Tejas Dahiya

Learned relationships from healthy sensor data enable highly accurate diagnosis and significant fuel correction, reducing CO2 emissions accounting error by over 80% in heavy-duty diesel vehicles.

Retaining molecular anisotropy can dramatically enhance prediction accuracy in coarse-grained simulations, revealing that geometry matters more than previously thought.