V. G. Berman

Skolkovo Institute of Science and Technology

Papers

3

Total Citations

24

H-Index

3

About

V. G. Berman is at the forefront of integrating large language and multimodal models into embodied robotic systems, with a primary focus on quadruped robots for assistive and interactive applications. Their most impactful work, "CognitiveDog" (2024, 16 citations), pioneers the use of Large Multimodal Models (LMMs) to enable a quadruped robot to not only converse with humans but also physically manipulate objects in its environment, representing a significant leap in human-robot interaction. Berman further advances assistive robotics with "DogSurf" (2024, 5 citations), which employs a GRU-based surface recognition system to help visually impaired individuals navigate safely by detecting slippery terrain and providing haptic feedback. In "LLM-MARS" (2023, 3 citations), they introduced a novel framework for multi-agent robot systems that leverages LLMs for dynamic dialogue and behavior tree generation, allowing robots to interpret operator commands and generate context-aware actions. Berman’s work uniquely bridges natural language understanding, computer vision, and physical robot control, establishing a new paradigm for intelligent, communicative, and helpful robotic assistants.

Research Focus

Key Achievements

3
H-Index
3
Papers
24
Total Citations
8
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: 2024 (2 Papers)
🤝 Key Collaborators: 14
🏛 Institutions: Skolkovo Institute of Science and Technology

Top Papers

  1. 1
  2. 2
  3. 3

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago