Javier Felip

About

Javier Felip is a roboticist whose research focuses on enabling robots to grasp and manipulate objects under real-world uncertainty. His core contributions lie at the intersection of sensor-based control, perception under incomplete observation, and human-robot collaboration. Felip’s most influential work, “Mind the gap - robotic grasping under incomplete observation” (109 citations), tackles the fundamental challenge of picking up unknown objects from a tabletop when the robot’s knowledge of the world is partial. He introduced a novel approach to object shape prediction that “closes the knowledge gaps” in the robot’s understanding, allowing for successful grasping despite missing sensory data. His earlier work on robust sensor-based grasp primitives (72 citations) developed a controller that uses feedback from contact-based sensors to handle uncertainty during grasping with a three-finger hand. Felip also contributed to multi-sensor fusion for contact detection and localization, and to the design of Tombatossals, the UJI humanoid torso, a platform for reproducible research in autonomous sensor-based tasks. More recently, he has explored intuitive human-robot collaboration using real-time approximate Bayesian inference. With over 200 total citations, Felip’s work has advanced the practical deployment of robots in unstructured environments, making him a key figure in sensor-based manipulation.

Research Focus

Key Achievements

Top Papers

1. 1
   Mind the gap - robotic grasping under incomplete observation

   109 citations · 2011
2. 2
   Robust sensor-based grasp primitive for a three-finger robot hand

   72 citations · 2009
3. 3
   Sensors and Methods for the Evaluation of Grasping

   10 citations · 2012
4. 4
   Multi-sensor and prediction fusion for contact detection and localization

   10 citations · 2014
5. 5
   Embodiment independent manipulation through action abstraction

   7 citations · 2010
6. 6
   Simulation of robot dynamics for grasping and manipulation tasks

   6 citations · 2012
7. 7
   Tombatossals: A humanoid torso for autonomous sensor-based tasks

   6 citations · 2015
8. 8
   Intuitive & Efficient Human-robot Collaboration via Real-time Approximate Bayesian Inference

   5 citations · 2022
9. 9
   Contact localization through robot and object motion from point clouds

   3 citations · 2013

Key Collaborators

Contact & Links