Abstract
The real world is full of uncertainties, and implementing any policy invariably encounters constraints. To foster more robust decision making, it is crucial to develop policies that inherently account for uncertainties and underlying constraints. This talk delves into the design of stochastic model predictive control (SMPC) strategies that ensure system stability and performance despite challenges like packet dropouts and intermittent observations. By integrating estimation techniques, such as Kalman filtering, with control mechanisms, these approaches maintain mean-square boundedness in systems affected by unreliable communication channels. Furthermore, the incorporation of deep learning into MPC frameworks is explored to manage uncertainties in nonlinear systems, employing dual-timescale adaptation methods to uphold stability during the learning process.

Short CV
Prabhat K. Mishra is an Assistant Professor at the Indian Institute of Technology Kharagpur(IIT Kgp) where he does research at the intersection of control theory and artificial intelligence for safety-critical applications and cyber-physical systems. He received a Ph.D. from the Indian Institute of Technology Bombay (IITB) and a Swiss Government Excellence Scholarship to study at Ecole polytechnique federale de Lausanne (EPFL) for a year. He was a Postdoctoral Associate at the Massachusetts Institute of Technology (MIT) and a Postdoctoral Research Associate at the University of Illinois at Urbana-Champaign (UIUC) before moving to IIT Kgp.