Decision Making in EV/Mobility
The electrification of mobility is fundamentally reshaping energy demand across urban infrastructure and enterprise operations. Electric vehicles (EVs) are evolving beyond their traditional role as transport assets to become mobile energy buffers, capable of dynamic interaction with buildings, charging infrastructure and the power grid. This talk will present a unified approach to leveraging EVs for cost-effective energy management across two complementary domains: smart buildings and logistics operations. In the first part, the talk will explore multi-agent reinforcement learning (MARL) frameworks for the joint control of HVAC systems and EVs within buildings. By modeling EVs as stochastic energy buffers, the approach will discuss how to capture the uncertainty in vehicle availability while ensuring thermal comfort and respecting state-of-charge constraints, which enables adaptive load shaping under time-of-day pricing regimes, leading to significant energy cost savings. The second part will focus on enterprise-scale energy optimization, where EV fleet routing is tightly coupled with energy procurement decisions across multiple sources. The talk will discuss hybrid iterative optimization frameworks that integrate approximation techniques for energy sourcing and fleet routing, ensuring scalability across diverse service locations and fleet sizes. Together, these approaches will illustrate how EVs can be orchestrated as intelligent agents within cyber-physical systems for scalable and optimized operations.
Date and Time
Location
Hosts
Registration
Add Event to Calendar
Speakers
Dr. Prasant Misra of TCS Research
Decision Making in EV/Mobility
The electrification of mobility is fundamentally reshaping energy demand across urban infrastructure and enterprise operations. Electric vehicles (EVs) are evolving beyond their traditional role as transport assets to become mobile energy buffers, capable of dynamic interaction with buildings, charging infrastructure and the power grid. This talk will present a unified approach to leveraging EVs for cost-effective energy management across two complementary domains: smart buildings and logistics operations. In the first part, the talk will explore multi-agent reinforcement learning (MARL) frameworks for the joint control of HVAC systems and EVs within buildings. By modeling EVs as stochastic energy buffers, the approach will discuss how to capture the uncertainty in vehicle availability while ensuring thermal comfort and respecting state-of-charge constraints, which enables adaptive load shaping under time-of-day pricing regimes, leading to significant energy cost savings. The second part will focus on enterprise-scale energy optimization, where EV fleet routing is tightly coupled with energy procurement decisions across multiple sources. The talk will discuss hybrid iterative optimization frameworks that integrate approximation techniques for energy sourcing and fleet routing, ensuring scalability across diverse service locations and fleet sizes. Together, these approaches will illustrate how EVs can be orchestrated as intelligent agents within cyber-physical systems for scalable and optimized operations.
Biography:
Dr. Prasant Misra is a TCS Distinguished Senior Scientist in the Data and Decision Sciences division of TCS Research (Office of the CTO), and a Visiting Faculty in the Department of Cyber Physical Systems at the Indian Institute of Science, Bangalore. He received his Ph.D. in Computer Science and Engineering from the University of New South Wales (UNSW), Sydney, in 2012, and completed his postdoctoral fellowship at RISE SICS Stockholm (formerly the Swedish Institute of Computer Science) in 2013. His current research is centered around modeling, optimization and decision intelligence (Physical AI) for cyber-physical systems operating at the intersection of mobility and energy domains.
Address:TCS Distinguished Senior Scientist, Data and Decision Sciences division, TCS Research (Office of the CTO), Bangalore, Karnataka, India