Anil Aswani

University of California, Berkeley

Papers

3

Total Citations

185

H-Index

3

About

Anil Aswani’s research lies at the intersection of control theory, machine learning, and robotics, with a focus on creating safe, data-driven systems that can learn and adapt in real time. He is best known for pioneering **learning-based model predictive control (LBMPC)** , a framework that rigorously integrates statistical learning with control engineering while preserving guarantees on safety, robustness, and convergence. His landmark 2012 paper on the onboard implementation of LBMPC on a quadrotor helicopter (174 citations) demonstrated how such algorithms can operate under real-time constraints, bridging theory and practice in autonomous systems. Aswani has also advanced **robot imitation learning**, notably through work on dynamic regret convergence and adaptive regularization for on-policy methods like DAgger (2021), addressing fundamental questions about how robots can efficiently learn from human supervisors without catastrophic performance swings. Beyond these contributions, his comparative studies of optimization algorithms for learning-based MPC (2014) provide practical guidance for deploying these methods in robotics and automation. With a career marked by rigorous theoretical foundations and impactful experimental validation, Aswani’s work continues to shape how autonomous systems learn from data while maintaining the safety and reliability essential for real-world deployment.

Research Focus

Key Achievements

3
H-Index
3
Papers
185
Total Citations
62
Avg Citations/Paper
🏆 Most Cited Paper
Learning-based model predictive control on a quadrotor: Onboard implementation and experimental results
174 citations · 2012
📈 Most Prolific Year: 2012 (1 Papers)
🤝 Key Collaborators: 7
🏛 Institutions: University of California, Berkeley

Top Papers

  1. 1
  2. 2
  3. 3

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 12 days ago