A.H. Khalil

Papers

Total Citations

H-Index

About

A.H. Khalil is a robotics researcher focused on bridging the gap between advanced control theory and practical real-time deployment. Their primary research area lies in Model Predictive Control (MPC), specifically addressing the critical computational bottleneck that prevents MPC from being widely used in fast, dynamic robotic systems. Khalil’s major contribution is the development of **TransformerMPC**, a novel framework that leverages the attention mechanism of transformer neural networks to dramatically accelerate MPC computation. By learning the optimal control policy from data, TransformerMPC bypasses the need for solving complex optimization problems at every time step, enabling real-time control on resource-constrained hardware. This work has garnered early recognition with 3 citations, signaling its potential to influence both the control and machine learning communities. Khalil’s research represents a significant step toward making sophisticated, optimization-based control practical for agile robots, autonomous vehicles, and other latency-critical applications. Their work stands out for its elegant fusion of deep learning and classical control, offering a promising path toward more intelligent and responsive autonomous systems.