Hacker News Reader

Comment by deathhand | Hacker News Reader

Sadly this illustrates the current state of Robotics - HARD. This was 3 years with multiple people's lives. The industry as a whole needs to go down the same route as computers did which are common platforms that fit together that get the job done. Sadly there needs to be a behemoth in this space first(1960s IBM) that does all this work that society as a whole then can profit from(Facebook today)

View on HN

I'd say this reflects how problems in the physical world are hard in general. The robotics industry does have common platforms in the form of robotic arms used in automotive manufacturing. But they take away all the messiness and noise of the physical world by locking humans away and only allowing the arms to manipulate carefully designed prefab components. It's a similar situation in the world of computing because our silicon chips don't make full use of the analog design space. Rather, they stick to digital circuits with clearly defined voltage levels and clocking margins. By not pushing silicon design to its absolute limits and sticking with artificial restrictions that improve reliability, computer programmers don't have to worry about analog voltage drops across the chip and focus only on ones and zeros. Creating this same artificial environment for robots would involve turning the messy world we live in into a factory floor.

These multiple people were professor and their students. The main value for that particular project wasn’t coconuts nor product development, it was education. In that light the project is awesome, pretty sure the students learned tons of valuable things about hardware design debugging and testing, embedded software, coconut farming, and many others.

From product development point of view, I think a team of 3-5 smart and experienced people is likely to deliver better result in a couple of months, given reasonable amount of funding to purchase commercially available components.

Reply on news.ycombinator.com