Robot manufacturers such as ABB, Fanuc and Keyence expect a slow but steady expansion into the food industry as well as new types of manufacturing and adjacent niches of distribution and agriculture.
Japanese company Yaskawa Electric are currently experimenting with rice and tofu. Yes, you read it correctly. Yaskawa Electric is tackling the convenience store, a $7bn dollar market capitalisation in Japan.
As funny as it sounds, rice, tofu, and vegetables have an irregular shape making them hard to grasp. Imagine yourself trying to make/create a sushi roll with chop sticks. Tedious and time consuming, something that business owners cannot afford.
QC? (Quality Control)
Within the manufacturing area there is an increase in robotic quality control inspections. While products roll off the robot assembly line there are still people at the end of the line that are inspecting the final product.
An example could be testing laptops that have to be inspected visually as well as functionally to see if all the buttons functions are working properly. For this reason Omron, specialised in sensing and control for factory robots is working on an image sensor and artificial intelligence combination to improve visual quality control and detect scratches and scuffs on a finished product.
Ask a robotics companies what they can do and they will enumerate endless possibilities. A more interesting question is to ask robotics companies what cobots cannot do at the current time.
A. 10kg limit
Cobots cannot exert a large force. Kg for kg the average human is usually ten times stronger than a cobot.
B. Environment sensing
Image sensing or computer vision has evolved exponentially over the last decade, however cobots are still having a hard time distinguishing what an object is and deciding how to move it.
C. Cybersecurity issues
To increase a cobot intelligence a big AI system must be implemented. Since AI systems are physically “clunky” all the cobots have to be connected to the network sparking a possible security problem.