Ishank Gulati BTech student at Malaviya National Institute of Technology

Clustering using Genetic Algorithm

Introduction This post is in continuation to one of my earlier post Image Segmentation using K-Means Clustering. This post intends to demonstrate an algorithm which tries to resolve the drawbacks of K-Means Clustering by inducing parallelism in computation. For this project I have referred Genetic Algorithm-based Clustering technique research paper by Ujjwal Maulik and Sanghamitra Bandyopadhyay. Frequently used Notations Notation Description n_samples Number of samples in dataset n_features Dimension of a sample n_clusters Number of clusters into which image is to be segmented Genetic Algorithm Genetic Algorithms are randomized search and optimization techniques guided by the principles of evoluion and... Read More

Image Segmentation using K-Means Clustering

Introduction Images are considered as one of the most important medium of conveying information. Understanding images and extracting the information from them such that information can be used for other tasks is an important aspect of Machine Learning. There are enormous applications of the same such as use of images in navigation of robots, in self driving cars, medical diagnosis etc. Image Segmentation is used to seperate different objects in the image such that the segmented objects can be used in further computation and analysis. There are several algorithms available for this purpose and in this post we will use... Read More

Machine Learning based Obstacle Avoider

Introduction This project aims at developing an Artificial Neural Network controlled autonomous robot which can avoid any kind of obstacles in its path. I have used Firebird V robot developed by Nexus Robotics for demonstration. A webcam is mounted on top of robot which is further connected to a laptop. Image captured by webcam will be processed on laptop using a python script. After processing, these images will be fed to the trained Neural network which will control the navigation of robot. The output of ANN is wirelessly transmitted to robot using XBee module. For this project I have referred... Read More

Caretaker Robot

Introduction This project is a part of e-yantra+ Robotics Competition (eYRC+), a unique annual competition for undergraduate students in science and engineering colleges organised by IIT Bombay. We were given a robotic kit, consisting of Firebird V robot and other components required to complete the task. e-Yantra Robotics Competition Plus has designed a theme to bring our awareness to the issue of catering to the needs of patients, for example, automating the task of serving them, even in the absence of human supervision. In this theme, the arena is an abstraction of a floor in a hospital, which includes a... Read More

Pathfinding using A* Algorithm

Introduction We are given a set of test images, each containing A grid of 100 squares of size 40x40 pixels. Start marked with a orange square. End marked with a pink square. Obstacles marked with black squares. The squares are identified by the coordinate (x,y) where x is the column and y is the row to which the square belongs, as shown in the picture. Assuming a robot moves from Start to End by moving either horizontally or vertically, the objective is to find the shortest path from Start to End. The length of the path is determined by the... Read More