWIR camera, i.e. FLIR Lepton.
• Four VMs
• The platform and dashboard at AmazonEC2.
VMs in the Cloud
CPU Model Intel Xeon E5
CPU Frequency/ Mhz 2400
Cores 4/8
Thread per core 1
Memory/MB 8192
OS Ubuntu 16.04 (Linux 4.4)
People Counting Snapshot
People Counting Video
Conclusion
• Cloud robotics has been significantly pushed the barriers of robotics.
• It is not still a proper model for real-time, latency sensitive smart city
applications.
• We propose our hybrid cloud robotics model
• We present a scalable platform, named C2RO platform, for parallel
computing
• The application of this platform for a high-profile urban project was reported
References
1. B. Tang et al., “Incorporating Intelligence in Fog Computing for Big Data Analysis in Smart Cities,” in IEEE Trans. on Industrial Informatics, March 2017
2. B. Kehoe, et al., “A Survey of Research on Cloud Robotics and Automation,” in IEEE Trans. On Automation & Eng., VOL. 12, NO. 2, April 2015
3. J. Wan, et al., “Cloud Robotics: Current Status and Open Issues,” in Special Section in IEEE Access: The Plethora of Research in Internet of Things (IoT),
2016
4. N. Gangid, B. Sharma, “Cloud Computing and Robotics for Disaster Management,” in IEEE Inter. Conference on Intelligent Systems, Modelling and
Simulation, 2016
5. J. Salmeron-Garcia, et al., “A Tradeoff Analysis of a Cloud-Based Robot Navigation Assistant Using Stereo Image Processing”, in IEEE Trans. On Automation
Science and Engineering, Vol.12, No.2, April 2015
6. L. Riazuelo et al., “RoboEarth Semantic Mapping: A Cloud Enabled Knowledge-Based Approach,” in IEEE Trans. On Automation Science and Eng., VOL. 12,
NO. 2, April 2015
7. K. Sugiura, K. Zettsu, “Rospeex: A Cloud Robotics Platform for Human-Robot Spoken Dialogues,” in IROS, Sept. 2015, pp. 6155-6060.
8. J. Mahler et al., “Dex-Net 1.0: A Cloud-Based Network of 3D Objects for Robust Grasp Planning Using a Multi-Armed Bandit Model with Correlated
Rewards,” in ICRA, May 2016
9. S. Dey, A. Mukherjee, “Robotic SLAM - a Review from Fog Computing and Mobile Edge Computing Perspective,” in IEEE International Conference on
Mobile and Ubiquitous Systems: Computing Networking and Services, 2016
10. M. Satyanarayanan, “Edge Computing for Situational Awareness,” in IEEE Local and Metropolitan Area Networks (LANMAN), 2017
11. K. Bilal, A. Erbad, “Edge Computing for Interactive Media and Video Streaming,” in IEEE International Conference on Fog and Mobile Edge Computing
(FMEC), 2017.
12. S. Kamburugamuve, L. Christiansen, G. Fox, “A framework for real-time processing of sensor data in the cloud”, in Journal of Sensors, 2015
13. H. He, et al., “Cloud based Real-time Multi-Robot Collision Avoidance for Swarm Robotics”, in International Journal of Grid and Distributed Computing, Vol.
9, No.6, 2016, pp. 339- 358
14. S. Farokhi et al., “Performance Boost with Hybrid Cloud Robotics,” in International Conference on Intelligent Robots and Systems, Sept. 2017
15. A. Anjomshoaa, et al., “Quantifying the Anthropogenic Heat in Urban Areas Using Thermal Images,” in CSCI, 2016
Contact Us
• Soodeh Farokhi, Founder and CTO, C2RO Robotics, soodeh.farokhi@c2ro.com
• Nazli Khan Beigi, Communication Specialist, C2RO Robotics, nazli.khanbeigi@c2ro.com
• www.c2ro.com
Additional Slides
Use cases of C2RO Hybrid Cloud Robotics
Use cases of C2RO Hybrid Cloud Robotics
Object Recognition (OR):
• The robotic platform is equipped with a single-board computer (SBC), i.e.
Raspberry Pi 3, and an RGBD sensor.
• In case that OR is performed only on the SBC, the performance of the
process is ~0.07fps, and the user sees new images with detected objects
every ~13 seconds.
• While using our proposed model, the entire process improves to ~50fps,
which translates into a real-time visualization of the detection process.
Use cases of C2RO Hybrid Cloud Robotics
Simultaneous Localization and Mapping (SLAM)
• The robotic platform is equipped with a single-board computer (SBC), i.e.
Raspberry Pi 3, and an RGBD sensor.
• If the map is built locally on the SBC, the SLAM algorithm update rate is
~5Hz;
• While by using our hybrid cloud robotics model, the map update rate was
enhanced to 30Hz.
Additional Slides
Parallel Cloud Computation
Cloud Computation Topologies
1. Carries out all computation in one processor. Though this reduces the
transmission of multi-robot state information between processors, but due
to huge computations, the processor would be very busy and this leads
into slow computation.
2. Assigns each computational component to one processor. This would
require transfer of robots’ state information between processors, in
addition to an imbalance of processing between delay-sensitive and non-
delay-sensitive components. So, the overall delay of this topology was
rather high too.
3. Parallel processing topology that unlike other parallel algorithms, would
focus on entity or agent level parallelization and would study mainly
computation resource scaling based on the computation load.
Parallel Cloud Computation
The topology is to prevent the delays caused by non-delay sensitive functions
by combining the processes of the similar essence in one processor.
Also, although each processor could run multiple instances in parallel, we
made sure that the proper data types are sent to the processor instance that
catches the right robot information.
Hence, by proposing an innovative load balancing technique, we provide an
efficient, low latency parallel computation algorithm
Parallel Cloud Computation
The same algorithm is implemented on SBC, a desktop computer server, a desktop
computer GPU, and C2RO cloud, including 4, 4, 4, and 20 processing cores, respectively.
Hence, by proposing an innovative load balancing technique, we provide an efficient,
low latency parallel computation algorithm.

Weitere ähnliche Inhalte

Was ist angesagt?

Microactuations thermal force
Microactuations thermal forceMicroactuations thermal force
Microactuations thermal force
Haider Alkaisy
 
Spintronics
SpintronicsSpintronics
Spintronics
Mayank Vora
 
Principles of Multiscale Modelling of Materials
Principles of Multiscale Modelling of Materials  Principles of Multiscale Modelling of Materials
Principles of Multiscale Modelling of Materials
Altair
 

Was ist angesagt? (20)

Spintronics Introduction (BASIC)
Spintronics Introduction (BASIC)Spintronics Introduction (BASIC)
Spintronics Introduction (BASIC)
 
Microactuations thermal force
Microactuations thermal forceMicroactuations thermal force
Microactuations thermal force
 
Spintronics : New Era in Nano Tecnology technology
Spintronics : New Era in Nano Tecnology technologySpintronics : New Era in Nano Tecnology technology
Spintronics : New Era in Nano Tecnology technology
 
Quantum Sensor .pptx.pdf
Quantum Sensor .pptx.pdfQuantum Sensor .pptx.pdf
Quantum Sensor .pptx.pdf
 
Spintronics
SpintronicsSpintronics
Spintronics
 
Spintronics report
Spintronics reportSpintronics report
Spintronics report
 
Photoluminescence
PhotoluminescencePhotoluminescence
Photoluminescence
 
Spintronics ppt
Spintronics pptSpintronics ppt
Spintronics ppt
 
spintronics ppt
spintronics pptspintronics ppt
spintronics ppt
 
Cloud robotics
Cloud roboticsCloud robotics
Cloud robotics
 
Giant magnetoresistance and their applications
Giant magnetoresistance and their applicationsGiant magnetoresistance and their applications
Giant magnetoresistance and their applications
 
Spintronics ( IEEE presentation )
Spintronics ( IEEE presentation )Spintronics ( IEEE presentation )
Spintronics ( IEEE presentation )
 
Smart Dust - A Little Wonder
Smart Dust - A Little WonderSmart Dust - A Little Wonder
Smart Dust - A Little Wonder
 
quantum computing and Quantum Communications
quantum computing and Quantum Communicationsquantum computing and Quantum Communications
quantum computing and Quantum Communications
 
Spintronics
Spintronics Spintronics
Spintronics
 
Thermal sensor and its application
Thermal sensor and its applicationThermal sensor and its application
Thermal sensor and its application
 
Sensor technology
Sensor technologySensor technology
Sensor technology
 
Piezoelectric Materials and Applications
Piezoelectric Materials and ApplicationsPiezoelectric Materials and Applications
Piezoelectric Materials and Applications
 
Principles of Multiscale Modelling of Materials
Principles of Multiscale Modelling of Materials  Principles of Multiscale Modelling of Materials
Principles of Multiscale Modelling of Materials
 
Squid2
Squid2Squid2
Squid2
 

Ähnlich wie Real-Time Cloud Robotics in Practical Smart City Applications

A survey of fog computing concepts applications and issues
A survey of fog computing concepts  applications and issuesA survey of fog computing concepts  applications and issues
A survey of fog computing concepts applications and issues
Rezgar Mohammad
 
slam_research_paper
slam_research_paperslam_research_paper
slam_research_paper
Vinit Payal
 
Zpryme Report on Cloud and SAS Solutions
Zpryme Report on Cloud and SAS SolutionsZpryme Report on Cloud and SAS Solutions
Zpryme Report on Cloud and SAS Solutions
Paula Smith
 
Cloud Robotics.pptx
Cloud Robotics.pptxCloud Robotics.pptx
Cloud Robotics.pptx
ShadekulIslamShovo
 

Ähnlich wie Real-Time Cloud Robotics in Practical Smart City Applications (20)

Cloud Robotics: It’s time to offload their brain on Cloud, for better Robotic...
Cloud Robotics: It’s time to offload their brain on Cloud, for better Robotic...Cloud Robotics: It’s time to offload their brain on Cloud, for better Robotic...
Cloud Robotics: It’s time to offload their brain on Cloud, for better Robotic...
 
IoT A Fog-Cloud Computing Model
IoT A Fog-Cloud Computing ModelIoT A Fog-Cloud Computing Model
IoT A Fog-Cloud Computing Model
 
Seminar report
Seminar reportSeminar report
Seminar report
 
13 9246 it implementation of cloud connected (edit ari)
13 9246 it implementation of cloud connected (edit ari)13 9246 it implementation of cloud connected (edit ari)
13 9246 it implementation of cloud connected (edit ari)
 
云计算及其应用
云计算及其应用云计算及其应用
云计算及其应用
 
Edge computing and its role in architecting IoT
Edge computing and its role in architecting IoTEdge computing and its role in architecting IoT
Edge computing and its role in architecting IoT
 
A survey of fog computing concepts applications and issues
A survey of fog computing concepts  applications and issuesA survey of fog computing concepts  applications and issues
A survey of fog computing concepts applications and issues
 
Ambient Intelligence perspective from IoT insight
Ambient Intelligence perspective from IoT insightAmbient Intelligence perspective from IoT insight
Ambient Intelligence perspective from IoT insight
 
A Review- Fog Computing and Its Role in the Internet of Things
A Review- Fog Computing and Its Role in the Internet of ThingsA Review- Fog Computing and Its Role in the Internet of Things
A Review- Fog Computing and Its Role in the Internet of Things
 
slam_research_paper
slam_research_paperslam_research_paper
slam_research_paper
 
Infrastructure of services for a smart city
Infrastructure of services for a smart cityInfrastructure of services for a smart city
Infrastructure of services for a smart city
 
Fog computing technology
Fog computing technologyFog computing technology
Fog computing technology
 
Fog compute good
Fog compute goodFog compute good
Fog compute good
 
Mobile Fog: A Programming Model for Large–Scale Applications on the Internet ...
Mobile Fog: A Programming Model for Large–Scale Applications on the Internet ...Mobile Fog: A Programming Model for Large–Scale Applications on the Internet ...
Mobile Fog: A Programming Model for Large–Scale Applications on the Internet ...
 
Zpryme Report on Cloud and SAS Solutions
Zpryme Report on Cloud and SAS SolutionsZpryme Report on Cloud and SAS Solutions
Zpryme Report on Cloud and SAS Solutions
 
Evaluating Cloud & Fog Computing based on Shifting & Scheduling Algorithms, L...
Evaluating Cloud & Fog Computing based on Shifting & Scheduling Algorithms, L...Evaluating Cloud & Fog Computing based on Shifting & Scheduling Algorithms, L...
Evaluating Cloud & Fog Computing based on Shifting & Scheduling Algorithms, L...
 
Cloud Robotics.pptx
Cloud Robotics.pptxCloud Robotics.pptx
Cloud Robotics.pptx
 
IoT.pdf
IoT.pdfIoT.pdf
IoT.pdf
 
internet architecture.pdf
internet architecture.pdfinternet architecture.pdf
internet architecture.pdf
 
A Comparison of Cloud Execution Mechanisms Fog, Edge, and Clone Cloud Computing
A Comparison of Cloud Execution Mechanisms Fog, Edge, and Clone Cloud Computing A Comparison of Cloud Execution Mechanisms Fog, Edge, and Clone Cloud Computing
A Comparison of Cloud Execution Mechanisms Fog, Edge, and Clone Cloud Computing
 

KĂźrzlich hochgeladen

+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
?#DUbAI#??##{{(☎️+971_581248768%)**%*]'#abortion pills for sale in dubai@
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Victor Rentea
 

KĂźrzlich hochgeladen (20)

Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
 
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusExploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with Milvus
 
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptx
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
 
Vector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptxVector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptx
 
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
 
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
 
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)
 

Real-Time Cloud Robotics in Practical Smart City Applications

  • 1. Real-Time Cloud Robotics in Practical Smart City Applications Nazli Khan Beigi, Bahar Partov, Soodeh Farokhi C2RO: Collaborative Cloud Robotics Montreal, Canada
  • 2. Paper Contribution • Cloud robotics as a means of resource optimization in smart city applications. • C2RO platform contributing in • Parallel computing, • Cloud and edge computing, • Machine learning, • Computer vision.
  • 3. What is Cloud Robotics? • The major trend in today’s robotics since its emergence in 2010. • Network-connected robots offloading • Intensive and complex computation tasks • Data sharing available in a centralized location. • Advancements in the cloud computing and big data: • To push the barriers in artificial intelligence applications.
  • 4. C2RO Platform • C2RO cloud robotics platform uses real-time stream processing technology • Virtually connects, stores, and processes • The energy-efficient and low-cost mobile robotic devices or sensors.
  • 5. Cloud Robotics Challenge: Latency End-to-end latency = processing time + network response time Robotics applications classifications: • Hard-real time applications, e.g., collision avoidance • Soft real-time applications, e.g., object recognition • Delay tolerant applications, e.g., mapping
  • 6. Latency, Processing Time In a collision avoidance application, for instance, the PT includes: • Video capturing • Video compression/ decompression • Object recognition • Algorithm Implementation
  • 7. Latency, Network Response Time • Latency in lower layers of communication, i.e., from devices to the edge of the network • Latency in higher levels of the communication network, i.e., from edge to the cloud: • Professional data stream network (DSN) • To reduce the routing latency • To provide redundant paths • A partnership with PubNub, globally scaled DSN • With 14 data centers and 99.999% service level agreements (SLAs).
  • 8. Ultimate Solution to Latency: Hybrid Cloud Robotics • The real-time processing is dynamically distributed among resources • on-board, • on the edge, • on the cloud. • Edge computing, i.e., fog, to fill the latency and network incompetency gap in real-time applications.
  • 9. C2RO Hybrid Cloud Robotics Computation Model Examples • LIDAR sensor | car processing unit | cloud; • Vacuum cleaner or toy | personal computer or home network router | cloud; • Surveillance camera | significant robot | cloud.
  • 10. Experimental Setup: • Real time Object Recognition • C-SLAM: Localization and Autonomous Mapping
  • 11. Application of C2RO Platform in Smart Cities • Smart cities have scattered massive number of sensors • Complicated infrastructure to connect and process • Cloud robotics: an efficient computing means in intensive data processing applications. • A high-profile collaborative urban “People Counting” project done in collaboration with Philips and MIT. • Street lights are transformed the into multifunctional digital urban platforms. • Cloud connected smart thermal sensors are added
  • 12. People Counting System Configuration • Raspberry Pi 3 SBC, • Radiometric-capable LWIR camera, i.e. FLIR Lepton. • Four VMs • The platform and dashboard at AmazonEC2. VMs in the Cloud CPU Model Intel Xeon E5 CPU Frequency/ Mhz 2400 Cores 4/8 Thread per core 1 Memory/MB 8192 OS Ubuntu 16.04 (Linux 4.4)
  • 15. Conclusion • Cloud robotics has been significantly pushed the barriers of robotics. • It is not still a proper model for real-time, latency sensitive smart city applications. • We propose our hybrid cloud robotics model • We present a scalable platform, named C2RO platform, for parallel computing • The application of this platform for a high-profile urban project was reported
  • 16. References 1. B. Tang et al., “Incorporating Intelligence in Fog Computing for Big Data Analysis in Smart Cities,” in IEEE Trans. on Industrial Informatics, March 2017 2. B. Kehoe, et al., “A Survey of Research on Cloud Robotics and Automation,” in IEEE Trans. On Automation & Eng., VOL. 12, NO. 2, April 2015 3. J. Wan, et al., “Cloud Robotics: Current Status and Open Issues,” in Special Section in IEEE Access: The Plethora of Research in Internet of Things (IoT), 2016 4. N. Gangid, B. Sharma, “Cloud Computing and Robotics for Disaster Management,” in IEEE Inter. Conference on Intelligent Systems, Modelling and Simulation, 2016 5. J. Salmeron-Garcia, et al., “A Tradeoff Analysis of a Cloud-Based Robot Navigation Assistant Using Stereo Image Processing”, in IEEE Trans. On Automation Science and Engineering, Vol.12, No.2, April 2015 6. L. Riazuelo et al., “RoboEarth Semantic Mapping: A Cloud Enabled Knowledge-Based Approach,” in IEEE Trans. On Automation Science and Eng., VOL. 12, NO. 2, April 2015 7. K. Sugiura, K. Zettsu, “Rospeex: A Cloud Robotics Platform for Human-Robot Spoken Dialogues,” in IROS, Sept. 2015, pp. 6155-6060. 8. J. Mahler et al., “Dex-Net 1.0: A Cloud-Based Network of 3D Objects for Robust Grasp Planning Using a Multi-Armed Bandit Model with Correlated Rewards,” in ICRA, May 2016 9. S. Dey, A. Mukherjee, “Robotic SLAM - a Review from Fog Computing and Mobile Edge Computing Perspective,” in IEEE International Conference on Mobile and Ubiquitous Systems: Computing Networking and Services, 2016 10. M. Satyanarayanan, “Edge Computing for Situational Awareness,” in IEEE Local and Metropolitan Area Networks (LANMAN), 2017 11. K. Bilal, A. Erbad, “Edge Computing for Interactive Media and Video Streaming,” in IEEE International Conference on Fog and Mobile Edge Computing (FMEC), 2017. 12. S. Kamburugamuve, L. Christiansen, G. Fox, “A framework for real-time processing of sensor data in the cloud”, in Journal of Sensors, 2015 13. H. He, et al., “Cloud based Real-time Multi-Robot Collision Avoidance for Swarm Robotics”, in International Journal of Grid and Distributed Computing, Vol. 9, No.6, 2016, pp. 339- 358 14. S. Farokhi et al., “Performance Boost with Hybrid Cloud Robotics,” in International Conference on Intelligent Robots and Systems, Sept. 2017 15. A. Anjomshoaa, et al., “Quantifying the Anthropogenic Heat in Urban Areas Using Thermal Images,” in CSCI, 2016
  • 17. Contact Us • Soodeh Farokhi, Founder and CTO, C2RO Robotics, soodeh.farokhi@c2ro.com • Nazli Khan Beigi, Communication Specialist, C2RO Robotics, nazli.khanbeigi@c2ro.com • www.c2ro.com
  • 18. Additional Slides Use cases of C2RO Hybrid Cloud Robotics
  • 19. Use cases of C2RO Hybrid Cloud Robotics Object Recognition (OR): • The robotic platform is equipped with a single-board computer (SBC), i.e. Raspberry Pi 3, and an RGBD sensor. • In case that OR is performed only on the SBC, the performance of the process is ~0.07fps, and the user sees new images with detected objects every ~13 seconds. • While using our proposed model, the entire process improves to ~50fps, which translates into a real-time visualization of the detection process.
  • 20. Use cases of C2RO Hybrid Cloud Robotics Simultaneous Localization and Mapping (SLAM) • The robotic platform is equipped with a single-board computer (SBC), i.e. Raspberry Pi 3, and an RGBD sensor. • If the map is built locally on the SBC, the SLAM algorithm update rate is ~5Hz; • While by using our hybrid cloud robotics model, the map update rate was enhanced to 30Hz.
  • 22. Cloud Computation Topologies 1. Carries out all computation in one processor. Though this reduces the transmission of multi-robot state information between processors, but due to huge computations, the processor would be very busy and this leads into slow computation. 2. Assigns each computational component to one processor. This would require transfer of robots’ state information between processors, in addition to an imbalance of processing between delay-sensitive and non- delay-sensitive components. So, the overall delay of this topology was rather high too. 3. Parallel processing topology that unlike other parallel algorithms, would focus on entity or agent level parallelization and would study mainly computation resource scaling based on the computation load.
  • 23. Parallel Cloud Computation The topology is to prevent the delays caused by non-delay sensitive functions by combining the processes of the similar essence in one processor. Also, although each processor could run multiple instances in parallel, we made sure that the proper data types are sent to the processor instance that catches the right robot information. Hence, by proposing an innovative load balancing technique, we provide an efficient, low latency parallel computation algorithm
  • 24. Parallel Cloud Computation The same algorithm is implemented on SBC, a desktop computer server, a desktop computer GPU, and C2RO cloud, including 4, 4, 4, and 20 processing cores, respectively. Hence, by proposing an innovative load balancing technique, we provide an efficient, low latency parallel computation algorithm.

Hinweis der Redaktion

  1. Messages of 30kBytes, up to 50ms latency