Cognicept Systems develop a human-in-the-loop teleoperation service to reduce robot failure rates

Highlights

• Cognicept Systems has developed a human-in-the-loop teleoperation service that can be utilised to reduce mobile robot failure rates.
• This remote robot technology allows humans to intervene and access robot cameras when they fail and set them back on course.
• Cognicept estimate robot failure rates could be reduced from 25% to just 1%.

The recent firing of 120 robots and rehiring of human staff to replace them at the Henn Na Hotel in Sasebo, has sparked the debate about the failure rate reality of robots, and the need for human intervention for robots to operate economically.

Robot failure can occur for a number of reasons, including path-finding failures and localisation failures, all which can be rectified with human intervention. However, with the number of robots on the rise, this level of human intervention will also increase.

Michael Sayre, CEO of Singapore start-up, Cognicept Systems, alongside founders Ruchit Rami and Alok Pathak, have developed a human-in-the-loop teleoperation service that can be utilised with mobile robots. In their Singapore teleoperations centre, 14 remote operators can intervene when a robot fails and access the view through the robot’s cameras. The remote operators are able to assume control and steer the robot out of trouble or back on course. Cognicept estimate robot failure rates could be reduced from 25% to just 1%.

Considerations to be taken into account include legal and privacy issues as well as operator burnout from switching between multiple robots.

About Cognicept Systems

Cognicept Systems is equipment uptime-as-a-service. We enable Human-in-the-loop (HITL) error handling with our telerobotic networking technology and human remote robot trainers. Our system allows a “confused” robot to dial out to a human operator, who can see what the robot sees, and control it intuitively, bringing the robot back to normal operation. Cognicept makes unpredictable applications reliable and enables use-cases that were previously impossible.

Register Interest