Postdoctoral Researcher · University of São Paulo
Ricardo V. Godoy
Building Physical AI — robots that are intuitive to control, safe around people, and improved through experience.
Research Framework
Physical AI as a Closed Loop
Three interconnected pillars that reinforce each other: better embodiment provides better demonstrations, better control enables safe execution, and learning turns data into scalable skills.
🦾
Capturing Intent
Robot Embodiment for AI
EMG, LMG, MoCap & vision interfaces that decode high-fidelity human intent for natural robot control.
🎯
Shared Control & Safety
AI Embodiment Control
Shared-autonomy frameworks translating intent into safe, collision-free robot actions in real time.
🧠
Learning & Adaptation
AI Brain for Physical World
Learning policies that generalize across objects, tasks, and platforms — bridging sim and reality.
Embodiment → demonstrations →
Control → safe execution →
Learning → scalable skills →
Embodiment
Impact
Research Highlights
🏆
Best Paper Award — IEEE LARS 2025
📄
38+ publications in RA-L, TNSRE, ICRA, IROS, Scientific Reports
🤝
Collaborations with Boston Dynamics, ANYbotics, ETH Zurich, and others
🎓
Co-supervising 3 PhD, 1 MSc, 6 BSc students
🏫
Assistant Professor — Albert Einstein Hospital (2025)
🔧
Petrobras R&D Project — Technical Leader (team of 30–50 researchers)
Selected Work
Featured Publications
38+ papers in top robotics venues. Showing selected recent work.
2026
Descriptor: Parasitoid Wasps and Associated Hymenoptera Dataset (DAPWH)
2026
Combining Augmented Reality with Semi-autonomous, Lightmyography Based Control to Improve Usability of Prostheses
2026
Automated identification of Ichneumonoidea wasps via YOLO-based deep learning: Integrating HiresCam for Explainable AI
2026
A Leaf-Level Dataset for Soybean-Cotton Detection and Segmentation
2026
Viewpoint-Agnostic Grasp Pipeline using VLM and Partial Observations
2026
From Perception to Assistance: VLM-Driven Potential Fields for Shared-Control Manipulation
Demos
Robots in Action
Key experiments and demonstrations from recent work.
