Kobuki autodocking

[Wouter Lammens]

I have some questions the precise working principle of the Kobuki autodocking algorithm <

http://wiki.ros.org/kobuki/Tutorials/Testing%20Automatic%20Docking>:

- What is M_PI in this part (line 397 and further) of the code <

https://github.com/yujinrobot/kobuki/blob/indigo/kobuki_random_walker/include/kobuki_random_walker/random_walker_controller.hpp> and how it is given a value? 

I still have some questions that remain unanswered in the tutorial page.

- Can more than one infrared sensor on the Kobuki base receive a signal from the docking station, at the same time. I was wondering this since the sensors are oriented in a special way (see tutorial page)?

- What do all the numbers mean that are shown when I subscribe to the docking IR topic? On this page <

https://yujinrobot.github.io/kobuki/doxygen/enAppendixProtocolSpecification.html> I see certain numbers (

0x01 for NEAR_LEFT state
0x02 for NEAR_CENTER state et cetera). But on the topic there are different numbers: 

48 2 61 56 16 25 24 for the middle sensor,

2 4 8 16 32 for the other two. So what do these numbers mean?

- How does the Turtlebot keep a straight line to its docking station? Is there some kind of feedback mechanism that makes sure it goes straight? In the tutorial there is nothing mentioned about this part... I don't really see how this can work, since I understand that the signals are binary: e.g. the Turtlebot is either in a far field or not. Is there also a control mechanism  when the Turtlebot moves from a left or right field to the central field?

Kind regards,

Wouter

[Daniel Stonier]

On Thu, Apr 14, 2016 at 5:55 AM, Wouter Lammens <wouter....@student.kuleuven.be> wrote:

I have some questions the precise working principle of the Kobuki autodocking algorithm <

http://wiki.ros.org/kobuki/Tutorials/Testing%20Automatic%20Docking>:

- What is M_PI in this part (line 397 and further) of the code <

https://github.com/yujinrobot/kobuki/blob/indigo/kobuki_random_walker/include/kobuki_random_walker/random_walker_controller.hpp> and how it is given a value? 

It's a not very cross-platform definition for pi by the gnu libc library.

 

I still have some questions that remain unanswered in the tutorial page.

- Can more than one infrared sensor on the Kobuki base receive a signal from the docking station, at the same time. I was wondering this since the sensors are oriented in a special way (see tutorial page)?

- What do all the numbers mean that are shown when I subscribe to the docking IR topic? On this page <

https://yujinrobot.github.io/kobuki/doxygen/enAppendixProtocolSpecification.html> I see certain numbers (

0x01 for NEAR_LEFT state
0x02 for NEAR_CENTER state et cetera). But on the topic there are different numbers: 

48 2 61 56 16 25 24 for the middle sensor,

2 4 8 16 32 for the other two. So what do these numbers mean?

They are quite noisy and on/off. Note that the enums are bit values, so 56 = 32 + 16 + 8 = FAR_RIGHT + FAR_LEFT + FAR_CENTER. i.e. all three signals are detected.

The actual algorithm is simply described on the wiki page - it simply shoots for the centre and then follows a straight line to the dock. The actual algorithm is exposed in kobuki_dock_drive with the ros wrapper to it in kobuki_auto_docking which uses these values directly. Poke through that to ascertain the feedback mechanism in its gory detail.

Daniel

.

- How does the Turtlebot keep a straight line to its docking station? Is there some kind of feedback mechanism that makes sure it goes straight? In the tutorial there is nothing mentioned about this part... I don't really see how this can work, since I understand that the signals are binary: e.g. the Turtlebot is either in a far field or not. Is there also a control mechanism  when the Turtlebot moves from a left or right field to the central field?

Kind regards,

Wouter

--
Kobuki's nest: kobuki.yujinrobot.com
---
You received this message because you are subscribed to the Google Groups "Kobuki Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kobuki-users...@yujinrobot.com.
To post to this group, send email to kobuki...@yujinrobot.com.
Visit this group at https://groups.google.com/a/yujinrobot.com/group/kobuki-users/.