Saviz Mowlavi

Mitsubishi Electric (United States)

Papers

1

Total Citations

9

H-Index

1

About

Saviz Mowlavi is a robotics researcher whose work centers on state estimation and sensor fusion for legged robots, with a particular focus on bridging the gap between classical control theory and modern deep learning. His most notable contribution, the OptiState framework, introduces a hybrid approach that integrates Kalman filtering with gated networks and transformer-based vision, enabling more robust state estimation under the highly dynamic and sensor-limited conditions typical of legged locomotion. This work, published in 2024, has already garnered 9 citations, reflecting its timely relevance to the field. Mowlavi’s research addresses a critical challenge: how to effectively combine proprioceptive and exteroceptive data to maintain accurate robot state awareness during fast, agile movement. By fusing optimization-based methods with learned modalities, he offers a practical pathway toward more resilient and perceptive robotic systems. His contributions are particularly valuable for researchers working on real-world deployment of legged robots in unstructured environments, where sensor noise and motion dynamics pose persistent obstacles. Mowlavi’s work exemplifies the growing trend of hybrid architectures that leverage the strengths of both classical and learning-based approaches.

Research Focus

Key Achievements

1
H-Index
1
Papers
9
Total Citations
9
Avg Citations/Paper
🏆 Most Cited Paper
OptiState: State Estimation of Legged Robots using Gated Networks with Transformer-based Vision and Kalman Filtering
9 citations · 2024
📈 Most Prolific Year: 2024 (1 Papers)
🤝 Key Collaborators: 5
🏛 Institutions: Mitsubishi Electric (United States)

Top Papers

  1. 1

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago