Virtual Expo 2024 | Virtual Expo

E-Commerce App

The E-Commerce Application project endeavors to craft a sophisticated and user-centric mobile application tailored for online shopping, harnessing the powerful combination of Flutter and Firebase technologies. By capitalizing on Flutter's versatility in developing cross-platform applications, and Firebase's robust features for cloud storage and authentication, the project aims to deliver an immersive and intuitive shopping experience to users. Through meticulous design and implementation, the application will seamlessly integrate essential functionalities such as user authentication, product catalog browsing, and cart management thereby ensuring a seamless and gratifying shopping journey for every user

Envision

CompSoc

Analysis of Swept and Unswept Wings in Subsonic and Supersonic Flows

As aircraft have progressed to higher speeds, nearing the speed of sound, the formation of shockwaves on wing surfaces has led to increased drag, necessitating additional thrust to maintain velocity. To enable supersonic flight, wing designs have been angled in a swept configuration, which results in a slower local airflow and reduced pressure compared to unswept wings. This strategic sweeping allows aircraft to achieve higher speeds before encountering shockwave formation, enabling flight at velocities exceeding Mach 1. Wing sweep, a technique implemented to mitigate the adverse effects of local supersonic flows during high-velocity flight, serves to alleviate the detrimental impact of shockwaves caused by said flows on an aircraft's aerodynamic performance. By sweeping the wing backward, the incoming airflow can be delineated into two distinct components: one flowing over the airfoil and another traversing along the wing's leading-edge perpendicular to the airfoil. While the former contributes to both lift and drag generation, the latter has negligible aerodynamic influence. In subsonic regimes for swept wings, the utilization of only a fraction of the incoming airflow results in diminished lift production compared to unswept wings. However, as velocities approach and surpass the speed of sound, the reduced airspeed over the airfoil in a swept wing confers an advantage by impeding the rapid escalation of drag experienced by unswept configurations. Theoretical postulations of these principles are to be empirically validated throughout the course of this project.

Envision

Piston

Analysis and Enhancement of Properties of Wing Through Tubercles

This project explores the integration of tubercles, inspired by the unique features of humpback whale fins, onto the wings of aircraft to enhance aerodynamic performance. Tubercles are small, rounded protrusions found along the leading edge of humpback whale fins, which are believed to play a significant role in improving lift and reducing drag during locomotion. In this study, we investigate the impact of incorporating tubercles onto the wings of an aircraft through computational simulations and experimental testing. Our findings suggest that the addition of tubercles can potentially enhance the aerodynamic efficiency of aircraft wings, leading to improved lift-to-drag ratios and overall performance. These results offer valuable insights into the application of biomimicry in aerospace engineering and may contribute to the development of more efficient and sustainable aircraft designs. If successful, the integration of tubercles into aircraft wing designs could revolutionize the way we approach aerodynamic optimization, potentially paving the way for more fuel-efficient, environmentally sustainable aircraft. Furthermore, this study underscores the importance of biomimicry as a source of inspiration for innovative engineering solutions, bridging the gap between nature and technology for the benefit of human progress

Envision

Piston

Stegosaurus - Hiding in Plain Sight

Stegosaurus is a command-line tool for Linux-based operating systems that allows for the encryption and concealment of messages or files within other files, such as images or audio files, utilizing C++. This project was a dive into steganography, which involves the technique of concealing messages in such a way that the existence of the message remains undetectable; encryption, which is for converting the data that is being hidden into a form that is not directly understandable, and data compression.

Envision

CompSoc

Introduction to Web Development with Django

This project aims as an introduction to web development using Django. Once learning the necessary skills required, work is being done on IEEE NITK's website, which is a Django powered application.

Envision

CompSoc

C2 - GottaQuizEmAll

GottaQuizEmAll is a specialized quiz website focused on Pokémon, providing users with a robust platform for in-depth exploration of Pokémon data. The website facilitates access to detailed Pokedex statistics, enabling users to delve into the intricate characteristics of various Pokémon species. Authenticated users can create and engage in customized quizzes. The platform features leaderboards, allowing users to compete and showcase their expertise. With an intuitive interface, GottaQuizEmAll aims to deliver a technically sophisticated experience for users keen on enhancing their understanding of Pokémon.

Envision

CompSoc

Motion Alarm Module using 555 timer

The Motion Detection Alarm System is a solution designed to enhance security and awareness in various environments through the integration of motion sensing with visual and auditory feedback mechanisms. Utilizing motion sensors to detect movement within its designated range, the module triggers a response mechanism that delivers both visual feedback through LED indicators, and auditory feedback through a buzzer, to promptly notify users of detected motion.

Envision

Diode

A Deep Learning Based Framework For Robust Traffic Sign Detection Under Challenging Weather

"Addressing the crucial need for robust traffic sign detection in adverse weather conditions, this project presents a novel Convolutional Neural Network (CNN) architecture. Leveraging deep learning techniques, transfer learning, and data augmentation, the proposed model surpasses existing methods by enhancing performance and generalization across diverse environmental scenarios. Implemented using Python, TensorFlow, and Keras, with GPU acceleration for efficiency, the model demonstrates state-of-the-art accuracy on real-world datasets like the German Traffic Sign Detection Benchmark (GTSDB). Thorough evaluation metrics ensure credibility, while transparent references enable reproducibility, highlighting the project's contribution to advancing autonomous vehicle technology."

Envision

Diode

Piston

Design and Implementation of Obstacle Avoiding Line Follower Robot

This Project is about the functioning of a Line Follower Robot which follows a line until it exists using infra-red sensors. The robot works using PID control algorithm and is made using Arduino UNO

Envision

Diode

Bird Call Classification

This project aimed to develop a deep learning model for classifying different species of birds based on audio recordings of their vocalizations. The dataset was obtained from Kaggle and consists of 5 different species of audio samples each representing a single class. Python’s librosa library was used to process the audio samples and extract log mel-spectrogram image representations from the audio files. These 2D spectrograms, which encode the time-frequency patterns of the bird vocalizations, were then normalized. The normalized spectrogram images served as input to a convolutional neural network (CNN) model built using the TensorFlow framework. After training for 15 epochs, the test accuracy was about 83.33%; the trained CNN demonstrates the feasibility of using deep learning on audio spectrograms for acoustic bird species classification. Potential improvements could involve data augmentation, regularization, and ensemble methods to better generalize the model's performance across diverse recording conditions.

Envision

Diode

Modified 555 Timer IC Using Only Two Comparators

This project proposes a modified 555 timer IC with only two comparators by eliminating the discharge transistor and SR flip-flop. A periodic square waveform can be generated by utilizing the interconnection of two comparators to provide 'Set' and 'Reset' conditions to each other. The simulation and experimental results confirm that the clock frequency generated by the modified 555 timer circuit is significantly higher and duty cycle adjustment can be made more precise and flexible than the commercially available 555 timer circuit. A frequency in the band of 1MHz to 89MHz can be generated by replacing the resistors, capacitor, and comparator specification in all possible configurations. As a result, a lower cost and higher efficiency on-chip reference clock generator can be realized to fulfill versatile future application needs.

Envision

Diode

Vision Kinect

Tetris is a classic puzzle video game that was created by Alexey Pajitnov, a Russian Computer Scientist, in 1984. In Tetris, geometric shapes called Tetriminos fall from the top of a rectangular playfield. The player's objective is to manipulate these falling Tetriminos by rotating and moving them horizontally as they descend, with the goal of 'clearing' horizontal lines. When a line has no gaps, it disappears and the blocks above it cascade downwards, freeing up space for more tetrominoes. Vision Tetris is a modern-day spin of this classic, by making the game hands-free! By taking inspiration from gaming console motion sensing like that of Xbox Kinect, we aim to enable players to play Tetris via the movement of their hands! We achieve this by leveraging modern computer vision object detection techniques, to detect different hand gestures representing different in-game actions. We wrap this system with a web application, to make it easily usable for anyone.

Envision

CompSoc

One Shot Face-Recognition using Siamese Network

Learning effective features for machine learning tasks can be computationally demanding, particularly when dealing with limited data scenarios. Performing accurate predictions based on a single example per class is a challenging scenario for one-shot learning. This paper explores the use of Siamese neural networks for addressing one-shot image recognition tasks. Siamese networks are structured to learn discriminative features by comparing pairs of input images, allowing for the assessment of similarity and dissimilarity between them. Utilizing a convolutional architecture, our approach achieves state-of-the-art performance in one-shot image recognition, surpassing other deep learning models. This re- search contributes to advancing the field of one-shot learning methodologies and underscores the importance of leveraging Siamese networks for feature extrac- tion and similarity assessment in the context of image recognition with minimal training data.

Envision

Diode

Velocity Profile Simulation in MATLAB

This project explores the visualization and analysis of fluid flow dynamics within pipe systems using MATLAB simulations and plots. By examining velocity profiles and pressure distributions, the study seeks to describe the intricate characteristics of fluid flow and evaluate performance metrics such as Reynold's number and flow rate by understanding fluid dynamics principles. The objective is to visualize and decipher fluid flow behaviour's underlying mechanisms. Ultimately, this aspires to equip engineers with invaluable insights into optimizing pipe system designs and enhancing operational efficiencies across diverse engineering applications.

Envision

Piston

Numerical Investigation of Optimal Fin Pitch in Motorcycle Engines.

Maintenance of temperature conditions is essential for internal combustion engines. Hence the system must be so designed to achieve the same. In the current work, specifically the effect of fin pitch has been investigated using Computational Fluid Dynamics (CFD). Various fin pitches are investigated for different wind speeds and based on the obtained data, the optimal fin for the chosen geometry is reported.

Envision

Piston

Design and Implementation of a Smart School Bell System

This project presents the development of a Smart School Bell System using Arduino microcontroller technology. The system automates the process of bell ringing in educational institutions by allowing users to input class schedules via a keypad interface. It utilizes an Arduino Uno microcontroller, a liquid crystal display (LCD), a keypad, and a buzzer to manage class timings and generate audible alerts. The system enhances operational efficiency and eliminates the need for manual bell ringing, offering a convenient solution for educational institutions.

Envision

Diode

Image Quality Manipulation Using Autoencoders

In our project, we use the capabilities of deep autoencoders to enhance image quality. By using super-resolution and noise removal techniques, our project aims to tackle two of the most important problems with Image Quality. Deep autoencoders unveil intricate details within images, making them significant tools in applications requiring image quality preservation and restoration such as medical diagnostics, surveillance, and satellite imagery. This project is a right step into the realm of image processing, promising impactful solutions for real-world applications.

Envision

CompSoc

Advanced Recognition of License Plates (ARLP)

Advanced Recognition of License Plates (ARLP) is an Envision project (I03) under IEEE NITK (2024) that focuses on developing a robust Vehicle Number Plate Recognition (NPR) system for various automobiles. Utilizing a custom CNN model inspired by various architectures worldwide and incorporating transfer learning with Inception ResNet V2, we evaluated and compared the effectiveness of both approaches. Additionally, we employed TesseractOCR and EasyOCR for text extraction from number plates. The objective is to create a specialized NPR system that efficiently and accurately identifies Indian license plates for applications, including automated toll collection, traffic monitoring, and law enforcement.

Envision

Piston

D03: Simulation of a Solid-State Guitar Amplifier

In this Project, we aim to simulate a solid-state Electric Guitar Amplifier on LTSpice, consisting of 3 main circuits: Tone, Volume and Distortion. The three circuits are assembled at the end to make up a functional Electric Guitar Amplifier. The Tone stage will help adjust the tone by varying a Capacitor to pick up needed frequencies. The Volume Stage helps increase the signal volume by amplification, and the Distortion Stage is implemented by using a positive feedback Op-Amp in the circuit. Finally, an Active Noise-Cancelling (ANC) Circuit has been added to the amplifier, which has been implemented using a 3-stage amplification circuit involving a pre-amplifier, a delay or all-pass filter, and a summing amplifier. The main objective of this project is to develop knowledge and expertise of LTSpice in simulating complex analog RC circuits and gain appreciable insights into implementing different types of operational amplifiers in various circuits.

Envision

Diode

Bipedal Robot Simulation using MATLAB

In this project, we aim to simulate a bipedal (two-legged) robot’s walking mechanism using MATLAB, Simulink, and Simscape Multibody. MATLAB is a high-level programming language and interactive environment widely used for numerical computing, data analysis, and visualization. It provides a powerful platform for engineers and scientists to develop algorithms, model systems, and conduct computational experiments. Simulink is a graphical programming environment in MATLAB that allows users to model, simulate, and analyze multi-domain dynamical systems using a block diagram representation.

Envision

Diode

Piston

Forecasting of Air Quality using Machine Learning

This project utilizes various Machine Learning (ML) models in tandem with various python libraries such as NumPy, Pandas, Matplotlib, and others, to forecast the Air Quality Index (AQI) of a selected location. Our aim here is to predict the subindices of various components that influence air quality by leveraging ML models for the same and thus accurately predict the AQI. This enables us to estimate the amount of pollution in air and study its effect on various buildings, materials, and the environment as a whole. The data used in project has been sourced from the Central Pollution Control Board (CPCB) website. Selenium, a tool for Web Automation, was helpful in scouring the web and extracting hourly data containing concentrations of various gases, humidity levels, and other atmospheric parameters at a specific place. Collected data is then analyzed and used to train models, eventually allowing us to forecast our desired variable.

Envision

Piston

Etherballot

"EtherBallot" represents a groundbreaking initiative in the realm of voting systems, integrating blockchain technology to revolutionize the way elections are conducted. This project delves into the innovative use of decentralized ledgers to ensure the security and transparency of voting processes. Through the exploration of blockchain's inherent features, EtherBallot strives to create a voting system that is resistant to manipulation and fraud, fostering trust and credibility in electoral outcomes. By embracing cryptographic principles, EtherBallot aims to redefine the landscape of democratic practices, promoting inclusivity and accessibility in the electoral sphere.

Envision

CompSoc

Filmizz : The Cinematic Magic

FilmWizz is your ultimate destination for all things cinema. With a simple search, users can discover a treasure trove of movies, sorted by genre for easy navigation. From timeless classics to the latest blockbusters, FilmWizz has it all. Dive deeper into your favorite films with detailed information such as box office earnings, IMDb ratings, cast lists, and more. Whether you're a casual moviegoer or a die-hard cinephile, FilmWizz is your go-to source for movie magic.

Envision

CompSoc

GPS Tracker Using Arduino

In this project, we aim to create a simple method of keeping track of the target's location using a microcontroller - specifically the Arduino UNO and a GPS module. The data collected on by the module is uploaded on a cloud service called ThingSpeak which is commonly used for IoT applications. ThingSpeak also has supports RESTful API which we have to used to create a simple android app to download the latitude and longitude of the tracker at the press of a button.

Envision

Diode

Modal Analysis of Aircraft Wing

This project seeks to make a 3d model of an aircraft wing and perform modal analysis on it. For designing of the wing we've used Fusion and for Modal Analysis we've used Ansys.

Envision

Piston

GameDev and AI Bot Development in Godot

This project aims to design and develop a 2D tank battle game using the Godot game engine and implement a Reinforcement Learning (RL) agent to autonomously play the game. The game will feature a map where players control tanks with the objective of destroying enemy flags while navigating obstacles. The development process will involve creating the game environment, including tank movement mechanics, weapon systems, map creation and graphical assets. The game will allow players to compete against AI-controlled opponents. The RL bot will be trained using NEAT algorithm to learn optimal strategies for navigating the battlefield, aiming, and shooting at enemy tanks while avoiding incoming fire. The bot will undergo extensive training in simulated environments to improve its performance and adaptability to different game scenarios.

Envision

CompSoc

IoT-Based Weather Monitoring System Using Arduino

This report presents an exploration into the integration of Arduino microcontrollers within IoT-based weather monitoring systems, offering a comprehensive overview of its design, implementation, and potential applications. Beginning with an overview of the significance of weather monitoring in diverse sectors, the report underscores the need for innovative solutions. It then introduces the Arduino platform as a versatile tool for constructing IoT devices, providing insights into its features, capabilities, and compatibility with various sensors. Subsequently, the report delves into the design and architecture of an IoT-based weather monitoring system using Arduino. It discusses the selection and integration of sensors such as temperature, humidity, and atmospheric pressure sensors, along with the utilization of Arduino boards for data processing and communication. The report presents a practical implementation of the IoT-based weather monitoring system using Arduino. It details the hardware setup, software configuration, and deployment considerations, providing a step-by-step guide for replication and experimentation. In conclusion, this report emphasizes the role of Arduino in democratizing IoT development and fostering innovation in weather monitoring. It underscores the versatility and adaptability of Arduino-based solutions in addressing diverse weather-related challenges and driving progress towards a more resilient and sustainable future.

Envision

Diode

Serpentine Saga

Meet Link: https://meet.google.com/igi-fext-uyq SERPENTINE SAGA(SLITHER.IO) is a multiplayer online video game available for Ios,android and web browsers, developed by Steve Howse. Players control an avatar resembling a snake, which consumes multi-colored pellets, both from other players and ones that naturally spawn on the map in the game, to grow in size. We aim to implement a clone of this game in order to learn about Game Development, Python programming and Networks programming.

Envision

CompSoc

Electronic Thermostat

This report presents the design, construction, and implementation of an electronic thermostat utilizing NTC (Negative Temperature Coefficient) thermistors, transistors, and relays.

Envision

Diode

Gesture Controlled Virtual Mouse Using ESP32CAM

Controlling the mouse on the PC screen by processing external camera feed from a microcontroller using OpenCV libraries

Envision

Diode

RecipeRaven

Develop a robust recipe management system using React, tailored to efficiently organize and share culinary creations

Envision

CompSoc

Multi-Tool Rotatable Arm

This project focuses on designing and implementing a versatile 360-degree robotic arm, beginning with simulation-based development before hardware construction. Through simulation tools, the arm's design and control mechanisms are optimized for adaptability across various scenarios. Academic goals include honing skills in Arduino programming, CAD modeling, and simulation techniques. Hands-on engagement with Arduino microcontrollers and CAD software facilitates practical experience in coding and mechanical design principles. Additionally, the project provides a platform for learning simulation and testing methodologies relevant to electronic circuits, fostering interdisciplinary collaboration and skill development in robotics and mechatronics.

Envision

Diode

Piston

Image Inpainting using GAN

Image inpainting, the process of reconstructing missing or damaged parts of an image, has witnessed significant progress with the advent of Generative Adversarial Networks (GANs). This project explores recent developments in image inpainting techniques, particularly focusing on GAN-based approaches, applied to the widely used CelebA dataset. The CelebA dataset consists of celebrity images with diverse poses, expressions, and backgrounds, making it a challenging testbed for inpainting algorithms. We delve into various GAN architectures and training strategies employed to achieve compelling inpainting results on CelebA images.

Envision

CompSoc

Design and Analysis Of Regenerative Braking System using Simulink

This report presents a comprehensive analysis of a regenerative braking system designed for an electric vehicle using MATLAB/Simulink. The general concept of the specific mechanism and advantages and limitations of RBS have been evaluated in the study from existing literature sources. The study focuses on evaluating various performance aspects, including state of charge, field current, armature current, and electrical torque. The system comprises essential components such as a battery, DC machine, and four gates/switches. Through simulation and analysis, the effectiveness of regenerative braking in recovering kinetic energy and its impact on the overall performance of the electric vehicle are investigated.

Envision

Piston

API Tester

API Tester Mentors Aahil Rafiq : 7975657621 , aahil.rafiq136@gmail.com Rohan Chandra : 8639884378 , rohanchandra0405@gmail.com INTRODUCTION We're creating an API Tester from scratch. Beginning with HTML, CSS, and JavaScript, we'll explore internet fundamentals, REST APIs, and HTTP requests. The project involves setting up a Node.js server, making API requests, and transitioning to React for improved UI. In the final phase, we'll focus on basic authentication to enhance application security. OBJECTIVE Master basic frontend skills: HTML, CSS, and JavaScript. Build a basic REST API using Node.js and Express. Understand APIs, how to make API calls, and the key elements of a request. Learn to troubleshoot CORS errors. Explore React fundamentals, including how React works, basic React hooks, component splitting, and using UI libraries. Bonus: Set up basic authentication from scratch for added learning and security. METHOD Start with the basics: learn HTML, CSS, and JavaScript. Understand how the internet works, know the difference between frontend and backend, and learn about REST APIs and HTTP requests. Learn to set up a Node.js server using Express.js. Practice making simple requests from the frontend to the backend. Dig deeper into API requests, their contents, and what they mean. Practice making requests using tools like the JS Fetch API or Axios. Learn React and switch from a basic HTML frontend to a more dynamic React frontend. Finally, explore basic authentication. Add this security feature to our application for better protection. Prerequisites Good to have : basic html , css and javascript Interest in web development REFERENCES How do Websites Actually Work? HTML Tutorial for Beginners: HTML Crash Course CSS Crash Course For Absolute Beginners JavaScript Crash Course For Beginners Docs : html , css , JS Frontend web development - a complete overview Backend web development - a complete overview RESTful APIs in 100 Seconds // Build an API from Scratch with Node.js Express How to build a REST API with Node js & Express n0rdy - Understanding CORS Http Request and Response. An HTTP (Hypertext Transfer Protocol)… | by Raza | Nov, 2023 | Medium axios docs Do You Know Enough JavaScript To Learn React official website to learn react Google Meet Link: https://meet.google.com/pxw-qqmx-nim

Envision

CompSoc

OPTICharge

One of the most crucial challenges faced by the Li-ion battery community concerns the search for the minimum time charging without damaging the cells. This goal can be achieved by solving a large-scale constrained optimal control problem which relies on accurate electrochemical models. However, these models are limited by their high computational cost as well as identifiability and observability issues. As an alternative, simple output-feedback algorithms can be employed, but their performance strictly depends on trial and error tuning. Moreover, particular techniques have to be adopted to handle safety constraints. With the aim of overcoming these limitations, we propose an optimal-charging procedure based on deep reinforcement learning. In particular, we focus on a policy gradient method to cope with continuous sets of states and actions. First, we assume full state measurements from the Doyle-Fuller Newman (DFN) model, which is projected to a lower-dimensional feature space via Principal Component Analysis(PCA).

Year Long Project

Diode

Piston

Design and optimization of Swiss-roll type liquid cooling micro-channel for thermal management of power battery

When an electric vehicle ascends a slope while accelerating, it often demands high-rate discharge from its battery module to meet driving needs. Failure to promptly dissipate heat can diminish the battery module's lifespan and efficacy due to elevated temperatures and uneven thermal distribution, potentially leading to thermal runaway. The current thermal management system in battery packs faces challenges in maintaining optimal operating temperatures, resulting in reduced efficiency and safety. Integrating Swiss roll liquid cooling microchannel aims to address these issues by enhancing heat transfer and lowering overall battery pack temperatures. Hence, mitigating the maximum temperature and temperature variance of the battery module is crucial. This project presents a thermal management system for a Swiss roll-type battery, wherein we maintain the coolant inlet flow rate, coolant type, and inlet temperature constant while altering the design of the micro-channel by altering the flow type to enhance heat dissipation efficiency.

Year Long Project

Piston

Building Information Modelling

We will use BIM for designing and increasing the efficiency of the project by improving collaboration between different team members throughout the project life cycle and producing a central model. It is done by modelling building elements, implementing interiors and circulation, views and visualization. Construction modelling is done to enable teams to simulate potential designs, estimate unnecessary costs, and identify flaws between the architectural models. Revit is a BIM software for project documentation. It performs complete projects in engineering and architecture. Unlike AutoCAD, Revit inserts real elements like walls, doors, stairs, columns, floor, ceiling etc., The Project demonstrated that utilizing Building Information Modelling (BIM) with Revit for residential building floor plans offers substantial benefits. BIM improved efficiency by streamlining tasks and ensuring accuracy across the project. The 3D visualization capabilities of BIM aided in identifying and resolving potential issues early on. Overall, BIM proved to be a valuable tool for creating and managing construction project information effectively.

Year Long Project

Piston

Design And Analysis of Bridge Using STAAD Pro

We've designed a Bridge for class A Loading (As per IRC 6), using STAAD Pro.

Year Long Project

Piston

ARTBot

ARTbot is an innovative swarm robotics project aimed at merging art with technology. This project introduces a novel concept where users interact with a canvas window to design shapes and patterns. The swarm bots trace the path fed in the canvas. This project was created using ROS2 and OpenCV. ARTbot showcases the seamless integration of robotics and art and demonstrates the potential of swarm robotics in interactive and collaborative environments.

Design and Analysis of multi-storey buildings using STAAD Pro

In order to compete in the ever-growing competent market, it is very important for a structural engineer to save time. This attempt is made to analyze and design a multistoried building by using a software package STAAD Pro. For analyzing a multi storied building and some other structures one has to consider all the possible loadings and see that the structure is safe against all possible loading conditions.

Year Long Project

Piston

Phase Field Modelling

Mesoscale modelling of the microstructure evolution in materials, during solid-solid phase transformation. Index Terms — Microstructure, Python

Year Long Project

Piston

Path Planning of a Robot using Deep Reinforcement Learning

This project aims to develop Deep Reinforcement Learning algorithm for mobile robot navigation in ROS2. Robots learn to travel through a simulated environment and reach a random goal point while avoiding obstacles by using TD3 neural network. Obstacles are detected by LIDAR (Light Detection and Ranging) sensor and a goal is given to the robot in polar coordinates. We are trying to "find the optimal sequence of actions that lead the robot to a given goal”.

Year Long Project

Diode

Piston