• Present May-2020

    Research Assistant

    University of Nebraska-Lincoln, Department of Computer Science & Engineering

  • May-2020 Aug-2019

    Teaching Assistant

    University of Nebraska-Lincoln, Department of Computer Science & Engineering

  • Aug-2019 Jan-2017

    Research Assistant

    Mississippi State University, Center for Advanced Vehicular Systems

  • Jul-2016 Feb-2016



  • Jan-2016 Dec-2015

    Scientific Trainee

    Inter University Center for Astronomy and Astrophysics


  • Ph.D. In Progress

    PhD in Computer Science

    University of Nebraska-Lincoln

  • MS 2019

    MS in Computer Science

    Mississippi State University

  • B.Tech2015

    B.Tech in Computer Science & Engineering

    Rajiv Gandhi University of Knowledge Technologies

  • PUC2011


    Rajiv Gandhi University of Knowledge Technologies

Man is made by his belief. As he belives, so he is

- The Mahabaratha


I am interested in all streams of Artificial Intelligence such as Robotics, Computer Vision, and Machine Learning. I also passionate about other fields such as Augmented Reality and Virutal Reality. I have involved with several projects which involes with the domains mentioned above. You can see about those projects in my projects section. By the way, I am planning to join in Doctoral program in Fall - 2019. By then I will get more insight into my major researh area.


  • Artificial Intelligence
  • Robotics
  • Computer Vision
  • Machine Learning
  • Augmented Reality
  • Virutal Reality


Ms. Nagarjuna Devi


Ms. Sowmini Devi


Dr. Edward Swan


Dr. Song Zhang


Dr. Yang Zhao


Dr. Christopher Archibald


"A good teacher can inspire hope, ignite the imagination, and install a love of learning"

I am fortunate to have a great teachers in my life. Everyone inspired me a lot, taught me a lot. Whenver I think about my mentors I always remember the quote said by Jame Levine i.e. "I was lucky that I met the right mentors and teachers at the right moment".

Filter by type:

Sort by year:

Automatic Identification of Broiler Mortality using Image Processing Technology

Veera V. R. M. K. R. Muvva, Yang Zhao, Pratik Parajuli, Song Zhang, Tom Tabler, Josehp Purswell
Conference Papers 10th International Livestock Environment Symposium (ILES X), (p. 1), ASABE, 2018


Identifying dead birds is a time and labor consuming task in commercial broiler production. Automatic mortality identification not only helps to save the time and labor, but also offers a critical procedure/component for autonomous mortality removal systems. The objectives of this study were to 1) investigate the accuracy of automatically identifying dead broilers at two stocking densities through processing thermal and visible images, and 2) delineate the dynamic body surface temperature drops after euthanasia. The tests were conducted on a weekly basis over two 9-week production cycles in a commercial broiler house. A 0.8mx0.6m floor area was fenced using chicken wires to accommodate experimental broilers. A dual-function camera was installed above the fenced area and simultaneously took thermal and visible videos of the broilers for 20 min at each stocking density. An algorithm was developed to extract pixels of live broilers in thermal images and pixels of all (live and dead) broilers in visible images. The algorithm further detected pixels of dead birds by subtracting the two processed thermal and visible images taken at the same time, and reported the coordinates of the dead broilers. The results show that the accuracy of mortality identification was 90.7% for the regular stocking density and 95.0% for the low stocking density, respectively, for 5-week old or younger broilers. The accuracy decreased for older broilers due to less body-background temperature gradients and more body interactions among birds. The body surface temperatures dropped more slowly for older broilers than younger ones. Body surface temperature requires approximately 1.7 hour for 1-week old broiler to reach 1°C above the background level, while over 6 hours for 4-week and 7-week old broilers. In conclusion, the system and algorithm developed in this study successfully identified broiler mortalities at promising accuracies for younger birds ( less than 5-week old), while requires improvement for older ones.

Towards Training an Agent in Augmented Reality World with Reinforcement Learning

Veera Venkata Ram Murali Krishna Rao Muvva, Naresh Adhikari, Amrita D. Ghimire
Conference Papers InControl, Automation and Systems (ICCAS), 2017 17th International Conference on 2017 Oct 18 (pp. 1884-1888). IEEE.


Reinforcement learning (RL) helps an agent to learn an optimal path within a specific environment while maximizing its performance. Reinforcement learning (RL) plays a crucial role on training an agent to accomplish a specific job in an environment. To train an agent an optimal policy, the robot must go through intensive training which is not cost-effective in the real-world. A cost-effective solution is required for training an agent by using a virtual environment so that the agent learns an optimal policy, which can be used in virtual as well as real environment for reaching the goal state. In this paper, a new method is purposed to train a physical robot to evade mix of physical and virtual obstacles to reach a desired goal state using optimal policy obtained by training the robot in an augmented reality (AR) world with one of the active reinforcement learning (RL) techniques, known as Q-learning.

A Collaborative Filtering Recommender System with Randomized Learning Rate and Regularized Parameter

V. V. R. Murali Krishna Rao Muvva
Conference Papers InCurrent Trends in Advanced Computing (ICCTAC), IEEE International Conference on 2016 Mar 10 (pp. 1-5). IEEE.


Recommender systems with the approach of collaborative filtering by using the algorithms of machine learning gives better optimized results. But selecting the appropriate learning rate and regularized parameter is not an easy task. RMSE changes from one set of these values to others. The best set of these parameters has to be selected so that the RMSE must be optimized. In this paper we proposed a method to resolve this problem. Our proposed system selects appropriate learning rate and regularized parameter for given data.

A Novel Method to Achieve Optimization in Facial Expression Recognition Using HMM

D. Ngarjuna Devi, M. V. V. R. Murali Krishna Rao
Conference Papers InSignal Processing And Communication Engineering Systems (SPACES), 2015 International Conference on 2015 Jan 2 (pp. 48-52). IEEE.


Human-Computer Interaction is an emerging field of Computer Science where, Computer Vision, especially facial expression recognition occupies an essential role. There are so many approaches to resolve this problem, among them HMM is a considerable one. This paper aims to achieve optimization in both, the usage of number of states and the time complexity of HMM runtime. It also focusses to enable parallel processing which aims to process more than one image simultaneously.

  • image

    Deep Learning based Marker Detection with Drone

    Identifying certain kind of marker with deep learning through drone camera

    This is final project of Cyber Physical Systems (Fall - 2019, UNL) course. This project aims to identify the two types of markes through deep learning based object detection. The built model is attached to Parrot Drones camera to track the marker.

    The proposed method is tested in the simulation using TUM simulator, proceeding for real drone testing.

    Keywords - UAV, TUM Simulator, Parrot AR Drone, Object Detection, Deep Learning

    Technologies - Python, OpenCV, ROS, Gazebo, Tensorflow

  • image

    Multi Drone Landing

    Multi Drone landing using TUM Simulator

    This is final project of Robotics Today (Fall - 2019, UNL) course. This project addresses to land multiple drones using single fiducial marker. Drones lands on the surrounding area of fiducial marker such that other drones still use the marker to track the landing spot. Contour identification is used to detect the marker, and PD controller is used to control the drone.

    The proposed method is tested in the simulation using TUM simulator, proceeding for real drone testing.

    Keywords - UAV, TUM Simulator, Parrot AR Drone

    Technologies - Python, OpenCV, ROS, Gazebo

    Simulation result can be found below.

  • image

    Dead Chickens Removing using Machine Vision

    Broiler house mortality identification using image processing.

    This experiment is about to identifying the dead chickens in the commerical broiler house using image processing. This is a colloborative project of Agriculture Department and Computer Science department. Moniterd by Dr. Zhao and Dr. Zhang. I have worked for implementing the computer vision sytem to identify the dead chickens. This work is recently published in ASABE. Please look in my publications page for more details abou this project.

    We are now working to integrate this computer vision module to robot arm to pick up the dead chickens.

    Technologies - OpenCV, Matlab.

  • image


    Autonomous Quadcopter for Mail Delivery.

    This project is about to implement an autonomous quadcopter for mail delivery. We (I, Naresh, Anh, and Guoming) have implemented an DIY (Do It Yourself) quadcopter by buying the required parts, and we made it to fly by using the communication between Raspberry pi and a PC. Now we are working for it to fly by itself and through 3d path planning and SLAM.

    Keywords - Quadcopter

    Hardware - Raspberry Pi, Carbon Fiber frame, 2300KV Brushless Motors, 12A ESC, 6*3 Carbon Fiber Propollers

    Technologies - Python, OpenCV

  • image

    Unammaned Ground Vehicle for Poultry House

    Implementing an unmanned ground vehicle which is used by ABE researhcers.

    An unammned ground vehicle (Which was used by ABE Students to do some experiment in poultry house) was implemented by me. At first it was made as an autonomous robot, later it was converted to RC-based robot due to some experimental conditions.

    Hardware - SuperdroidRobotics Chasis, Sabertooth Motor Shield, Arduino, Specktrum DXe Transmitter & Reciver.

    Technologies - Arduino.

  • image

    AR alignment in HoloLens

    Aligning a virtual object in the real world.

    This project is about to aligning the virutal object in real world in a stable manner, which is the fundamental aim of Augmented Reality. I am working under Dr. Swan's supervison for this project.

    We are using HoloLens SLAM (Simultaneous Localization and Maping) to align object in a very stable manner in real world, Vuforia tracking for tracking the marker, and Unity for building the application.

    Hardware - HoloLens.

    Technologies - Unity, C#

  • image

    AR Robot

    Robot path planning in Augmented Reality (Contain both real and virutal obstacles) world.

    This is final project of AI Robotics (Spring - 2017, MSU) course. In this I implemented a robot which can navigate in the AR world, which contain both real world and vitual world. In this, the robot need to reach the goal position where virtual robot is there by using path planning and with the hlep of sensors. Ultrasonic range finder deals with real world obstacles, ODG R7 (AR glasses) with pixy sensor guides the virutal world. We have used Depth First Search (DFS) and Q - leanring (One of the active reinforcment learnings) for guiding the robot to reach the goal. You can see the results in the provided video there. This paper is accepted to one of the IEEE - Robotics and Automation Society, you can see details about this in my publications page.

    Keywords - Robotics, Augmented Realtiy, Path Planning, Reinforcement Learning, Depth First Search.

    Hardware - RoboIndia Chasis, CMU Pixy Sensor, ODG R7 Glasses, Ultrasonic range finders, Arduino UNO.

    Technologies - Arduino

    You can find the output video files below.

  • image

    Animal Classifier

    Deep learning based image classifier to classify animals.

    This is our Machine Learning (Fall - 2017, MSU) final project. We (I and Anh) implemented an image classifer to classify seven varities of animals using deep learning.

    Keywords - Deep Learning, Convolutiong Neural Networks.

    Technloiges - Python, Anaconda, Keras.

  • image

    Would you Survive in Titanic?

    This app can tell what are survival chances, if you were travel in Titanic.

    This is the final project of Visual Data Analysis with R (Spring - 2017, MSU). In this project I have implemented an SVM based classifier to identify whether the passenger surrvied in the titanic or not. After that I made some modifications and made an app which can predict your chance of surviival on different classes with different fare rates.

    Keywords - Data Processing, Support Vector Machine (SVM), Kernel-SVM.

    Technlogies - R.

    Know your predictions here

  • image


    A stargety based agent to play the RISK game.

    This our Artificail Intelligence (Fall - 2016, MSU) final project. In which we (I and Naresh) need to implement an angent (Our agent name is 'PeaceAgent') to play the RISK game. A competition was held with all the other agents of the class. Our agent secured 6th position in two-player competition and 2nd position in three-player competition.

    Keywords - Startegy.

    Technoloiges - Python.

  • image

    Reinforcement learning for Triwizard Cup

    Agents to reach the triwizard cup by learning itself about the world.

    This my Algorithms (Fall - 2016, MSU) course final project. In this project, I have implemented a Q-learning based trined agent to reach Triwizard cup by successfully avoiding the death eaters and obstalces. Python is used to implement the code Unity along with C# used for make the interface.

    Keywords - Reinforcement Learning, Q- learning.

    Technolgies - Python, Unity, C#.

    You can download the sample interface here.

  • image

    Movie Recommender System

    Recommending movies to users based on their previous interests.

    We (I and Naveen) have done this project as our final year project of undergradutaion. This project is guided by Ms. Sowmini Devi. In this we have implemetned a movie recommender system using the 'Imputation of missing values' of Machine Learing. A collaborative filtering approach is used to implement this model.

    Keywords - Colloborative Filtering, Matrix Factorization, Machine Learning.

    Technologies - C, Matlab.

  • image

    Blood Bank Management System

    A online web portal which integrates blood donors and recipents.

    For the 'Software Engineering' project we (I along with two of my classmates, Naveen and Graceleena) have implemented a web based blood bank management system, which can integrate the donors and recipents.

    Technlogies - HTML, JavaScript, PHP, MySQL


  • Chancellor's Fellowship Award, University of Nebraska-Lincoln, 2019
  • Student Research Travel Award Mississippi State University, for IELSX - 2018


  • Got 4th rank in the ranking system of SYSS, whose aim is to identify the rural background merit students.
  • Selected for RGUKT 6 year integrated B.Tech course among all the SSC pass students in our state.


  • Member of Upsilon Pi Epsilon
  • Student member of IEEE
  • Student member of IEEE - Computer Society
  • Student member of IEEE - Robotics & Automation Society
  • Student member of IEEE Young Professionals
  • Student member of ACM



  • 25th IEEE Conference on Virtual Reality and 3D User Interfaces, 2018.
  • Internatinoal Symposium on Mixed and Augmented Reality, 2019.

Contact & Meet Me

I would be happy to talk to you whehter you want to give me an advice on the work I am doing, or you need one for the work you are doing. I like constructive critisim, if you think any of the works I am doing in a wrong manner please let me know. If you want to meet personally, please look below to know where you can find me.

  •    mvvrmkr@gmail.com
  •    krishna@huskers.unl.edu
  •    KrishnaMuvva

At NIMBUS (Nebraska Intelligent Unmanned Systems) Lab

You can find me at NIMBUS Lab, which is located in the second floor of Schorr Center.

I come to NIMBUS Lab almost every week day.