0
$\begingroup$

Rosanswers logo

Hi,

Just tried to see if hector slam algorithm works on raspberry pi3 with ubuntu MATE 16.04 w/ ROS kinetic. Unfortunately seems that the computational power of the board is not enough to handle the data stream while map is created so our discover is that raspberry is not enough powerful (we tried the commands on another pc and the map is created successfully).

So now our question is: Is it possible to stream the data coming from laser scanner (connected to Pi3) to another powerful pc. Do proper elaborations and then send final commands to the raspy again (cmd_vel..). If yes which connection should we use? And how to do that? Any hints?

Thanks


Originally posted by matteopantano on ROS Answers with karma: 32 on 2017-01-15

Post score: 0

$\endgroup$

1 Answer 1

0
$\begingroup$

Rosanswers logo

HI

Short answer is yes. The pi has plenty enough power - I would ensure you have a high bandwidth WiFi channel - I use a 300Mps adapter on a Raspberry Pi 2 for the set up you mention - works fine. 150Mps is a bit slow.

Mark


Originally posted by MarkyMark2012 with karma: 1834 on 2017-01-15

This answer was ACCEPTED on the original site

Post score: 0


Original comments

Comment by gvdhoorn on 2017-01-15:
Additionally: make sure to compile things with optimisations enabled. With CMake / catkin: -DCMAKE_BUILT_TYPE=Release (or RelWithDebInfo).

Comment by matteopantano on 2017-01-15:
How can stream data from raspy and then analyze it in the other PC?

Comment by MarkyMark2012 on 2017-01-15:
Simply install ROS on the other PC and run a Hector SLAM node there listening to the topics it required.

This tutorial should help re: running ROS on multiple machines - link text

Comment by matteopantano on 2017-01-16:
Thanks guys! :)

Comment by gvdhoorn on 2017-01-16:
If you feel your question was answered, please don't close it, but accept the answer by @MarkyMark2012 by ticking the checkmark to the left his answer. Thanks.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.