Cloud robotics
2 likes944 views
The document introduces cloud robotics using ROS (Robot Operating System). It discusses how cloud computing can help address limitations in robot processing and storage capabilities by moving these functions to remote servers. Key benefits highlighted include lower costs, longer battery life, and the ability to share knowledge between robots. While cloud robotics is still developing, some implementations so far include RoboEarth, Google's street view robot, and ROS frameworks used with robots like PR2 and Nao.
Cloud robotics
- 1. Welcome
- 2. CLOUD ROBOTICS WITH ROS ( ROBOTIC OPERATING SYSTEM ) Titto Thomas Roll No.47 S7 A College of Engineering, Chengannur
- 3. Robotics Robots - mechanical intelligent agent which can perform tasks on its own, or with guidance They can be industrial, mobile, service , humanoids..etc Touches every instant of life in one way or the other Robotics - deals with design, construction, operation, structural disposition, manufacture and application of robots Dynamically growing industry and its in need of alternative and new ways College of Engineering Chengannur 1
- 4. Where do we stand ? Require great computational capabilities to be able to process – in real time – images, faces, voice, sounds, obstacles many more Necessary to have a very powerful – and high energy consuming – processor and peripherals on board. Necessary to carry huge and heavy batteries as well College of Engineering Chengannur 2
- 5. ROS (Robotic Operating System ) Software framework for robot software development developed in 2007 under the name switchyard by the Stanford Artificial Intelligence Laboratory in support of the STAIR Based on graph architecture & is geared toward a Unix-like system Doesn't have to ”reinvent the wheel ” College of Engineering Chengannur 3
- 6. ROS is.. Peer-to-peer Multi-lingual Tools-based Thin Free and Open- Source Exponentially growing Used in many applications including Willow Garage’s PR2, Aldebaran’s Nao and Shadow’s robotic hand College of Engineering Chengannur 4
- 7. Cloud Computing Using multiple server computers via a digital network, as though they were one computer the services available are considered part of cloud College of Engineering Chengannur 5
- 8. Cloud Provides.. Device and location independence Significant workload shift from the local computers Cost is reduced as capital expenditure is converted to operational expenditure Sharing of resources and costs across a large pool of users Highly reliability and efficiency Improved security due to centralization of data Easiness in supporting and improving a software Scalable - dynamic provisioning of resources College of Engineering Chengannur 7
- 9. ROS in the Cloud College of Engineering Chengannur 7
- 10. High Level Overview Server Computer Cluster jobs & sensor data Cloud Computing Framework Ros ROS Messages Messages over http Public Cloud ROS Framework Private (Internet) Cloud (Intranet) ROS Messages over http College of Engineering Chengannur 8
- 11. Architechture In Layers Storage / Compute Nodes Distributed File System and necessary services Cloud Framework Master & Subscriber Nodes ROS Nodes College of Engineering Chengannur 9
- 12. ROS Messaging Example Publication Topic Subscription Node Node Service Invocation College of Engineering Chengannur 10
- 13. Cloud Enabled Robots Can.. Understand Perceive Share & REACT College of Engineering Chengannur 11
- 14. Why should we ?? Off the shelf hardware means affordable robots Lower the barrier to entry for robotics Scalable CPU, memory, and storage shared knowledge database hardware upgrades are invisible & hassle-free Longer battery life College of Engineering Chengannur 12
- 15. lighter, easier-to- maintain hardware (akin to desktop PC vs. a thin- client “netbook”) Robot experiences / history / behavior outcomes / learned skills can all be published or data mined Expanding the knowledge beyond “Physical Body” Cheaper , Lighter & Smarter College of Engineering Chengannur 13
- 16. Needs Improvisation Cloud Computing is still taking baby steps, so will have to wait for the platforms to develop cloud-based applications can get slow or simply become unavailable leaving the robot “Brainless” Tasks that involve real-time execution require onboard processing College of Engineering Chengannur 14
- 17. Implementations till now.. ● RoboEarth - to develop a “World Wide Web for robots” ● ASORO’s Cloud Computing Infrastructure ● Google’s initiative with Android ● Nao Humanoid Robots ● Google’s street View Robot ● DAviCi framework ● Project Phondox ● Turtlebot from google College of Engineering Chengannur 15
- 18. Check it out.. College of Engineering Chengannur 16
- 19. THANK YOU..!!
- 20. ANY QUESTIONS ??