Close

Gokul narayanan

Robot Software Engineer

Download Resume

About Me

"Robot software engineer with 3 years of experience in building robots and deploying it in real-world. A visionary individual who always seeks inspiration from solving challenging issues in the society for the betterment of humankind. Have worked on grounbreaking projects ranging from surgical robots, medicare robots to largescale automation systems. Currently, my research focus is on learning for motion planning."

Experience

SIEMENS

Advanced Robotics Intern

  • Designing and developing a modular automation solution to stitch garments using collaborative robot (UR3) which will increase the production rate by 20%.
  • Implemented pose estimation of garments on a worktable using Intel realsense D400.
  • Implemented stacking of garments using UR3 manipulator and Robotiq suction gripper.
  • Developing prototypes to fold clothes and stitch multiple clothes together.
  • Developing state machines in ROS for automating the garment stitching task using manipulator.
  • Performance comparison of motion planning algorithms (sampling based vs search based vs optimization based).

Worcester Polytechnic Institute

Research Assistant

  • Developed low level drivers for interfacing Variable friction finger gripper.
  • Developed an 'Modified A* algorithm' for performing in-hand manipulation of objects using the gripper .
  • Developed a framework integrating motion planner and visual servoing to handle uncertainities while performing the task.
  • Submitted the work as a paper for ICARA 2020, which is currently under review.
  • Developed a symbolic task planner integrated with a hierarchical reinforcement learning framework to perform loco-manipulation task.

ASIMoV Robotics

Senior Robot Software Engineer

  • Supervised a team of three people in developing a software stack of an autonomous mobile robot.
  • Developed custom hardware interface node using ROS controllers for a 6DOF manipulator.
  • Developed inverse kinematics package for constrained motion planning problem.
  • Benchmarked various OMPL planners on a 6 DOF manipulator for constrained motion planning problem.
  • Implemented Cartesian path planner for waypoint navigation of a 6 DOF manipulator.

ASIMoV Robotics

Robot software engineer

  • Built and deployed India’s first mobile humanoid robot assistant ‘Intelligent Robot Assistant(IRA)’ in HDFC bank.
  • Developed the local and global path planner for the robot to navigate in a cluttered environment.
  • Developed task planner for the robot using state machines in ROS.
  • Developed software modules for the robot to assist patients in hospitals.

Education

Worcester Polytechnic Institute

August 2018-August 2020

Masters in Robotics engineering

Coursework

  • Motion planning, Human robot interaction, Robot Dynamics and Control, Advanced Robot Navigation.
  • PSG College of Technology

    August 2012 - May 2016

    Bachelor in Robotics and Automation Engineering

    Coursework

  • Computer vision, Introduction to AI, Foundation of robotics, Neural networks
  • Projects

    Sewing Robot

    Automating the sewing industry using an UR5 robot. Developing the software pipeline to automate the stitching process from estimating the garments pose to stitching them using sewing machine. Building prototypes to fold clothes and stitch multiple clothes together. Currently working on this project as an intern in SIEMENS and in close collaboration with a startup called SEWBO.

    '

    Intelligent Robot Assistant(IRA)

    IRA was India's first humanoid robot which was deployed in HDFC bank, India for customer service. IRA can detect customers entering the bank and can interact with them through a touch screen. It can also give various information about the bank to customers and can guide the customers by taking them to various counters in the bank by autonomously navigating the bank environment. This project was developed during the period when I was working in ASIMOV robotics.

    View Project

    Within Hand manipulation using Variable friction finger gripper

    Developed a hybrid planning method integrating offline motion planning with online visual servoing to perform within hand manipulation of objects using variable friction finger gripper. Evaluated the performance of the system with inaccuracies in the modelling of the system and uncertainities in the motion of the object. Submitted the work in ICRA 2020 which is under review. Currently developing a reinforcement learning based algorithm to manipulate objects without the need for explicit modelling.

    View Project

    Sayabot-A Service robot

    Sayabot is a service robot that was designed to work seamlessly alogside humans with capablities like autonomous navigation, natural language processing and face recognition.It was designed to work on wide array of places such as hospitals, airports,banks, hotels, corporate offices, hotels, retails store. This project was developed during the period when I was working in ASIMOV robotics.

    View Project

    Craniobot

    This project was funded by Biotechnology Industry Research Assistance Council (BIRAC). The aim of the project was to develop a minimally invasive robotic surgery system that can do craniofacial surgery. For the initial research phase, the prototype was developed in a 10x bigger version. And I was involved in developing the ROS based software architecture system for the prototype using Moveit Interface. This project was developed during the period when I was working in ASIMOV robotics.

    View Project

    Remote manipulator control

    The aim of the project was to develop an teleoperating interface for a 6DOF manipulator. I was involved in developing the motor drivers and controllers in ROS. And also developed the Moveit interface for the manipulator to solve the constrained motion planning problem. This project was developed during the period when I was working in ASIMOV robotics.

    View Project

    Comparsion of trajectory tracking controllers using UR5 robot

    The aim of the project was to compare the performance of the various trajectory tracking controllers such as Inverse dynamic controller, Robust feedback linearization controller and Passivity based controller under different scenarios. All these controllers where developed in Matlab and were tested on a UR5 manipulator with a benchmark trajectory dataset. This project was done as a course project for Robot controls subject.

    View Project

    EKF based localization for mobile robot

    The aim of the project was to implement the Extended Kalman Filter based localization for a mobile robot from scratch, to better understand the mathematical principles behind the algorithm. We developed a virtual environment in Gazebo with landmarks. Then we extracted the position of landmarks in the environment by processing the data from the 2D lidar fixed on the mobile robot. Developed the motion model and the observation model for the mobile robot in C++. Finally, we were able to estimate the position of the robot in the environment using the EKF algorithm. This project was done as a course project for Advanced robot navigation subject.

    View Project

    Skills

    Get in Touch