Narendalai's Side Projects

Udacity Android Developer Nanodegree Projects

Jan 2018 - Nov 2018 [certificate]

• TrackVan Capstone Project: This app is designed for van owners for tracking its usage within different drivers. For example, in a high school, the principal can be the owner of the van and use this app to track and manage the van's usage. The van can be shared between football team, soccer team, basketball team and international center. List the main features of your app: user authentication with sign in and sign out functinalities; van admin - create, update, delete van; share van unique code to users by email or msg; user - include, disclude van; to use a van, has to include it first; users can do view van status, start using van, return van. Data persistence is accomplished by Firebase user authentication and Firebase realtime database. [github]
• Gradle for Android and Java Final Project: In this project, I created an app with multiple flavors that uses multiple libraries and Google Cloud Endpoints. The finished app will consist of four modules. A Java library that provides jokes, a Google Cloud Endpoints (GCE) project that serves those jokes, an Android Library containing an activity for displaying jokes, and an Android app that fetches jokes from the GCE module and passes them to the Android Library for display. [github]
• XYZ Reader App project: Completed a content reading app per the starter code provided. [github]
• Android Baking App project: Created an app to view video recipes that incorporated media loading, verifying user interfaces with UI tests, integrated third party libraries and provided a complete UX with home screen widget. Following features are accomplished: used MediaPlayer/Exoplayer to display videos; handled error cases in Android. added a widget to the app experience. leveraged a third-party library in the app. used Fragments to create a responsive design that works on phones and tablets. [github]
• Popular Movies Android App: Built an Android app to help users discover popular and highly rated movies on the web. This app utilizes core Android user interface components and fetches movie information using themoviedb.org web API. The app displays a scrolling grid of movie trailers and launches a details screen whenever a particular movie is selected. Allows users to save favorites, play trailers, and read user reviews. In addition, it is optimized for tablet users. [github]
• Sandwich Club Project: In this project, I completed the Sandwich Club app to show the details of each sandwich once it is selected. Built a ScrollView layout, populated its fields by parsing JSON data recieved. Utilized Android image downloading and caching library Picasso. [github]

Udacity Self-Driving Car Nanodegree Projects

Jun 2017 - Apr 2018 [certificate]

• Finding Lane Lines on the Road: Detected highway lane lines on a video stream. Used OpenCV image analysis techniques to identify lines, including Hough Transforms and Canny edge detection. [github]
• Advanced Lane Finding: Built an advanced lane-finding algorithm using distortion correction, image rectification, color transforms, and gradient thresholding. Identified lane curvature and vehicle displacement. Overcame environmental challenges such as shadows and pavement changes. [github]
• Vehicle Detection and Tracking: Created a vehicle detection and tracking pipeline with OpenCV, histogram of oriented gradients (HOG), and support vector machines (SVM). Optimized and evaluated the model on video data from a automotive camera taken during highway driving. [github]
• Traffic Sign Classification: Built and trained a deep neural network to classify traffic signs, using TensorFlow. Experimented with different network architectures. Performed image pre-processing and validation to guard against overfitting. [github]
• Use Deep Learning to Clone Driving Behavior: Built and trained a convolutional neural network for end-to-end driving in a simulator, using TensorFlow and Keras. Used optimization techniques such as regularization and dropout to generalize the network for driving on multiple tracks. [github]
• Extended Kalman Filter Project: Implemented an Extended Kalman Filter algorithm in C++ capable of tracking a bicycle's motion in two dimensions. Managed to process Lidar and Radar generated data. [github]
• Unscented Kalman Filter Project: Implemented an Unscented Kalman Filter algorithm in C++ capable of accurately and performantly tracking a turning object. [github]
• Kidnapped Vehicle Project: Implemented a 2 dimensional particle filter in C++ capable of localizing a vehicle within desired accuracy and time. [github]
• PID Controller Project: Implemented a PID controller in C++ to maneuver a vehicle around a track. [github]
• Model Predictive Control Project: Implemented Model Predictive Control to drive a vehicle around a track even with additional latency between commands. [github]
• Path Planning Project: Built a path planner that navigates a vehicle through traffic on a highway. Used the concepts taught in the module - environmental prediction, behavioral planning, and trajectory generation - to build the planner. [github]
• ROS System Integration Project: Utilized ROS subscribers and publishers, integrated different components of self-driving car. Annotated traffic light images and generated TFRecords file for training the classifier. Fine-tuned a traffic light classifier adapted TensorFlow Object Detection API, which can determine traffic light in real time in Udacity simulator. [github]
• Semantic Segmentation Project: Trained and tuned a Fully Convolutional Network to segment road images using pre-trained VGG model and Kitti Road dataset. [github]

Udacity Deep Learning Nanodegree Projects

Jan 2017 - May 2017 [ github | certificate ]

• Bike Sharing Usage Prediction: Build and train neural networks from scratch to predict the number of bikeshare users on a given day.
• CIFAR-10 Image Classification with CNN: Implemented a convolutional neural network to classify images from the CIFAR-10 dataset.
• TV Scripts Generation by RNN: Trained a recurrent neural network on scripts from The Simpson's to generate new text.
• Language Translation: Trained a sequence to sequence model on a dataset of English and French sentences that can translate new sentences from English to French.
• Face Generation with GAN: Built a deep convolutional generative adversarial network to generate new images of human faces.

Hospital Management System

Aug 2015 - March 2016 [github]

• Research requirements, understand workflows.
• MS SQL Server database design and programming.
• Develop desktop applications with VB.NET in Visual Studio.
• IIS server setup, software implementation.
• Realized streamlined workflow from registration, prescription, medicine to payment and receipt print.