About

Abdeslam Boularias is a robotics and machine learning researcher whose work sits at the intersection of autonomous manipulation, reinforcement learning, and perception. His research has made substantial contributions to some of the most challenging problems in modern robotics: enabling robots to grasp and manipulate unknown objects in cluttered environments, learn from physical interaction, and interpret natural language commands for navigation. Boularias pioneered approaches that allow robots to learn entirely from scratch through trial and error, eliminating dependence on pre-built object models — a breakthrough demonstrated in his widely cited 2015 work on manipulation in clutter (103 citations). His 2017 self-supervised object detection system cleverly leveraged physics simulation and multi-view pose estimation to sidestep the burdensome data-labeling process that plagues deep learning applications in robotics (100 citations). His Deep Interaction Prediction Network (DIPN) further advanced the field by enabling robots to mentally "imagine" the consequences of pushing actions before execution. Beyond manipulation, Boularias has explored inverse reinforcement learning in dynamic settings like table tennis and developed compliant, vision-driven strategies for high-precision assembly. With a body of work spanning probabilistic learning, physics-based modeling, and language-grounded navigation, his research consistently pushes robots toward greater autonomy and adaptability in real-world environments.

Research Focus

Key Achievements

18
H-Index
60
Papers
1,047
Total Citations
17
Avg Citations/Paper
🏆 Most Cited Paper
Learning to Manipulate Unknown Objects in Clutter by Reinforcement
103 citations · 2015
📈 Most Prolific Year: 2020 (8 Papers)
🤝 Key Collaborators: 88
🏛 Institutions: Carnegie Mellon University, Rutgers, The State University of New Jersey, Max Planck Institute for Intelligent Systems, Max Planck Society, Max Planck Institute for Biological Cybernetics, Rutgers Sexual and Reproductive Health and Rights

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 · 33 days ago