Re: [ros-users] Capability of ARM dev board for ROS host

Top Page
Attachments:
Message as email
+ (text/plain)
Delete this message
Reply to this message
Author: User discussions
Date:  
To: ros-users
Subject: Re: [ros-users] Capability of ARM dev board for ROS host
On 01/14/2012 02:17 AM, Austin Hendrix wrote:
> My personal robot is running the WG navigation stack with a Hokuyo
> URG-04LX-UG01 on a pandaboard. I'm running significantly reduced
> settings for update rates, map size and resolution, local and global
> planners, and it performs moderately well. I don't have a vision
> system yet, so I don't know how well it will perform with that
> additional load.


I am running ROS and navigation stack on Pandaboard too (but no laser
input). Slowest process is move_base which occupies 100% of CPU all
time. All other nodes are idle. I had no motivation yet but at some
point I dig into move_base sources and look what I can optimize and
maybe I just throw some less needed functionality out.

My robot does most of decicions based on vision and I assumed
transporting just images is something I can avoid. So I took existing
camera node and injected all neccessary detection code in it. So my code
publishes just ~10 coordinates for each frame which is piece of cake to
handle for rest of system.

Whatever is slow for you, try to find exact stack and node and then
problem becomes much easier.

    Tõnu