Re: [ros-users] is a native ROS device possible?

Joq, this is true and we think it's a major aspect why you can't find
something similar out there yet.

At ROSCon, Morgan addressed basically the same idea we have in mind
developing SOMANET and rosxmos. The only difference is that he dreams of a
dynamic firmware, which means an OS-based concept where you can run
executables after booting the device. SOMANET nodes download a freshly built
custom firmware image including your application software at any time you
want to deploy a new version (whether from the Synapticon OBLAC cloud or an
update tool for your PC).

For some reasons, we generally want to avoid OSs for the
distributed/real-time part of any cyber-physical system, which SOMANET shall
be a complete platform for. Reliability, (real-time) performance and Stuxnet
& Co. being some of them.

Nik

-----Ursprüngliche Nachricht-----
Author: Jack O'Quin
Date: 2012-06-09 04:55 +200
To: User discussions
Subject: Re: [ros-users] is a native ROS device possible?

A big challenge is loading up-to-date firmware versions into the device.

(I believe Morgan mentioned that in his talk.)
--
joq