A 3D structured-light imaging system creates the human-like capabilities deployed in the design of a Robomotive humanoid robot jointly developed by Yaskawa Motoman, Beltech, and Robotiq. The so-called eyes of the system make it possible to identify a product and its surroundings (packaging or a pallet) and determine the product's orientation within the 3D space.
In integrating humanoid adaptive servo-grippers with 3D vision and smart software, the goal for the design is a new generation of robots that can lower tooling costs over conventional systems.
The human-like arms and grippers use common controls to work together or independently. For example, each arm lift a different product, and the arms can work together to assemble them. A seven-axis design (as opposed to the typical six-axis design) provides the flexibility to work with objects of different sizes.
An interesting aspect of this system is its targeting of factory automation applications. Potential uses include small bin picking setups where parts are constantly changing and robotic solutions would have been ruled out in the past. The system has already been deployed in small batch automation processes with large mix of parts, and tasks normally accomplished using manual labor. Applications where the humanoid robot excels include tasks where there is advantage of the two arms working in tandem, or if it is difficult for a one-armed robot to accomplish the motions in a limited working area.
A key technological advantage of this system, according to its makers, is that users don't need to spend time or money changing the hardware environment; the Robomotive can be placed in a workstation and trained to do the task a human would do. The user can load a lot of programs, so tasks can be switched easily and quickly.
This humanoid robot (featured in the Robomotive video below) highlights how this technology is being adapted to industrial environments. The ability to use 3D vision and coordinate two independent arms opens up new possibilities for automating repetitive manual tasks.