Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

added another camera for side view on quadrotor and tf on rviz #59

Open
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

iKrishneel
Copy link
Collaborator

Side Facing Camera
25m

Downward Facing Camera
frame0000

@k-okada
Copy link
Member

k-okada commented Mar 17, 2016

thanks, please add captured screen of rviz so that everyone can share how we can see world through the sensors.

@iKrishneel
Copy link
Collaborator Author

added the screen shot

@k-okada
Copy link
Member

k-okada commented Mar 17, 2016

can you complete this task with this camera?
I suppose

  • you can find the target on the starting phase
  • after you reaching to the target, it disappeared from the view, because facing camera only look at front side and downside camera is too narrow
  • you have to control the robot without knowing where the target is..
  • at some point, the target appeared in the downside camera, if you're lucky.

@k-okada
Copy link
Member

k-okada commented Mar 17, 2016

possible solutions are

  • put one more camera
  • add tilting actuator to the camera
  • change camera direction.

you can check thew camera view with http://jsk-visualization.readthedocs.org/en/latest/jsk_rviz_plugins/plugins/camera_info.html

@iKrishneel
Copy link
Collaborator Author

This side camera at the moment we are using to detect the moving vehicle. Yes as mentioned this is not sufficient for whole task, but at the moment we are trying to experiment simple method of detecting the target and moving to the track.

At the moment we are planning to equip the UAV with 2 or 3 RGB camera and Fish-Eye to cover wide field of view, perhaps to keep the arena in view always.

@k-okada
Copy link
Member

k-okada commented Mar 17, 2016

please try to put camera a little bit down direction, not parallel to the
horizontal plane, because you do not have to see upside. if you have 100
degree view , you can set 50 degree.

◉ Kei Okada

On Thu, Mar 17, 2016 at 8:28 PM, Krishneel [email protected] wrote:

This side camera at the moment we are using to detect the moving vehicle.
Yes as mentioned this is not sufficient for whole task, but at the moment
we are trying to experiment simple method of detecting the target and
moving to the track.

At the moment we are planning to equip the UAV with 2 or 3 RGB camera and
Fish-Eye to cover wide field of view, perhaps to keep the arena in view
always.


You are receiving this because you commented.
Reply to this email directly or view it on GitHub
#59 (comment)

@tongtybj
Copy link
Member

It seems that the head direction of drone is wrong... it is inverse. So when I put the command "go forward", the drone flyes backward acctually...
The reason is we did not set the correct yaw direction like "x" in https://github.com/start-jsk/jsk_mbzirc/blob/master/jsk_mbzirc_tasks/launch/jsk_mbzirc_task_1.launch#L30

And it seems like that the "yaw" is not given as rosparam in https://github.com/tu-darmstadt-ros-pkg/hector_quadrotor/blob/indigo-devel/hector_quadrotor_gazebo/launch/spawn_quadrotor.launch#L28

@tongtybj
Copy link
Member

Sorry for my ignorance....
Found the param "Y" in https://github.com/ros-simulation/gazebo_ros_pkgs/blob/jade-devel/gazebo_ros/scripts/spawn_model#L64

@iKrishneel In fact the side camera is set backward. We should set the drone ariframe in the correction direction first, then correct the camera direction in https://github.com/iKrishneel/jsk_mbzirc/blob/hector-camera/jsk_mbzirc_tasks/urdf/quadrotor_with_hokyo30lx_and_downward_cam.urdf.xacro#L32

@iKrishneel
Copy link
Collaborator Author

@tongtybj sorry that is correct, I did change that on my file but didnt send the PR. I think this PR should not be merged but as we talked, Gimbal will be best solution.

@k-okada
Copy link
Member

k-okada commented Mar 29, 2016

I don't think gimbal is a good idea, because of it requires more weight, if
you have high resolution image (xga is fine) with wide lens, it can provide
virtual gimbal
Please check opt camera http://www.optnagano.co.jp/product/nm_camera.html

2016年3月28日月曜日、[email protected]さんは書きました:

@tongtybj https://github.com/tongtybj sorry that is correct, I did
change that on my file but didnt send the PR. I think this PR should not be
merged but as we talked, Gimbal will be best solution.


You are receiving this because you commented.
Reply to this email directly or view it on GitHub
#59 (comment)

◉ Kei Okada

@tongtybj
Copy link
Member

@iKrishneel OK, acctually the direction of drone shoud be change about Yaw, becuase the recent head dirrection is wrong, this request is unrelated with the issue of camera. So, could you please send a pr of the correction of drone direction.

@k-okada I think both of idea is worth to try.
As to the gimbal solution, actually, the whole weight we design for drone is about 5Kg. and the wieght of gimbal for board type camera is less than 100g. That is the addition weight due to the gimbal is not a big matter for the flight time problem.
For the eyefish camera, thank you for sharing the info of opt camera. I will try the existing item in JSK like nm33. And I also watch another commercial board type camera(ids) like https://jp.ids-imaging.com/store/produkte/kameras/ids-interface/usb-3.0-gige/ids-housing/%E3%83%9C%E3%83%BC%E3%83%89%E3%83%AC%E3%83%99%E3%83%AB/page/3/mode/list.html.
I used to buy some board camera from here, they can provide a suitable lens according to our request.

@k-okada
Copy link
Member

k-okada commented Mar 30, 2016

if you put existing wide lens camera, you can try task 1 on simulation
immediately, I believe running task on simulation is the higher priority
then designing gimbal, compile opencv, ....

◉ Kei Okada

On Tue, Mar 29, 2016 at 7:03 PM, 趙 漠居 [email protected] wrote:

@iKrishneel https://github.com/iKrishneel OK, acctually the direction
of drone shoud be change about Yaw, becuase the recent head dirrection is
wrong, this request is unrelated with the issue of camera. So, could you
please send a pr of the correction of drone direction.
@k-okada https://github.com/k-okada I think both of idea is worth to
try.

As to the gimbal solution, actually, the whole weight we design for drone
is about 5Kg. and the wieght of gimbal for board type camera is less than
100g. That is the addition weight due to the gimbal is not a big matter for
the flight time problem.
For the eyefish camera, thank you for sharing the info of opt camera. I
will try the existing item in JSK like nm33. And I also watch another
commercial board type camera(ids) like
https://jp.ids-imaging.com/store/produkte/kameras/ids-interface/usb-3.0-gige/ids-housing/%E3%83%9C%E3%83%BC%E3%83%89%E3%83%AC%E3%83%99%E3%83%AB/page/3/mode/list.html
.
I used to buy some board camera from here, they can provide a suitable
lens according to our request.


You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
#59 (comment)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants