Robot Thinkstock

Learning in the Sky: Collaborative Robots Embrace Cloud Computing

C2RO is a startup company that is introducing cloud computing to enable machine learning and collaborative solutions for robotic networks.

C2RO is a new innovative startup company that launched in October of 2016 out of the startup incubator TandemLaunch. Its mission is to introduce a new way for robots to collaborate via cloud computing. The C2RO developed a robot-agnostic software platform that streams in real-time data processing to provide cloud server-based artificial intelligence (AI) -enabled software solutions. This helps the robots by augmenting their perception and collaboration capabilities.

The software increases the intelligence and autonomy of the robots. It enables collaboration between several different robots to perform complex tasks. The robots connect via the software and their sensor data processes in real time in the cloud. This increases the way robots share knowledge, work in a joined network, enhance their sensor perception, and even helps robots that have limited sensing and computational resources.

“[C2RO has] developed a software platform that connects robots in the cloud—and processes their sensor data in real time—in a secure and reliable manner,” said Soodeh Farokhi, the company’s founder and technology lead, who holds a Ph.D. in cloud computing. “Using our software-only solution, robots can share knowledge using the latest advancements in artificial intelligence and computer vision algorithms.”

The C2RO platform uses the machine learning and vision systems of robot one to help teach other robots via its cloud connected software platform.

Cloud Robotics

“Cloud robotics” was coined in 2010 to describe network-connected robots using parallel computation and data sharing. The interconnection with other computer hardware and environments enable data sharing, and these environments are customizable based on customers’ needs. Connecting robots to the cloud dramatically enhances their efficiency and capabilities because it opens up possibilities for collaboration and collective learning.

C2RO offers a Multi-Robot Collaboration module, which connects multiple robots (regardless of type or manufacturer) so they can work together. There are two main benefits to collaboration: efficiency, such as when multiple robot vacuums are used to clean a large area; and complementary sensing, such as when drones and land robots combine their sensors and viewing angles in a search-and-rescue scenario.

Connected robots also tend to be lighter and cheaper, since they do not require expensive processors or memory localized on board the robot. Inexpensive robots can be made smarter with C2RO’s product because intensive processing and computation is partially offloaded onto the cloud platform, which can process the data in real time due to our patent-pending technology.

Partners for Custom Solutions

Speaking with Farokhi, engineers will see benefits from working in the cloud. Programming robots becomes an easier task, even for those without expertise in machine learning or software engineering, because all of the modules have been pre-programmed into the cloud as a software-only solution. According to Farokhi, “All the user has to do is install a package on his/her robot, and the platform has a dashboard that acts like an app store or ‘marketplace’ in a way that the user can choose the modules they want for a monthly, subscription-based fee for each module per robot.”

Cloud-based robotic networks can be measured by its response time—the total time it takes from when a request is made by a user until a response is received. Response time is in turn determined by two factors: latency and processing time. Processing time is the amount of time a system takes to process a given request, while latency is the delay incurred in communicating a message.

Soodeh Farokhi

According to Farokhi, the C2RO platform tackles the twin problems of processing time and latency. “We have brought down the processing time of any request to mere milliseconds using our parallel computing technology,” he explained, “and to address latency, we have partnered with PubNub—a telecommunication provider—providing a speedy connection, heighten security, and less network delays.” PubNub is a data stream network company headquarter in San Francisco.

C2RO has also collaborated with Indiana University’s School of Informatics and Computing. Professor Geoffrey Fox serves as a technical advisor for the company. With regard to C2RO’s software platform, “It will allow my cuddly robot, vacuum cleaner, high-speed lawnmowers, and even any smart home devices to share backend services and generate new ideas by linking their data together,” Fox said. “Our intelligent systems engineering department is uniquely positioned to support C2RO.”

C2RO is currently signing up users for its beta release.

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish