+ Vivek Shankar Varadharajan
I am a roboticist and a researcher trying to understand, apply and evolve the concepts in robotics. In 2015, I received my M.Sc. degree in Automation and Robotics with a master thesis on An Aspiration set approach to an Orientering TSP under the supervision of Prof. Günter Rudolph from Technical University of Dortmund (Dortmund, Germany), I received my bachlors degree in Electronics and Instrumentation Engineering from Bannari Amman Institute of Technology, Anna University (Tamil Nadu, India) in 2012. At present, I am a Ph.D. student in computer engineering at the MIST Labs, École Polytechnique de Montréal (Montréal, QC, Canada). My research interests include machine learning, artificial intelligence, distributed robotics and Cyber-Physical systems.
Apart from my study, I love swimming and I do swimming in my time off.
I am always looking forward to new collaborations, if you find any of my work interesting and you wish to collaborate, feel free to write to me.
+ Current Projects
- Buzz in ROS ROSBuzz Creating a ROS package for Buzz:
ROSBuzz focuses on extending Buzz in ROS. Buzz is a heterogeneous programming language for programming a swarm of robots. It provides a means for coordinated task accomplishment with a swarm of Robots. The focus of this project is to implement Buzz Virtual machine (BVM) as a ROS Package. Current work on ROSBuzz focuses on testing the built package with state-of-the-art algorithms on a fleet of heterogeneous drone swarms. The drone swarm constitutes 3DR solos and DJI Matrix 100.
- Updatable Buzz Dynamically updating rules for the swarm after deployment:
Updatable Buzz is about building an updatable BVM. The focus of this project is to achieve continuous integration in a fleet of robots performing coordinated task. The main bottleneck of creating an update for a swarm, while in operation is to reach a consensus agreement.
- A distributed data storage mechanism SOUL: data sharing for robot swarms:
One of the main problems in designing distributed algorithms is that large amount of data has to be transported from one node to multiple others in a reliable manner, to utilize the full potential of the distribution. This project focuses on designing and exploring methods for efficient data storage and transport in robot swarms and distributed nodes.
- An approach to environmental model for robot swarms Distributed planning for robot swarms with an environmental model:
Environment is generally hostile for robot, there are several factor influencing the control of robots. In design of any control algorithm a nose factor is taken into consideration. If the robots are given a map of hazards for a given environment they are planning to navigate, they can effectively plan and avoid these pitfalls. This project focuses on developing methods to effectively model the environment for planned task execution.
- My doctoral research is focused on Software eco-system for long-term autonomy in robot swarms:
There exists a software ecosystem (App store) for smart phones and tablets, with which the end user can utilize the hardware capabilities of a phone to perform the assigned task just with a click (installing). Similarly, a reliable and easy configurable software ecosystem is to be created to exploit the capabilities of a swarm of autonomous robots. This could take a leap and make the swarm to be deployable in real world scenarios to accomplish magnanimous tasks by an user with limited knowledge.
+ Scholarship and Awards
- I am a recipient of BSFD student scholarship (\$40000), Awarded by Ecole Polytechnique de Montreal.
- I and the team from MIST lab won 2nd place at ImplementAI Hackathon, 2017. (If you are wondering what made us win, check out our submission here: DevPost)
- I won 3rd place at IEEE 3rd research boost 2017, poster presentation.
- I and the team from MIST lab won the Hackatown 2018 Hackathon with 40 teams competing. (If you are wondering what made us win, check out our submission here: DevPost)
+ Select Publications
- Vivek Shankar Varadharajan, Bram Adams and Giovanni Beltrame. The unbroken telephone game: keeping swarms connected. International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2019. (Accepted)
- Marcel Kaufmann, Vivek Shankar Varadharajan, and Giovanni Beltrame. A SelfAdaptive Data Handling System for Small Satellites and its Impact on Future Satellite Operations. International Astronautical Congress, Bremen, Germany. 2018.
- Vivek Shankar Varadharajan, David St-Onge, Christian Guss, and Giovanni Beltrame. Over-The-Air updates for robotic swarms. IEEE Software, 35(2):44–50, 2018.
- Vivek Shankar Varadharajan, David St-Onge, Ivan Svogor, and Giovanni Beltrame. A Software Ecosystem for Autonomous UAV Swarms. International Symposium on Aerial Robotics, Philadelphia, 2017.
- Aswinkarthik Ramachandran Venkatapatthy, Vivek Shankar Varadharajan, and Pritiviraj Ramalingam. Low Cost Augmentative Controller for SPWM Inverter with Photovoltic Utility. ICCGE, Hong Kong, IPCBEE vol.27, 2012.
+ Publications Under Review
- Vivek Shankar Varadharajan, David St-Onge, and Giovanni Beltrame. SOUL: data sharing for robot swarms. Autonomous Robots, 2019. Undergoing 2nd revision.
- Hassan Fouad, Vivek Shankar Varadharajan, and Giovanni Beltrame. Towards Energy Persistent Multi-Robot Missions. Robotics and Automation Letters, 2019. Under Review.