MIT researchers develop VR teleoperation robot control system

Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory have developed a virtual-reality system to allow remote operation of complex robotic machinery through the use of an Oculus Rift headset, according to a report from MIT.

The platform functions around a VR control room that allows the operator to control the robot by taking the perspective of being inside the ‘head’ of it, using controllers to gesture and coordinate the movements of the system, according to the report.

“A system like this could eventually help humans supervise robots from a distance. By teleoperating robots from home, blue-collar workers would be able to tele-commute and benefit from the IT revolution just as white-collars workers do now,” lead author of a related paper on the system Jeffrey Lipton said in a press release.

Researchers are hopeful that the system could be used to “game-ify” production positions and increase job opportunities for unemployed individuals.

The system operates around the Baxter humanoid robot from Rethink Robotics, according to the report, and uses a hybrid model of VR robot control.

The platform allows the user to constantly receive visual feedback while controlling it and to take the perspective of being inside the head of the unit – two features that researchers say combine the two main approaches previously used in VR teleoperation.

“This contribution represents a major milestone in the effort to connect the user with the robot’s space in an intuitive, natural, and effective manner,” Stanford University CS professor Oussama Khativ, who is not involved with the paper, said in a press release.

While using the system, operators interact with controls that appear in the virtual space to open and close the robots hands as well as pick up, move and release items. The operator can also plan its movements based on a live display of the robot’s arms.

The team has tested the system by remotely performing tasks such as picking up screws or stapling wires, and later picking up and stacking blocks.

Researchers claim that when tested against state-of-the-art systems, the CSAIL VR system was better at grasping objects 95% of the time and 57% faster at performing tasks.

The group is currently focused on making the system more scalable and expanding the number of robots that would be compatible with the platform.