An Electronic Oar for Controlling an Electronic Kayak

Let’s say that, hypothetically, you were to build an electronic kayak. That bad boy might — in theory — have an electric motor with a propeller to make the kayak zip through the water. And maybe it could even include some sort of machine learning model, so AI knows when to spin the motor to assist with paddling, kind of like a modern e-bike. That all sounds pretty cool, but how do you control it? Braden Sunwold actually built this hypothetical e-kayak and his solution was to build this electronic oar to control it.

This seems like a really obvious solution in retrospect, which is why it is so clever. Even though the e-kayak’s electric motor can provide propulsion, the user is still going to need a paddle. At the very least, they’ll need that for changing direction and fine adjustments. And it is likely they will do some paddling to supplement the motor, too. Instead of fumbling around with a smartphone or some kind of throttle box when they’re in the water, the user can just press a button on the paddle. More importantly, the e-kayak can pull data from the paddle to help the machine learning model determine how much power to give the motor and when.

It was necessary to create a waterproof enclosure for the electronic components and Sunwold turned to his friend, Jordan Godoy, for help with that. Jordan created a 3D model and then 3D-printed it. The enclosure is cylindrical to allow for the use of a standard o-ring seal. Clamps let it secure to the paddle’s shaft. A button on one end of the enclosure lets the user adjust speed and an Adafruit NeoPixel LED ring shines through the 3D-printed plastic to provide feedback.

Inside that enclosure, there is an Adafruit Feather M0 Basic Proto development board. It collects data from the button and a BNO085 IMU. It then sends the data to the e-kayak’s Raspberry Pi 4 Model B controller via nRF24 radio frequency transceivers. From there, it is up to the e-kayak’s Raspberry Pi to determine what to do with the motor. In automatic mode, it uses the machine learning model to attempt to assist the user’s paddling in a natural way. In manual mode, it simply provides thrust and the user can push the button on the paddle to change the speed.

Sunwold still has some more work to do to refine this concept and build a practical e-kayak, but he’s well on his way and this electronic oar is a big step forward.