Rekha Raja
Tata Consultancy Services
Plot No.A-44 & 45 first floor
Block- C & D, Sector-62
Noida-201301, Up, India
raja.rekha@tcs.com
+91 9205908076
RESEARCH INTERESTS
- Robot motion planning, multi robot trajectory planning
- Computer vision and machine Learning applications to robotics
- Visual servoing, reinforcement learning
- Redundant manipulators, dexterous manipulation
- Trajectory tracking control, robot kinematics and dynamics
EDUCATION
2010-2016 P.hD, Mechanical Engineering, Indian Institute of Technology (IIT) Kanpur, India
Thesis: Rough Terrain Motion Planning and Redundancy Resolution of an Articulated Rover Manipulator System
Supervisors: Prof. Ashish Dutta & Prof. Bhaskar Dasgupta. CGPA 7.75/10
2007-2009 M.Tech, Mechatronics, Indian Institute of Engineering Sc. and Technology (IIEST), India
Thesis: Obstacle Avoidance & Navigation of Autonomous Mobile Robot
Supervisors: Dr. Sankar Nath Shome and Dr. Sambhu Nandy. CGPA 7.67/10
2004-2007 B.Tech, Information Technology, Jalpaiguri Govt. Engineering College, Jalpaiguri, India
Thesis: PC to PC Wireless Communication
Supervisors: Dr. Swapan Kumar Sengupta. CGPA 7.98/10
EXPERIENCE AND EMPLOYMENT RECORDS
Nov 2015- Employer: Tata Consultancy Services (TCS), Noida, India
Nature of job: Research and Development.
Role: Scientist
Responsibilities: I am engaged with various activities such as
- Grasping and manipulation: In this work, our ultimate goal is build an autonomous robotic systems to automate the process of picking and stowing a wide variety of unknown objects from and to a rack. Grasping and manipulating unknown objects is a complex task and difficult to solve. I am primarily responsible for finding suitable robot end-effector trajectories for efficient pick and place of objects. Therefore, developed a motion planning algorithm for UR5 robot manipulator using RRT-Connect path planning algorithm available in MoveIt!.
- Currently, implementing and developing visual servoing algorithms on Gazebo, ROS platform to control the UR5, UR10 robot manipulator employing various methods such as path planning in image space using RRT, reinforcement learning and deep reinforcement learning to perform the planning and manipulation task.
- Multi-robot collaboration: working with multi-agent trajectory generation for warehouse automation where each robot has to assign task to reach multiple goal positions to pick the order and deliver to the customer in another location. This involve dynamic trajectory planning, which investigates the re-planning of activities in multi-agent warehouse automation.
- Assists team members and collaborators, in planning and designing research experiments; establishing priorities and recommending schedules, time tables.
- Presentation of research work and results to internal and external bodies.
- Publish and present research papers in conferences and journals.
Aug-Dec 2015 Employer: Computer Vision Lab, IIT Kanpur, India
Nature of job: Research and Development.
Role: Project Engineer
Responsibilities: There was many responsibilities such as
- To worked in an industrial project where I developed a computer vision algorithm based Support Vector Machine (SVM) to detect deviation in alignment of the weldsim during the pipe expansion process. The pipe expansion process is an essential step in the manufacturing process of large diameter pipes for the oil and gas industries. The algorithm worked in real time and produced alarms as well as proportional corrective signals.
- Assists graduate students with their projects and research work.
- Communicate with the client to understand their requirement. Present, demonstrate and deliver work to the client.
- To write and publish papers in conferences and journals.
Jan-Dec 2009 Employer: Central Mechanical Engineering Research Institute, India
Nature of job: Research and Development.
Role: Junior Research Fellow (JRF)
Responsibilities:
- To implement a path planning algorithm for mobile robot by avoiding obstacles. I was trained in programming languages and other computer related technology during and exposed to the interplay of electronics and mechanics which constitute a robotic system. As a part of project, I have implemented a Fuzzy Logic based path planning algorithm for a P3DX mobile robot which navigates in a constrained environment.
- To write and publish papers in conferences and journals.
RESEARCH PROJECTS
2016 Amazon Picking Challenge
[TCS] This work required to automate the process of picking and stowing a wide variety of objects from and to a rack. We developed algorithm for object recognition, pose detection, grasping, motion planning, task execution, and error detection. In the task the robots scored by how many items are picked and stowed in a fixed amount of time. I have developed motion planning algorithm using MoveIt! to perform the task. The trajectory is generated using RRT motion planning algorithm. The desired grasp position is obtained from grasping algorithm which is given as input to the motion planner to generate the trajectory from current configuration of the manipulator to the desired configuration.
2015 Automated pipe expansion process
[Computer I worked in an industrial project where, I developed a computer vision algorithm based on
Vision Lab] Support Vector Machine (SVM) to detect deviation in alignment of the weldsim during the pipe expansion process. The pipe expansion process is an essential step in the manufacturing process of large diameter pipes for the oil and gas industries. The algorithm worked in real time and produced alarms as well as proportional corrective signals.
2011 Visual motor control of 7DOF manipulator using KSOM
[EE671] The objective was to guide a robot manipulator to reach a target position in its workspace using visual feedback. The Visual Motor Control (VMC) considered here is a visual-servoing, in which a feasible inverse kinematic solution is obtained at position level. VMC primarily involves two tasks, namely extracting the coordinate information of the robot end-effector and the target from camera images and then finding out necessary joint angle movement for reaching the target. The second part is done using Kohonen SOM algorithm.
2011 Fuzzy Improved Mountain Clustering
[EE658] In this projects I reviewed three of the most representative off-line clustering techniques: K-means clustering, Fuzzy Cmeans clustering, Mountain clustering. With the concept of these clustering techniques we have developed a new algorithm that is called Fuzzy improved mountain clustering. The developed new fuzzy improved mountain clustering technique is similar to mountain clustering, except that instead of calculating the density/potential function at every possible position in the data space, it uses the positions of the data points to calculate the density/potential function, thus reducing the number of calculations significantly.
2011 CAD GUI for Curve and surface modelling
[ME751] The objective of this project is to make a suitable tool for drawing and study different types of curves and surfaces and their properties. Different types of curves and surfaces such as simple parametric, Bezier and spline and their properties are studied to make software to draw all these curves and surfaces, their modifications and properties. We have used MATLAB GUI platform to developed this software. For drawing these curves in Matlab the algorithms were developed and implemented. The software is made such that the drawing and modification of these curves can be done more interactively.
2010 Path planning of multiple mobile robots
[EE752] I developed an algorithm using potential field method for multiple mobile robots to reach a goal simultaneously avoiding the obstacles. Near the goal robots do not collide with each other as robots generated repulsive forces among them.
PUBLICATIONS
Journal Articles
- S. Kumar, A. Majumder, S. Dutta, R. Raja, S. Jotawar, A. Kumar, M. Soni, V. Raju, O. Kundu, E. Hassan, L. Behera, K. S. Venkatesh and R. Sinha, "Design and Development of an automated Robotic Pick & Stow System for an e-Commerce Warehouse." arXiv preprint arXiv:1703.02340 (2017).
- R. Raja, A. Dutta, and K. S. Venkatesh. "New potential field method for rough terrain path planning using genetic algorithm for a 6-wheel rover" Robotics and Autonomous Systems 72 (2015): 295-306, doi:10.1016/j.robot.2015.06.002.
- R. Raja, S N. Shome, S. Nandy, R. Ray, ''Obstacle Avoidance and Navigation of Autonomous Mobile Robot”, Advanced Materials Research, Vol 403–408, NOVEMBER/2011/4633-4642, (2011).
Conferences/Workshops/Posters
- R. Raja, A. Dutta, “Path planning in Dynamic Environment for a Rover using A* and Potential Field Method”, International Conference on Advance Robotics (ICAR), Hong Kong, 10-12 July, 2017.
- R. Raja, B. Dasgupta, A. Dutta, “Cooperative motion planning of redundant rover manipulators on uneven terrains”, International Conference on Advance Robotics (ICAR), Hong Kong, 10-12th July, 2017.
- R. Raja, S. Kumar, “Advances in motion planning for industrial robot manipulators”, ICRA workshop on dynamic motion planning, Singapore, 29th May, 2017.
- B. Hailu, R. Raja, A. Dutta, “Advanced KSOM based Redundancy Resolution of a Mobile Manipulator System for Motion on an Uneven Terrain”, International Conference on Advances in Robotics, Delhi, 28-2 July, 2017.
- R. Raja, S. Kumar, “A hybrid image based visual servoing for 6-D manipulator using kinect”, International Conference on Advances in Robotics, Delhi, 28-2 July, 2017.
- R. Raja, B. Dasgupta, A. Dutta, “Motion planning and redundancy resolution of a rover manipulator”, IEEE-WIE Conference, Bangladesh, Dec, 2015.
- A. Bhasin, R. Raja, A. Dutta, “Non-holonomic mobile manipulator kinematic control using hybrid simulated annealing”, IEEE-WIE Conference, Bangladesh, Dec, 2015.
- R. Raja, A. Dutta, “Motion planning of a mobile manipulator using fuzzy controller to dexterity measures”, IEEE-WCI Conference, IIT Kanpur, India, Dec, 2015.
- R. Raja, A. Dutta, B. Dasgupta, “KSOM based learning for cooperative motion of a redundant rover-manipulator system on uneven terrain”, International Conference on Control, Automation, Robotics and Vision, ICARCV 2014, Singapore, Dec 10-12, 2014.
- R. Raja, A. Dutta, B. Dasgupta, “Redundancy resolution of a redundant mobile manipulator”, Research Scholars' Day, Department of Mechanical Engineering, IITK on 21st March, 2014.
- R. Raja, A. Dutta, “Path planning algorithm for a Planetary Rover with Fuzzy Logic Adaptation and a Combination of A* and Potential Field Method”, IEEE Workshop on Computational Intelligence: Theories, Applications and Future Directions, IIT Kanpur, 14th July 2013.
- R. Raja, A. Dutta, “Path planning algorithm for a Planetary Rover with Fuzzy Logic Adaptation and a Combination of A* and Potential Field Method”, IEEE Workshop on Computational Intelligence: Theories, Applications and Future Directions, IIT Kanpur, 14th July 2013.
- R. Raja, A. Dutta, “Path Planning and obstacle avoidance of a mobile robot”, International workshop on Advances in Robotics, 5-7th July, 2012, IIT Delhi.
In Preparations
- R. Raja, A. Dutta, B. Dasgupta, “Parameterized KSOM for Redundancy Resolution of a 14-DOF Mobile Manipulators on Rough Terrain”, Journal of Applied Soft Computing [submitted].
- S. Kumar, A. Majumder, S. Dutta, S. Jotawar, R. Raja, M. Soni, V. Raju, O. Kundu, E. Hassan, L. Behera, K. S. Venkatesh and Rajesh Sinha, “Design and Development of an automated Robotic Pick & Stow System for an e-Commerce Warehouse”, IEEE Transaction on Autonomous Science and Engineering [submitted].
- Hitesh, R. Raja, A. Dutta, “Learning the Inverse Kinematics of a Mobile Manipulator System using Support Vector Machine for Motion on Uneven Terrain”, Expert Systems With Applications [submitted].
- R. Raja, D. Daina, P. Mithun, S. Kumar, P. Mithun, “Image based path planning for visual servoing of a manipulator”, Robotics and Autonomous Systems [Under preparation].
AWARDS/FELLOWSHIP
Dec, 2015 Young Researcher Award in the Area of Computational Intelligence from IEEE UP Section Computational Intelligence Society Chapter.
Dec, 2015 Best paper award in IEEE-WIE Conference, Bangladesh.
2010-2014 MHRD scholarship for pursuing Ph.D. at the IIT Kanpur, India.
Dec, 2014 Travel grant for attending 2014 IEEE International Conference on Control, Automation, Robotics and Vision, ICARCV, Singapore (sponsored by IIT Kanpur).
2007-2009 MHRD GATE scholarship for pursuing M.Tech.
INVITED TALK
- Invited talk entitled “Computational intelligence: Concepts and Applications”, QIP program, IIT Kanpur, 8th December 2017.
TEACHING EXPERIENCE
Spring 2015 Teaching Assistantships for ME769 - Advanced Topics in Robotics
Fall 2014 Teaching Assistantships for ME766 - Robot Motion Planning
Spring 2013 Teaching Assistantships for ME762 - Introduction to Robotics
Fall 2013 Teaching Assistantships for ME763 - Robot Dynamics and Control
Fall 2012 Teaching Assistantships for 101 - Engineering Graphics
SKILLS
Software
Programming Languages: C, C++, VC++, MATLAB, Python, ROS.
Tool/Library: OpenCV, PCL, Gazebo, MoveIt, VISP, Mathematica.
Operating System: Linux, Dos, Windows XP/7/8/10.
Query Language: SQL.
Mechanical design: Inventor.
Hardware
- UR10, UR5, PUMA, Barret Arm, SCARA manipulator.
- All Terrain Rover, Turtlebot, Pioneer 3-DX mobile robot, XWAM.
- Arduino hardware and software environment, Simple/smart servo motor control, Xbee.
- Microsoft Kinect, sonar, IR and laser sensors.
OTHER ACTIVITIES
- Head Student Section, State, IEEE Computational Intelligence Society, India, 2013 – 2015.
- IEEE member, IEEE-RAS member.
- Reviewer of RAS, Robotica, AIR, WIE, CIS.
PERSONAL DETAILS
Date of Birth: 25th June 1984
Sex: Female
Nationality: Indian
Marital Status: Married
Language Proficiency: English, Bengali, Hindi