[ros-users] Use cases for ROS on ARM based computers

Fri Jan 25 10:29:02 UTC 2013

I agree with Marcus’ arguments: ROS on ARM is part of one end of the scope we discuss in SIG Embedded (and there are indeed a couple of other SIGs that are not really alive). Even if it’s also a bit of a problem to cover the whole bandwidth of what embedded can be (x86 Embedded PC down to 8bit µC), it is still important to have a forum where all these targets are discussed together. 

So it could make sense to rethink where to locate this “new” group and I feel like adding it to SIG Embedded makes more sense, as the motivation to use ARM mostly is an embedded application. Of course you can look at it from the other side and say, ROS on ARM is the end of the classic ROS target world, sure. I think the people who use ROS on ARM need to decide on which side they stand: do you have embedded applications that drive you using ARM or are there other reasons why you use these targets?

In our company we have the need to run ROS on ARM (to be more precise on Xilinx Zynq Cortex A9 MP, Linaro-based), and all is about embedded and integrating the target in products.

--Nik

Von: ros-users-bounces at code.ros.org [mailto:ros-users-bounces at code.ros.org] Im Auftrag von Marcus Liebhardt
Gesendet: Freitag, 25. Januar 2013 02:18
An: User discussions
Betreff: Re: [ros-users] Use cases for ROS on ARM based computers

I understand your points and in a crowded environment I would strongly support it. However, many SIGs are/seem to be very quiet. That's why I am worried about new SIGs dying even after a powerful start as it happened to many SIGs before (just have a look at the Fuerte&Groovy SIGs).

I think, the chances for a SIG to stay active are higher the more people are interacting in it. And even though the focus of ROS on ARM might be different than what has been discussed on the embedded list so far, both groups could still benefit from synergy effects, such as overlapping topics, help from experts etc. And last but not least, it would be easier for ARM/embedded newcomers to find a starting point (e.g. I'm using an ARM board, but which group is the right one for me (ARM Cortex A8/9 vs M3/4)?).

Regards,

Marcus

PS: In case you still want to separate, please clearly point out, who is your target audience for the list. Thanks.

On Fri, Jan 25, 2013 at 2:28 AM, YAMOKOSKI, JOHN D. (JSC-ER)[OCEANEERING SPACE SYSTEMS] <john.d.yamokoski at nasa.gov <mailto:john.d.yamokoski at nasa.gov> > wrote:

+1 

Having a focused SIG would promote better attention to details/issues in the ARM community than being 'embedded' within a larger, more diffuse group.

While ARM platforms are often associated with embedded systems, that's not always the case. For instance I wouldn't classify NVidia's CARMA board as 'embedded'  ( http://www.nvidia.com/object/carma-devkit.html). Damn thing has more horsepower than my desktop. 

I don't what this to digress into a 'definition of embedded' thread. But as suggested, keeping ARM separate is no different than other operating system specific SIGs.

On 01/24/2013 06:01 AM, Kevin Walchko wrote:

I think embedded is potentially too broad, and looking at the mail group quickly there appears to be a lot of different topics discussed (bandwidth, C++ vs C, RTOS for ROS, etc).

There seems to be a lot of interest in specifically ARM and supporting ROS specifically on ARM. I think if a SIG is going to produce something, it helps to be focused. There appears to be a lot of work to just setup the infrastructure, test, and distribute ROS packages for ARM. In order to achieve this, I think it should be its own SIG. I see no difference between this and the OSX SIG I support … which is a lot of work.

On Jan 23, 2013, at 11:50 PM, Brian Gerkey  <mailto:gerkey at osrfoundation.org> <gerkey at osrfoundation.org> wrote:

> On Wed, Jan 23, 2013 at 3:59 PM, Marcus Liebhardt
>  <mailto:marcus.liebhardt at rnd.yujinrobot.com> <marcus.liebhardt at rnd.yujinrobot.com> wrote:
>> On Thu, Jan 24, 2013 at 3:14 AM, Tully Foote  <mailto:tfoote at willowgarage.com> <tfoote at willowgarage.com>
>> wrote:
>>>
>>> Hi Everyone,
>>>
>>> Thanks for the interest.  Based on the interest we have setup a SIG for
>>> this.  I encourage you to join if you have time to participate.
>>
>>
>> I wonder, if it is necessary to create an extra SIG for this. Although the
>> group of people interested in ROS on ARM has grown to a significant size,
>> there are still not so many people out there.
>> In order to avoid people getting scattered all over the place (and might
>> even lost), I suggest we use the existing Embedded SIG
>> (http://www.ros.org/wiki/sig/Embedded) for this.
>> It gathers people interesting in and working on different embedded levels,
>> such as ROS on ARMs running Linux distributions and ROS on/with "bare-metal"
>> embedded boards.
>> If this list/group gets to busy in the future, we still could split it up in
>> different groups later.
>
> As a lurker on the ros-sig-embedded list, I agree with Markus.  Unless
> we really have two distinct communities, "embedded" and "arm," it
> would seem that one group is enough.
>
>    brian.
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org <mailto:ros-users at code.ros.org> 
> https://code.ros.org/mailman/listinfo/ros-users

_______________________________________________
ros-users mailing list
ros-users at code.ros.org <mailto:ros-users at code.ros.org> 
https://code.ros.org/mailman/listinfo/ros-users

_______________________________________________
ros-users mailing list
ros-users at code.ros.org <mailto:ros-users at code.ros.org> 
https://code.ros.org/mailman/listinfo/ros-users

-- 
Marcus Liebhardt 
Control Engineer
Yujin Robot
주소: 대한민국 서울시 금천구 가산동 345-30 남성프라자 #601, 153-023.
Address: Door #601, Namsung-Plaza, 345-30 Gasan-dong, Guemcheon-gu, Seoul, 153-023, Republic of Korea
Website: http://www.yujinrobot.com <http://www.yujinrobot.com/> 
Email: marcus.liebhardt at rnd.yujinrobot.com <mailto:marcus.liebhardt at rnd.yujinrobot.com> 
Phone: +82-70-46577073 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ros.org/pipermail/ros-users/attachments/20130125/6d7f36eb/attachment-0004.html>