Robotics

What is your greatest lesson in robot making?

If you intend to share your experience, please consider selecting before going down the thread (if any :)), to reduce bias.

Scientist and engineer interested in Life.

Work on (GOF)AI, Agriculture, Biology, Energy, Robotics, Space---all topics climate-aware.

To me, controlling the environment has always been the best way to get results in robotics.

Be it designing a warehouse, structuring light, or just finding generic but concrete conditions (e.g. a typical home for Roombas).

creating and destroying; built flying cars & autonomous vehicles, now building /faust, more here: /july, web: july.rocks

Great list, shared experience.

Would you expand on the greatness of simulation environments?

Some that come to mind:
- getting to a product that works takes a really long time
- getting to a product that delivers value for customer is even longer
- Simulation environments are great but not for the reasons that you think
- root causing issues takes a long time (you don’t know if it’s sw or hw) - HIL tests go a long way
- changes in hardware once you deploy to customers is really hard
- building things that are useful is really hard
- premature optimization in software is better than premature optimization in hardware
- sensors are hard to debug 
- don’t try to reinvent everything