Papers

6

Total Citations

31

H-Index

3

About

Mikhail Litvinov is a robotics and artificial intelligence researcher whose work spans two decades of innovation at the intersection of human-robot interaction, multimodal systems, and generative AI. His early research, dating to the mid-2000s, explored virtual reality environments and intelligent man-machine interfaces for assistive medical systems, establishing a foundational interest in making technology accessible and responsive to human needs. In recent years, Litvinov has emerged as a pioneering contributor to the application of Large Multimodal Models (LMMs) in robotics. His most celebrated work, **CognitiveDog** (2024, 16 citations), introduced a quadruped robot capable of natural verbal communication and physical object manipulation using the Unitree Go1 platform — a landmark achievement in embodied AI. Building on this, his **CognitiveOS** framework proposes the first operating system designed to equip any robotic platform with cognitive, generative AI capabilities through a transformer-based multi-agent architecture. His **LLM-MARS** system further advanced natural language-driven multi-robot coordination, while **SwarmVLM** extended these principles to heterogeneous UAV-AGV warehouse systems. Litvinov's cumulative contributions reflect a coherent vision: bridging language, perception, and physical action to create robots that genuinely understand and interact with the human world.

Research Focus

Key Achievements

3
H-Index
6
Papers
31
Total Citations
5
Avg Citations/Paper
🏆 Most Cited Paper
CognitiveDog: Large Multimodal Model Based System to Translate Vision and Language into Action of Quadruped Robot
16 citations · 2024
📈 Most Prolific Year: 2025 (2 Papers)
🤝 Key Collaborators: 29
🏛 Institutions: Skolkovo Institute of Science and Technology, Baltic State Technical University Voenmeh

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago