Hi Christoph,
On 12/05/2013 10:58 PM, Christoph Schultz wrote:
Hi group,
I recently joined the group, because I downloaded the meta-ros layer
and wanted to share my experience with ROS on an OpenEmbedded Linux.
The initial compilation went through without any serious issue. So
good job :-)
You're welcome---nice to see that the tool chain works smoothly with yet
another board :)
[...]
I would be interested in the performance level of your hardware you
are using ROS with. Best case would be, that I finally find out that I
just forgot to move some important files to my tmpfs, so I am facing
still the same bottleneck. Worst case would of course be, that the CPU
performance is just not sufficient for this python based environment...
I am using ROS with a Trim-Slice computer. That is based on a NVIDIA
Tegra2 SoC and has a dual-core A9 with 1 GHz, and 1GB DDR2-800 RAM.
Some HW stats of my OpenEmbedded System:
400 MHz x86 Quark CPU
256 MB RAM
8 GB SD card
I believe your board could just be on the lower border (or unfortunately
below that), on which you can run Linux, roscore and some basic ROS nodes.
A reasonable alternative is to run roscore on another machine and only
run the ROS nodes on the frontier of your node network, e.g. the sensor
preprocessing and the low-level actuator control, on your Galileo board.
The ROSCon presentation by Morgan Quigley [1] provides a good overview
of the different solutions to use ROS on an embedded device.
Furthermore, the ros-sig-embedded mailing list [2] can give you more
feedback, as the issue you are facing is most probably not specific to
the OpenEmbedded layer, but the performance of roscore on an embedded
device.
Lukas
[1] http://vimeo.com/66971366
[2] https://groups.google.com/forum/#!forum/ros-sig-embedded
