Quad-Thruster Autonomous Underwater Vehicle

MAR 10

Tested the teleop for the first time. We were surprised that it worked so well on the first try !

MAR 10

Fully tested the webcam and the ball detection algorithm. It is not perfect but it should be good enough over time. Will need to try and fine tune the algorithm.

camera test shot 1 camera test shot 2

MAR 08

Completed initial driver for the ultrasound sensor.

MAR 08

Got camera working on beaglebone black. Had to deal with some USB connection issues.

first camera test shot

MAR 07

Started driver for ultrasound sensor. Used a voltage divider to get the sensor output voltage below 1.8V since Beaglebone Black only accepts 1.8V on its analog inputs.

MAR 04

Fully assembled the submarine with all electricals wired together. Tested and waterproofed motors.

back look of the submarine front look of the submarine top look of the submarine

MAR 01

Purchased a chip to convert 3.3V logic level of PWM pins on Beaglebone Black to 5.0V required by the Electronic Speed Controllers. The motors are now able to be controlled.

MAR 03

Attached the four brushless DC motors onto the submarine with Powerpole.

FEB 27

Modified the IMU node to send quaternions instead of euler angles to the controller.

FEB 25

Finished basic controller and keyboard teleop funcion for the submarine.

FEB 18

Programmed ESC for the brushless motors on BeagleBone Black.

FEB 17

Waterproofed the submarine, leaving the submarine in the water tank for 18 hours for testing.

waterproof test 1 waterproof test 2 waterproof the end

FEB 15

Finally cross-compiled and installed the IMU ROS node after many cross-compile and architecture issues.

FEB 14

Finally received the ultrasound sensor for the submarine and soldered wires to the various pins. Will now be able to start developing the driver for it.

ultrasound out of package

FEB 12

Recompiled the BeagleBone Black PWM kernel so that different duty cycles could be sent to all of the motors.

FEB 09

Assembled main parts.

FEB 02

Got the 3D printed hull and started to assemble physical parts.

initial assemble

FEB 01

Set up the cross-compile environment for the BeagleBone Black.

JAN 25

The ball tracking algorithm is successfully locating the ball.

ball detection on sample image

JAN 21

Partially completed soldering electrical connections.

JAN 20

Finished mechanical redesign which changed the location of the motors.