CA Technologies is adding collaborative robot research and development to its already-packed product dance card.
The New York-based company already has bustling businesses in mainframe software, automation, API distribution, security and several others. Nonetheless, on Jan. 17 CA revealed that it is moving to partner on something called cobotics--a collaborative robotics project--by working alongside Finland’s Tampere University of Technology and Finnish IT software and service company Tieto.
The project has been funded jointly by those three organizations along with Tekes, the Finnish Funding Agency for Innovation.
A cobot or co-robot (from “collaborative robot”) is a robot intended to physically interact with humans in a shared workspace. This is in contrast with conventional robots, which are designed to operate autonomously or with limited guidance.
Cobots were invented in 1996 by J. Edward Colgate and Michael Peshkin, professors at Northwestern University. A 1997 U.S. patent filing describes cobots as "an apparatus and method for direct physical interaction between a person and a general purpose manipulator controlled by a computer."
Goal is to Build Reliable Human-to-Robot Workflows
The CA-Tampere University cobotics collaboration project will attempt to solve the challenges of building safe, secure and effective human-to-robot workflows. Ensuring proper control and execution of these workflows and understanding their information requirements is central to maximizing the potential of robot-human collaboration, CA CTO Otto Berkes told eWEEK.
“Robotics has reached a stage where the value that robots provide can be significantly amplified by enabling them to truly collaborate with humans,” Berkes said. “Developing the right models and algorithms for that collaboration, while paying careful attention to safety and cybersecurity will define the next phase in the robotics field.”
“We believe that collaborative robotics is the fastest-growing area in the domain of robotics, and we believe that our expertise in communications, policy management, security and visualization, will enhance the project.”
The first phase of the project is expected to conclude at the end of December 2018.
eWEEK asked Berkes a few questions about this project.
Q: What exactly are human-to-robot workflows? Are these cobots being trained to anticipate a human’s direction?
A: “This is more about human-to-robot interaction,” Berkes told eWEEK. “The robot understands both its role, its position spatially, and its position in relationship to other robots and humans. They all collaborate, working on a task or tasks. For example, two robots can hold half a component each while the human connects the two halves. The workflow is to build the component, the robot task is hold the component and the human task is connect the component. They are complimentary.”
Q: What do you see as some of the early use cases?
“Simple ones are already in use, think about a robot arm that extends the human reach, but we will be seeing more sophisticated ones soon,” Berkes said. “One of the use cases we are investigating is where a robot is passing screws to a human to assemble a complex machine. The robot senses the screw is the wrong weight and will cause the machine to fail. The robot discards the screw, passes a good screw on and makes sure there are enough screws to finish the job, all without human intervention.”
eWEEK keeps a close eye on IT innovation such as this. For more information, go here.
Image courtesy of YouTube