About

Pedro Vicente is a robotics researcher specializing in autonomous manipulation, learning from demonstration, and adaptive body perception for humanoid robots. His work focuses on enabling robots to perform complex household tasks, such as table-cleaning and grasping, through deep learning and kinesthetic teaching. Vicente’s major contributions include developing methods for online body schema adaptation—where robots use internal mental simulation combined with multisensory feedback to continuously update their own kinematic models—and markerless visual servoing for hand-eye coordination during reaching and grasping. His papers, including the highly cited “iCub, clean the table!” (29 citations), demonstrate practical applications of deep neural networks for service robotics. Vicente also explores advanced topics like haptic exploration for safe 3D grasping using Bayesian optimization and reinforcement learning for dry-stacking irregular rocks, a task with potential extraterrestrial applications. His work on GPU-enabled particle optimization for hand pose estimation and self-calibration further highlights his impact in real-time robotic perception. With over a decade of research, Vicente’s contributions are foundational for developing autonomous robots that can adapt and learn in unstructured environments.

Research Focus

Key Achievements

8
H-Index
12
Papers
167
Total Citations
14
Avg Citations/Paper
🏆 Most Cited Paper
“iCub, clean the table!” A robot learning from demonstration approach using deep neural networks
29 citations · 2018
📈 Most Prolific Year: 2018 (4 Papers)
🤝 Key Collaborators: 14
🏛 Institutions: University of Lisbon, Instituto Superior Técnico, Instituto de Engenharia de Sistemas e Computadores Microsistemas e Nanotecnologias

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago