For email: use only the letters in perentheses and no spaces.
I am an INSPIRE Faculty fellow in the Robert Bosch Center for Cyber Physical Systems (RBCCPS) in IISc Bangalore. I received my Ph.D. degree in Mechanical Engineering (2016) from the Georgia Institute of Technology. I started my career in the field of robotics, especially in the domain of legged robots. I have worked with robots like AMBER1, AMBER2, DURUS-2D, and DURUS. My primary focus as a PhD student was on stability and control of walking robots. I have diversified to other fields like safety-critical control, and deep reinforcement learning for all kinds of robotic platforms. I currently have following major activities:
- I have been recently exposed to safety-critical control in all kinds of robotic platforms. Control barrier functions (CBFs) have proved to be extremely effective in real-time QP based controllers. I am currently focusing on introducing some notion of robustness for these safegaurding controllers. The notion of robustness for safety is still in its nascent stages. Watch the video below where a tunable gain (epsilon) is varied to make the drone more and more conservative in terms of safety. The brown square indicates the "safe" region, and the destination is outside this region. Due to GPS estimation errors, the drone overshoots away from the boundary, thereby making it unsafe. We have robustified the controller to overcome this overshoot via the notion of input-to-state safety. More details are in this letter:
- Deep learning is getting a lot of attention of late. I am currently in the phase of understanding deep reinforcement learning (D-RL) and see how it can be applied to legged robots. This is in collaboration with Dr. Bharadwaj Amrutur, Dr. Shalabh Bhatnagar and Dr. Ashitava Ghosal in the center. See video below of our most recent walking result! Gaits were generated via D-RL based training, and then implemented via motion primitives. More details are in these papers:
- I am interested both in theoretical and practical aspects of hybrid robotic systems. My PhD dissertation was about bridging the gap between theory and experiment by using input-to-state stabilizing control Lyapunov functions (ISS-CLFs). I am looking to extend it for various other robotic platforms and also develop novel robust controllers based on this idea. More details are in this paper:
S. Kolathaya and A. D. Ames. Input-to-State Safety With Control Barrier Functions. 2018. IEEE Control Systems Letters. Volume 3, Issue 1, Pages 108-113. .pdf .bib
A. Singla and S. Bhattacharya and D. Dholakiya and S. Bhatnagar and A. Ghosal and B. Amrutur and S. Kolathaya.
Realizing Learned Quadruped Locomotion Behaviors through Kinematic Motion Primitives.
in IEEE International Conference on Robotics and Automation.
S. Kolathaya et. al.
Trajectory based Deep Policy Search for Quadrupedal Walking.
in 28th IEEE International Conference on Robot & Human Interactive Communication (RO-MAN).
S. Kolathaya and J. Reher and A. Hereid and A. D. Ames. Input to State Stabilizing Control Lyapunov Functions for Robust Bipedal Robotic Locomotion. 2018. in 2018 American Control Conference (ACC), pages 2224-2230. .pdf .bib .video