# Move Base setup for Magni **URL:** \ **Category:** Uncategorized\ **Created:** [January 29, 2021, 5:50am UTC](https://forum.ubiquityrobotics.com/t/move-base-setup-for-magni/717 "2021-01-29T05:50:27Z")\ **Posts on this page:** 4\ **Page:** 1
**Author:** ![Ivano](https://forum.ubiquityrobotics.com/letter_avatar_proxy/v4/letter/i/edb3f5/32.png) [@Ivano](https://forum.ubiquityrobotics.com/u/Ivano)\ **Post date:** [January 29, 2021, 5:50am UTC](https://forum.ubiquityrobotics.com/t/move-base-setup-for-magni/717/1 "2021-01-29T05:50:27Z")
Hi, I am trying to get my move base to work for navigating my robot. I am getting my Lidar next week so I cant really test it yet but would like to create my launch files so long. I did use :~$ locate move\_base.launch and found a launch file that already exists on my PC with the following in: launch\> ode pkg=“move\_base” type=“move\_base” respawn=“false” name=“move\_base” output=“screen”\> rosparam file="$(find magni\_nav)/param/costmap\_common\_param.yaml" command=“load” ns=“local\_costmap” /\> rosparam file="$(find magni\_nav)/param/costmap\_common\_param.yaml" command=“load” ns=“global\_costmap” /\> rosparam file="$(find magni\_nav)/param/local\_costmap\_param.yaml" command=“load” /\> rosparam file="$(find magni\_nav)/param/global\_costmap\_param.yaml" command=“load” /\> rosparam file="$(find magni\_nav)/param/base\_local\_planner\_param.yaml" command=“load” /\> /node\> /launch\> (I Removed the \< infront of each line so it shows up on the post) I can send you what each yaml file contains if you’d like but id just like to know whether this launch file is the correct one to use, whether I should edit it or whether I should try to create a new one. ---
**Author:** ![Ivano](https://forum.ubiquityrobotics.com/letter_avatar_proxy/v4/letter/i/edb3f5/32.png) [@Ivano](https://forum.ubiquityrobotics.com/u/Ivano)\ **Post date:** [January 29, 2021, 6:00am UTC](https://forum.ubiquityrobotics.com/t/move-base-setup-for-magni/717/2 "2021-01-29T06:00:21Z")
I have started to try and create my own launch file using the instructions set out on this page [http://wiki.ros.org/navigation/Tutorials/RobotSetup](http://wiki.ros.org/navigation/Tutorials/RobotSetup). However, I am really confused as to what to put into the parameters of the odometry section. I know me not having the Lidar yet does make this task more difficult but I’d like to make some progress in the meantime. ---
**Author:** ![mjstn2011](https://forum.ubiquityrobotics.com/user_avatar/forum.ubiquityrobotics.com/mjstn2011/32/30_2.png) [@mjstn2011](https://forum.ubiquityrobotics.com/u/mjstn2011)\ **Post date:** [January 29, 2021, 9:17am UTC](https://forum.ubiquityrobotics.com/t/move-base-setup-for-magni/717/3 "2021-01-29T09:17:41Z")
You should look at this page to help you get started: > **[Lidar-Based Localization](https://learn.ubiquityrobotics.com/lidar_navigation)** > > Tutorials for Magni The ODOM is already setup for the motor node, ubiquity\_motor repository. If you are curious, you can look at the config file that has the defines ODOM uses is here: /opt/ros/kinetic/share/magni\_bringup/param/base.yaml ---
**Author:** ![inmoov](https://forum.ubiquityrobotics.com/user_avatar/forum.ubiquityrobotics.com/inmoov/32/473_2.png) [@inmoov](https://forum.ubiquityrobotics.com/u/inmoov)\ **Post date:** [July 17, 2021, 7:17am UTC](https://forum.ubiquityrobotics.com/t/move-base-setup-for-magni/717/4 "2021-07-17T07:17:18Z")
Hi, Since I found this post about move base , I wll post here, I had Magni for a while now and I still couldnt get Magni’s move base to work properly for path planning and simple obstacle avoidance even in Gazebo Simulation environment, it is going no where, are you planning to fix Magni’s move base at ALL? …(and Now I have to make tonnes and tonnes of efforts, spending months, weeks and nights trying to sort out Magni’s Move base to work properly, I know you have made move\_basic but that does not help at all in real life situation…with no path planning and no obstacle avoidance which is not much use …besides just going in a straight line with nothing blocking it… ) Can you please tell me know the proper parameters? for the parameter YAMLs below, especially for base\_local\_planner\_param.yaml , costmap\_common, global\_costmap and local\_costmap , I have been tweaking alot and lot and still Magni in Gazebo (turtlebot’s playground world environment for example), cannot get even pass ONE obstacle… and it just fails) ![image](https://frm.ubiquityrobotics.com/uploads/default/original/1X/ded234d1573b5da7736c4236a4e7a5ec3f47f126.png) below are the parameters I had been experimenting, testing, tweaking with no Luck!!! global costmap ![image](https://frm.ubiquityrobotics.com/uploads/default/original/1X/1720012302eb36bc467eb14303f23e94212edf61.png) local costmap ![image](https://frm.ubiquityrobotics.com/uploads/default/original/1X/2e493cdea2dc4a2387d607426a5239c735374fa8.png) I changed using the navfnROS Global planner because the default carrot planner just hits the obstacles and walls straight away where navfnROS at least will stop in front of an obstacle/wall , eventhough it does not follow the planned path to go around it … I had also tweaked the cost scaling factor, does not make much difference , inflation radius I changed to 0.1 and no still no luck These below keeps happening constantly… ![image](https://frm.ubiquityrobotics.com/uploads/default/original/1X/a212b21c867b8b7ea260b647867476ff844bef70.png) This is the simple playground world I use to simulate and even that failed , Magni just stuck with there with the Global and local planner ![image](https://frm.ubiquityrobotics.com/uploads/default/original/1X/997af4a084dd0d3ec244fe91e674a5f189a46b39.jpeg) ![image](https://frm.ubiquityrobotics.com/uploads/default/original/1X/92fedce201498fc482a023cdd0fba82906ba2c6b.jpeg) now if your Magni move base code comes at a cost, please let me know how much then? I will consider to buy instead spending months and months with no results…on my Magni, Also, I tried in Noetic, with the same move base config, it even got worse, it moves a little bit and far from the global planned path and just stuck there,…