For a number of weeks I have tried various methods of getting the iRobot base to move using the serial mini-Din pin to send serial commands to control the robot. I could not debug using neither the macbook nor the Xavier when working with Omar and we decided to try to use the Arduino method to by using the DB25 connector. However, the Arduino method also proved to have the same result as the prior two methods, which was that the robot base acknowledged that a device was connected to it and was able to send signals, but did not perform the actions sent via requests from the macbook, the Xavier, and the Arduino. This effort started around April 13th and lasted until April 20th, when the group decided that it was best for me to just send over the parts for the robot arm and the gripper to Omar, who constructed a simple robot using 2 servos to test his motion planning algorithm as he was the one who possessed the depth camera.
One problem that I ran into while building the full robot arm was that the servos behaved differently when the arm was attached. Below are videos that shows the servos working as intended without the arm and not working as intended with the arm.
As seen here, the servo that is responsible for the rotation of the arm itself is none-responsive when the arm is actually there. This issue is strange because the servo is listed to be able to handle 20kg of force with 5V and 1.2A, which is what I am running here.
Next week, I aim to help Omar integrate all the parts onto his backup robot.