`Zau_INESCTEC` branch

[JorgeFernandes-Git]

This branch will integrate all the necessary modifications required to apply the methodology that has been developed thus far, to the actual Zau robot that is situated in IILAB.

https://github.com/JorgeFernandes-Git/zau_bot/tree/Zau_INESCTEC

[JorgeFernandes-Git]

During the visit to IILab on Friday, the following skills and knowledge were acquired:

• Establishing a secure shell (SSH) connection with Zau.
• Familiarity with the process of powering on and off Zau.
• Competence in controlling the AGV.
• Ability to manipulate the manipulator arm.
• Understanding of Zau's ROS topics.
• Knowledge of Zau's commands and launch procedures.
• Utilizing Zau's odometry.
• Successful acquisition of an image using the Realsense camera mounted on the end-effector.
• Accurate measurement of actual dimensions to adapt the URDF file for use in calibration.
• Creation of credentials for establishing an SSH connection with Zau.

During the next visit to IILab, the following tasks will be completed:

• Verification of the established SSH credentials to connect with Zau.
• Intrinsic cameras calibration.
• Recording of a bag file using a single camera.
• Recording of a bag file using two cameras and odometry.

@rarrais and @miguelriemoliveira I will need a second camera for the AGV.

Considering the current setup, it would be ideal to use another Realsense D415 camera identical to the one utilized with the manipulator arm, as this would ensure compatibility since the existing camera drives are already functioning correctly.

I would appreciate hearing your thoughts on this matter and welcome any alternative suggestions you may have.

[JorgeFernandes-Git]

Hello @sergiodmlteixeira.

This is an update on the progress made since friday. This is an image of the adapted URDF file, which now accurately reflects the dimensions we took. Furthermore, I have made the necessary modifications to the cameras to accommodate the Realsense D415.

Regarding our previous concerns about compatibility between Melodic and Noetic versions of ROS, I have found information on a relevant ROS forum. According to the provided information, it is unlikely that any significant compatibility issues will arise as a result of running Melodic and Noetic versions together. https://answers.ros.org/question/305764/do-local-and-remote-machines-have-to-run-the-same-version-of-ros/

In light of these updates, I believe it would be prudent to schedule our next visit.

[miguelriemoliveira]

@rarrais and @miguelriemoliveira I will need a second camera for the AGV.

Considering the current setup, it would be ideal to use another Realsense D415 camera identical to the one utilized with the manipulator arm, as this would ensure compatibility since the existing camera drives are already functioning correctly.

Hi @JorgeFernandes-Git ,

If @rarrais cannot find one I can lend you one, but not that model or brand.

[rarrais]

Hi @JorgeFernandes-Git ,

We have available at the lab the following cameras: Intel Realsense VF0800, Orbbec Persee, Orbbec Astra s. @miguelriemoliveira , feel free to suggest which would be the best (or less bad!).

Thank you and sorry for the response delay.

[JorgeFernandes-Git]

I spoke with @miguelriemoliveira, who informed me that any camera would suffice for the task, including a common webcam.

I simply need to convert the image to a ROS message using the tool available at http://wiki.ros.org/usb_cam. In the case of ROS-compatible cameras, this step is unnecessary.

I am considering a visit to IILAB next week to determine which camera would be most suitable.

@rarrais, could you please provide insights on which camera would be more easily available? I will then proceed with attempting to configure the selected camera.

[miguelriemoliveira]

Hi,

I think the camera is not very important, since we are only using the rgb. In any case, if we can choose, I would select one with the same resolution as the other that is already installed.

Both must be intrinsically calibrated before recording a bagfile.