home

Welcome to FWD Skill Zone

Braccio Robotic Arm

TinkerKit Braccio is a fully operational robotic arm that can be controlled by Arduino micro-controller. Braccio robotic arm can be assembled in several ways for multiple tasks. Braccio can also support various objects on the end-effector which makes it versatile in its functions.

Voice Activated Robotic Arm

This project demonstrates an incorporation of an Arduino based voice-activation system to Braccio robotic arm. In this project, we will use the ELECHOUSE Voice Recognition Module to record 7 voice commands and then we will use these voice commands to tell Braccio Robotic Arm to perform few tasks.

Smart Robot Car

In this project, we will use the Elegoo Smart Robot Car. The Smart Car has features like Ultrasonic sensor for collision avoidance, Optical sensors for line following navigation, Bluetooth connection for Smart phone remote control operation, and an Infrared Remote Control.

Short Range Radar

This project demonstrates the design and construction of a Short Range Radar that will measure the distance from the front of the MaxSonar ultrasonic sensor to nearby objects. For this project, we need Ultrasonic sensor, Servo motor, Arduino UNO, and jumper wires.

Breadboard Power Supply

This Breadboard Power Supply will power breadboard projects to either 3.3Volts or 5Volts regulated. For input power, the BBPS uses either a DC barrel jack as a primary input or stripped wires on the secondary input terminals. The BBPS can take 6 to 30VDC power from a DC wall wart and outputs 5V or 3.3V regulated voltage. The user can select the desired voltage using a slide switch.

Dual-Polarity-Variable Power Supply

Almost all electronic devices and instruments are powered from a DC power source. This DC source can be either a battery or a power supply; however, these electronic devices require not only DC voltage sources but also a voltage that is filtered and regulated. There are different types of power supplies, but this project covers the design and construction of a linear power supply only. The goal of this project is to design, build, and simulate a Dual-Polarity-Variable Linear Power Supply with an external 5-Volt USB port, fixed ±12V, and a variable 1.25V to 10V output at a maximum current of 1A. The circuit board is designed in Eagle CAD software.

DC to DC Boost Converter

There are many different switching voltage regulator topologies like buck, boost, buck-boost, and chuck. However, this project covers the design and construction of a boost converter only. This converter is capable of boosting a 12V input voltage to a variable 12 to 24V output voltage (potentiometer controlled). The boost converter steps-up the input voltage by storing the electrical energy in an inductor and then releasing this energy to the load at a higher voltage level.

Deep Learning | Part 1 | Classification

Deep learning is a branch of machine learning that teaches computers to do what comes naturally to humans: learn from experience. Deep learning is especially suited for image recognition, which is important for solving problems such as facial recognition, motion detection, and many advanced driver assistance technologies such as autonomous driving, lane detection, pedestrian detection, and autonomous parking.
In this project, we will use a pre-trained Neural Network (CNN) to classify different objects. We will connect a web camera to our laptop, load a pre-trained CNN, continuously take pictures, resize the pictures and finally classify an object based on the selected CNN, in our case GoogLeNet.

Deep Learning | Part 2 | Transfer Learning

The purpose of a transfer learning is to fine-tune a pre-trained Convolutional Neural Network that will be used to perform classification on a new problem. According Mathworks, "AlexNet has been trained on over a million images and can classify images into 1000 object categories (such as keyboard, coffee mug, pencil, and many animals)." However, if we are looking for a classification accuracy over 99% on given objects, or if our image set is outside of the CNN, then we have to perform transfer learning. Transfer learning is commonly used in deep learning applications that we can take a pretrained network and use it as a starting point to learn new tasks. Fine-tuning a network with transfer learning is usually much faster and easier than training a network with randomly initialized weights from scratch. We can quickly transfer the learned features onto a new task using a smaller number of training images (aka ImageDataStore).

Design, Build, and Test Electronics Projects

Welcome to FWD Skill Zone