Tag Archives: self-learning helicopter

Self-learning ‘copter navigates with an ATmega644 MCU

Akshay Dhawan and Sergio Biagioni of Cornell University have designed a self-learning (RC) helicopter powered by an advanced machine learning algorithm paired with Atmel’s ATmega644 microcontroller (MCU).

Aside from Atmel’s ATmega644 MCU, key project components include:

  • Syma S107 Micro Helicopter
  • Custom PC Board (for MCU)
  • RS232 UART connector
  • Max233CP
  • Power Supply
  • Infrared Emitter 365-1056-ND
  • Infrared Receiver 160-1030-ND
  • Wooden platform
  • Balsa wood 24 inch dowel
  • White board (holds phototransistor circuit)

As HackADay’s Will Sweatman reports, the ‘copter is attached to a boom which restricts its movement down to one degree of motion. Meaning, the helicopter can only move up from the ground, rather than side to side or front to back.

“The goal is for the helicopter to teach itself how to get to a specific height in the quickest amount of time. A handful of IR sensors are used to tell the ATmega644 how high the helicopter is,” writes Sweatman.

“The genius of this though, is in the firmware. Akshay and [Sergio] are using an evolutionary algorithm adopted from Floreano et al, a noted author on biological inspired artificial intelligences.”

Essentially, the ‘copter creates random “runs” and then check the data. The runs that are closer to the goal are refined, while the others are eliminated in a process that emulates evolution via natural selection. In short, the project’s goal is for the ‘copter to start at Point A, go to Point C and hover. The allotted time is 10 seconds per run, with the helicopter expected to teach itself the routine as quickly as possible.

“A neural network is used to determine at what level the throttle should be at to achieve the highest Fitness Value. This network is a part of the Evolutionary Algorithm that runs in the firmware. Basically, it starts off with random values that generate random levels of throttle,” Sweatman explains.

“The values that achieve the highest Fitness Value get ‘mutated’, while the others are discarded. The mutations in the values are done at random and the process repeats. In the end, the firmware learns the best throttle levels to achieve the goal of being at Point C for the longest time in the allotted 10 seconds.”

Interested in learning more about the self-learning ‘copter? You can check out the project’s official Cornell page here.