Загрузка страницы

Making a "Smart" Self-Balancing Robot | Training a Neural Network to Replicate a PID Controller

First half of the video outlines the design and fabrication of the robot, and the second half (starts around 3:15) outlines the process of designing and training the neural network.

Why did I do this? It was a project for a research group developing a circuit component that can be used to replicate neural networks in hardware. This self-balancing robot is one project they hope to use these components with. Information on this group and their projects can be found here: https://adam.seas.gwu.edu/

The code for the project can be found here: https://github.com/SchwartzCode/Self_Balancing_Robot

Estimating angle from IMU data: http://philsal.co.uk/projects/imu-attitude-estimation
Tuning PID gains: https://www.youtube.com/watch?v=cjSw7sc2JKk&ab_channel=CircuitDigest
https://www.youtube.com/watch?v=uyHdyF0_BFo&ab_channel=ArcRobotics
PARTS:
2x 6V motor (https://www.dfrobot.com/product-1457.html?gclid=CjwKCAjw8J32BRBCEiwApQEKgW0RJa0U7znFqiDzMztA8LCHXYtYcGpPhKxFxsye__e1dTXuPue8QhoCRHYQAvD_BwE)
2x wheels (https://www.sparkfun.com/products/13259)
1x Arduino Uno (https://store.arduino.cc/usa/arduino-uno-rev3)
1x Arduino Motor Shield rev 3 (https://store.arduino.cc/usa/arduino-motor-shield-rev3)
1x Lead-Acid Battery (https://www.amazon.com/ExpertPower-EXP1270-Rechargeable-Lead-Battery/dp/B003S1RQ2S/)
1x IMU Chip (https://www.adafruit.com/product/3886)
IMU Library for Adafruit MPU6050: https://github.com/adafruit/Adafruit_MPU6050
Arduino MatrixMath Library: https://www.arduino.cc/reference/en/libraries/matrixmath/

Tags: Arduino, motors, project, beginner, intermediate, self-balancing robot, balancing, pendulum, inverted, Putty, Python, Pytorch, Numpy, MatrixMath, PID, controller, mechatronics

Видео Making a "Smart" Self-Balancing Robot | Training a Neural Network to Replicate a PID Controller канала Jank Station
Показать
Комментарии отсутствуют
Введите заголовок:

Введите адрес ссылки:

Введите адрес видео с YouTube:

Зарегистрируйтесь или войдите с
Информация о видео
18 января 2021 г. 5:41:07
00:06:32
Яндекс.Метрика