Robotics, manufacturing and the future

If there is one news article on call-to-action to read this Spring this is it

Europe and US have slowly hemorrhaged manufacturing capabilities in the last 40 years. Not just in cars, equipment, appliances, but also advanced manufacturing of microchips, sensor, processors.

Research funding to innovate in robotics and manufacturing come from all over the world, not just US, but a lot of the innovations until year 2020 or so were. Yet we failed to translate research into a revolution that can bring back to our home countries cheap and fast manufacturing.

Why?

1- because of a short investment cycle that forces to productize in 12-18 months in a field that required more time and more iterations

2- the BARRIER of physical embodied manipulation and locomotion data, and the lack of coherent robotic platforms

3- the scarce interaction between manufacturing and the robotic research products: often not fancy robots but simple contraptions win in a factory. One of a kind machines, but built out of manufacturing components that are often the same at different scales and operating on different materials

4- we failed to create a cheap factory of factories, where iterating on design can be fast, instead we shipped it all happily abroad, losing speed along the way

WHAT CAN WE DO?

• Co-locate research and manufacturing test facilities. Everyone loves a desk job, but robotics needs constant testing in the field
• More rapid prototyping at home. Production can still be elsewhere, but fast iterations on design are paramount
• Keep applied research money flowing: grants and government funding should come from active collaborations with industry
• Create factory cities, where fast manufacturing and prototypes can be developed all or most in one location
• Train the workforce to operate machines, to produce multiples of their manual capability, to think and devise and implement. Shift from passive jobs to more active roles and mentality: “how can I improve this?”, “”how can I make more money for the company that employs me?”

YES BUT:

• These plans may sound good, but unless there is strong home-based manufacturing and sales of robotics, they cannot succeed
• We still do not have proper brains for robots that can operate mostly autonomously - this may be an advantage, as it gives us some time to think and plan

about the author

I have more than 20 years of experience in neural networks in both hardware and software (a rare combination). About me: Medium, webpage, Scholar, LinkedIn.

If you found this article useful, please consider a donation to support more tutorials and blogs. Any contribution can make a difference!