KrishnaKumar
- krishnak@utexas.edu
- (512) 232-4406
- ECJ 9.227B
Physical AI; Learned world models (graph neural networks and operator learning); Differentiable simulations (MPM, DEM, LBM); Uncertainty-aware control for robotics in deformable environments; Digital twins for hazard and infrastructure response
Bio
Krishna Kumar is a core faculty member at the Oden Institute for Computational Engineering and Sciences. Kumar’s research develops differentiable world models for physical AI—methods that connect perception, learned simulation, and control to enable autonomous systems to manipulate deformable materials and navigate extreme environments. His group builds graph neural network simulators and neural operators that preserve physical conservation laws while running fast enough for real-time decision-making. Applications include autonomous earthmoving, hazard response, and infrastructure digital twins. He directs a $7M NSF-funded national ecosystem for AI integration in civil engineering and leads professional education programs reaching practitioners worldwide.
Educational Qualifications
Ph.D., University of Cambridge, 2015
M.S., Indian Institute of Technology Madras, 2010
B.E., Anna University, 2008
Select Awards & Honors
- CAREER Award – National Science Foundation (2024)
- Moncrief Grand Challenge Faculty Award – Oden Institute (2024)
- Ervin S. Perry Student Appreciation Award – Fariborz Maseeh Department of Civil, Architectural and Environmental Engineering (2023)
- Dean’s Award for Outstanding Engineering Teaching by an Assistant Professor – The University of Texas at Austin (2022-23)
Related Websites
Geoelements Research Group
Oden Institute for Computational Engineering & Sciences
Geotechnical Research Group
Select Publications