Home

I have recently graduated from EECS, MIT with a Masters degree. I worked with Prof. Pattie Maes in MIT Media Lab. Previously, before joining MIT Media Lab, I was working with Prof. Seth Teller in RVSN group in CSAIL.

My research work during my Masters at MIT was about developing a very interactive and intutive platform to make complex information (especially Science, Technology, Engineering and Mathematics material) truly accessible to blind and visually impaired students by using a tactile(braille) device. An important goal of this project was to develop tactile information mapping protocols through which any tactile (braille) device can best convey educational material and other graphical materials. For my excellent contribution in the field of blind literacy, I was awarded Honorary Mention in Touch of Genius Prize for Innovation 2015 by Nationa Braille Press, USA. I have also been awarded India's first Dr. APJ Abdul Kalam Award given by Police Housing Corporation, Bhopal, M.P.





Research

My publications are listed below:

  • Rahul Kumar Namdev, K. Madhava Krishna, C.V. Jawahar
    Multibody VSLAM with Relative Scale Solution for Curvilinear Motion Reconstruction [Video]
    Accepted in International Conference on Robotics and Automation (ICRA) 2013
  • Rahul Kumar Namdev, Abhijit Kundu, K. Madhava Krishna, C.V. Jawahar
    Motion Segmentation of Multiple Objects from a Freely Moving Monocular Camera [PDF][Video]
    International Conference on Robotics and Automation (ICRA) 2012, vol no. pp.4092-4099
  • Arun Kumar Singh, Rahul Kumar Namdev, Vijay Eathakota, K. Madhava Krishna
    A Novel Compliant Rover for Rough Terrain Mobility [PDF][Video]
    IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), vol. no. pp.4788-4793
  • Rahul Kumar Namdev
    A Research Platform for Kondo KHR-3HV Humanoid Robots.
    Technical Report : Humanoid Robotics Institute(Waseda University Tokyo)

Courses

The courses I have taken in IIIT-H include the following:
  • Computer Vision
  • Mobile Robotics
  • Embedded Robotics
  • Pattern Recognition
  • Data Structures
  • Probablity and Stochastic Processes
  • Linear Algebra
  • Computer Networks
  • Digital Communication
  • Digital Design with Hardware Description Languauge
  • Time Frequency Analysis
  • Signals and System
  • Microprocessos Based System Design
  • Computer Organisation
  • Electronics Workshop-I, II
  • Information Theory and Coding

Personal

Receiving the Gold-Medal for BTech+MS Dual Degree in ECE from Prof. Raj Reddy(Turing Award winner, Prof. of ECE, Carnegie Mellon University), Governor of Andhra Pradesh (India) and Dr D Balasubramanian.
    IIIT-H Gold-Medal Pic
Demonstrating the working of Rural Area Twitter at Hack U Yahoo! 2010. We were awarded the first prize for most innovative project idea.
    Demonstrating the Working of Rural Area Twitter at Hack U Yahoo! 2010

Major Projects

Multibody Visual SLAM: In this work, we developed a practical vision based Simultaneous Localization and Mapping (SLAM) system for a highly dynamic environment. We adopted a multibody Structure from Motion (SfM) approach, which is a generalization of classical SfM to dynamic scenes with multiple rigidly moving objects. We proposed an efficient solution of relative scale problem in a multibody VSLAM framework for reconstructing moving objects and stationary world in a unified common scale. A unified dynamic 3D map of scene involving multiple moving objects, structure of 3d world, camera trajectory and trajectory of moving objects is the output of our system. A part of this work has been recently communicated to ICRA-2013. In this paper we proposed a solution to relative scale problem for curvilinear motion reconstruction.
Technical Environment: C++, OpenGL, OpenCV, Eigen, Lapack, MRPT.
Advisor: Dr. K. Madhava Krishna and Dr. C.V. Jawahar.

Motion segmentation of multiple moving object from a moving monocular camera: In this project, we developed an incremental system for efficient dense motion segmentation of multiple moving objects from a moving monocular camera. We used multiple cues based on motion vectors from dense optical flow, motion clues based on multi-view geometry framework to achieve dense segmentation of moving objects. This work was accepted for publication in ICRA 2012
Technical Environment : MATLAB, C++, OpenCV.
Advisor: Dr. K. Madhava Krishna and Dr. C.V. Jawahar.

Pipelined Implementation of JPEG Encoder using the JPEG standard: In this project, we implemented a fully pipelined JPEG encoder using the JPEG compression standard. The architecture of the design exploits the principles of pipelining and parallelism to obtain high speed and throughput. The design was implemented in Verilog, HDL and synthesized for Xilinx.
Technical Environment: Verilog, Xilinx, Synopsis.
Advisor: Ms. Ipshita Chakrabarty.

Micro Line follower and obstacle avoiding Robot: We implemented a tiny line following Robot. We were using two micro-controllers in this robot, one for sensory feedback and other for motion and path planning. The communication between the the two microcontroller was USART and I2C.
Technical Environment: C++, ATMEL Micro-controllers, Sonars and few other sensors.
Advisor: Dr. K. M. Krishna.

Identifying a modulation Scheme: We implemented a modulation identification algorithm for identifying M-ary shift keying using wavelet transform and higher order statistical moments. With this algorithm we were able to predict which of the modulation schemes from the following M-ary GMSK, M-ary FSK, M-ary PSK, M-ary QAM are being used.
Technical Environment: MATLAB
Advisor: Dr. Ram Bilas Pachori.

Robotics Arm: The Robotic Arm had two servo motors and it was constrained to move in horizontal plane only. The task of this project was detect the object placed inside the outer radius of the Arm. The location of the obstacle was displayed on an LCD.
Technical Environment: C++, Atmel Micro-contollers, Servo Motors
Advisor: Dr. K Madhava Krihsna.

Multi Robot Formation Control: We built a multi robot Formation Control System. In this system there was a Leader which was an autonomous obstacle avoiding robot. We used ultrasonic sensors for detecting the obstacles. The first follower was following the leader by maintaining an IR communication link and the second follower was following the first and so on.
Technical Environment: C++, Atmel Micro-controllers, Sonar Sensor, IR Sensors
Advisor: Dr. K Madhava Krihnsa.

Other Projects

  • Establishing an Ad-hoc Network of Xbee RF Communication Module
  • AutoHome (OpenMoko): Controlling various electronics devices from cell phone
  • Distance Measurement Using Ultrasonic and Sharp I.R Sensor
  • Time Frequency Analysis of Musical Instuments
  • Computational Analysis of perspective in paintings
  • Controlling a Robot By Mobile using Dual tone Multiple frequency Decoder
  • Building An Ornithopter

Internship Experience

Developing a Research Platform for Humanoid Robot (KHR-3HV)
Humanoid Robotics Institute (Waseda University Tokyo)
May-July 2010
I developed a fully integrated wireless research platform including hardware, firmware and software. The hardware used was Renesas and DSPIC micro controllers along with some more micro controller chips. The Software was developed in C++. This software was controlling around 24 serial servos and was able to provide 18 degrees of freedom to the KHR-3HV humanoid robot.
Technical Environment : DSPIC and Renesas Micro-controllers, Humanoid KHR-3HV, C++
Advisor: Dr. Shuji Hashimoto (Vice President, Waseda University Tokyo )


Implementation of four wheeled active suspension rover with minimal actuation for rough terrain mobility August-December 2009
In this Internship we implemented a four wheeled active suspension rover with minimal actuation. The suspen- sion mechanism was derived from a planar four bar mechanism and required two actuators to control the internal configuration of the rover. This work has been published in IROS 2010.
Technical Environment : Nastran Simulation Software, ATMEL Micro-controllers, C++
Advisor: Dr. K. Madhava Krishna(Assistance Professor, Robotics research Lab )