Ikigai

The rise in computational power together with the pervasiveness of big data has created new computing modalities which has spurred an emergent digital economy, and has spawned new business value chains. Across these spates of recent developments, the question of resilience and reliability, safety and robustness, as well as the assured predictability of behavior in the (autonomous) systems that we design and build has become timely and crucial. Emergent autonomous AI systems must make decisions under imperfect circumstances based on environment information that is available from an array of perceptual inputs and their representation, reason through self-enumerated possible actions, and safely execute the ones that provide a meaningful state transition in the environment in which they act — all the while guaranteeing safety, stability, and resilience under external disturbances. Broadly, one can think of autonomous AI systems along two general threads: (i) situated autonomy, such as an automated airline ticket reservation system that reserves and makes decisions based on provided input by an external agent; and (ii) embodied autonomy, that is agents with perceptual sensors and proprioceptive actuators that effect a physical environmental change upon interaction; an example in this latter category are physical robots that sense the world directly through contact, and react adaptively to effect a change in representation of their environment via feedback. In both situated and embodied autonomy, we expect algorithms to deliver performance specifications without a glitch as we automate away tasks that humans have traditionally performed for over a century. My core research exists at the intersection of three intertwined fields that encapsulate important questions surrounding situatedness and embodiments: robotics, control theory, and machine learning. Robotics is crucial for the automation of physical embodiments. Control theory offers the crucial scientific recipé that establishes strong robustness and resilient foundations in otherwise flimsy AI models and policies (encompassing situatedness and embodiments). And machine learning provides the relaxation to devise generalist algorithmics that transcend narrow applications to broad usage. Motivated by what is possible with hitherto unexplored and large-scale engineering feats across these three disciplines, my research seeks to deconstruct the challenges that modern technology presents viz., (i) combining the elegance of mathematical rigor in imbuing resilience and robustness into the little margins that exist in safety-critical embodiments; (ii) reasoning possible automation capabilities into existence through abstract sub-task modules in a hierarchy of systems architectural principles; this is guided by clear conceptual analyses baked in the tradition of classical scientific rigor; and (iii) leveraging bio-inspiration to build emergent agent-centric situated AI and functional embodied automation systems. As I plan and act, I will continue to collaborate with various researchers with relevant expertise around the planet as we construct new insights on solving difficult automation problems, and broaden our understanding of the problems that are relevant to the future. Stimulating rapid growth in our understanding of critical structures to complex systems, we shall bring to bear new inventions to a safe and desired haven that fosters a neat co-existence with human beings in everyday living.

Further reading

Research Chronology

Select News Highlights

Archived news

Could we be at the precipice of continum soft robots that mimic the behavior of functional living organisms being versatile shoo-in for everyday robotic automation tasks? My new paper argues for and demonstrates the efficacy of a decentralized continuum dynamical system whose whole-body strain dynamics is resolved in an adaptive and hierarchical nonlinear control fashion. [Update Jan 2025]: Accepted at ACC in Denver. September 30, 2024
Recent talk at McGill's Mechanical Engineering and MILA, on a layered architecture and time-scale separation control scheme in the quest towards the real-time control of heterogeneous soft robotics devices. October 31, 2024
Will you be at the next ICRA conference? I am but a weave within the rich tapestry of excellent leaders in the soft robotics community, fashioned together to disseminate the latest findings -- new hardware fabrics, mathematical/ML models, and control tools -- in a workshop on biologically-inspired embodied intelligence for modern robots and devices. Please see the workshop webpage for details. August 24, 2024
Two papers accepted to CDC 2024, Milan! August 27, 2024
Please join the crescendo of ecstatic congratulations to Dr. Anurag Koul, a former postdoc in our lab and a close collaborator, who is transitioning his academic career into an applied scientist role at Amazon NYC. Bravissimo! August 8, 2024
Together with Volker Pohl, I am co-chairing the computational methods in control theory at the next Conference on Decision and Control in Milan (December 2024). Evviva! August 5, 2024
Raising a cheer of congratulations to Dr. Shaoru Chen, a former postdoc in our lab and collaborator, who recently sold his soul to LinkedIn joined LinkedIn! August 3, 2024
Organizing an MSR-wide workshop in Redmond, WA for folks working on group theory applications for model and policy and representation in modern learning algorithms. February 24, 2024
Recent presentation at Yale University based on our mixed H2/H-infinity stochastic RL policy optimization analysis. November 2, 2023

Mentoring Activities.

I enjoy working with/mentoring young students/scientists in teaching and research capacities alike. Before MSFT, I was an adjunct faculty at Brandeis up in Waltham, MA. Current or past students, interns and postdocs include:

Select Talks, Classes, and Conferences

Talk Highlights

Editorial/Community Services

Classes and Useful Presentations

Professional Activities.

(Strikethroughs signify societies I was a part of but whose membership I no longer belong).

Life, liberty, and the pursuit of the KOMs

As often as I can get away from the lab, I like to ride my (road/mountain/Grandma's) bike. I ride rain or shine, hell or highwater -- whether it's in the parching heats of sultry Texas, or the blustering cold winds of the Midwest and the Eastern Seaboard. States I have ridden so far include: NY, NJ, MA, NH, PA, DE, TX, IL, WI, IN, HI, WA. Below is a screenshot of my recent activities summary.


Recent Readings

Below are a list of books in my current reading list.

2024 Readings
Oldies and Goodies




This web presence was last updated on Wed, Mar 26, 2025.