I am broadly interested in Design and Validation of Autonomous Systems, which lies at the intersection of Formal Methods, Real-Time and Embedded Systems, Control Theory with applications in Robotics, Automotive, Industrial and Home Automation.

Autonomous Systems offer hope towards moving away from mechanized, unsafe, manual, often inefficient practices. Last decade has seen several small, but important, steps towards making this dream into reality. Today, Autonomous Systems have made its place almost everywhere: from driving, ensuring safety and efficiency in factory floors, assisting doctors in surgeries, wearables and home assistants, shaping better public policies, to making perfect Cheetos! Needless to say, designing and validation of Autonomous Systems is challenging: ensuring safety and correctness in presence of uncertainties, complex nature, heterogonous and distributed compute resources, imparting intelligence, model-implementation semantic gap, etc.

To this end, I have proposed several solutions to solve some of the issues mentioned above.
(i) Techniques to prove safety and correctness of systems under the presence of uncertainties.
(ii) Techniques to compute the amount of uncertainty an Autonomous Systems can tolerate.
(iii) Not all uncertainties are equal—some are more dramatic than others.
(iv) Quantitative techniques to analyze safety, where qualitative analysis fail.
(v) Monitoring-based techniques to analyze safety and correctness of Autonomous Systems, where traditional analysis doesn’t scale.
(vi) Optimal design approaches to design Autonomous Systems that uses heterogenous/distributed compute resources with violating safety and correctness.

Designing autonomous systems not only entail a variety of technical challenges, but also many societal and philosophical ones—an aspect that I intend to study more in future.

Towards achieving this goal, my list of publications can be found here.