Heni Ben Amor

About

Heni Ben Amor is a prominent roboticist whose research sits at the intersection of human-robot interaction, machine learning, and collaborative robotics. His work has fundamentally advanced how robots perceive, interpret, and respond to human partners during physical and cooperative tasks. Most notably, Ben Amor pioneered the framework of **Interaction Primitives** — a probabilistic approach enabling robots to learn, generalize, and adapt collaborative behaviors from human demonstrations — a contribution that has garnered over 200 citations and spawned significant follow-on research. Complementing this, his Intention-Driven Dynamics Model (166 citations) established a principled probabilistic method for robots to infer human intentions in real time, a critical capability for safe and efficient collaboration. Ben Amor has also contributed meaningfully to assistive and rehabilitation robotics (132 citations), exploring how adaptive robot systems can support aging populations and motor recovery. His more recent involvement in the large-scale **Open X-Embodiment** initiative (over 200 combined citations) reflects his engagement with foundation models for generalist robot learning across diverse platforms. Collectively, his portfolio — exceeding 1,280 citations across these works alone — marks him as a leading voice in building robots that are genuinely responsive, adaptive, and safe alongside human partners.

Research Focus

Key Achievements

Top Papers

1. 1
   Interaction primitives for human-robot cooperation tasks

   200 citations · 2014
2. 2
   Probabilistic movement modeling for intention inference in human–robot interaction

   166 citations · 2013
3. 3
   Physical Human-Robot Interaction: Mutual Learning and Adaptation

   132 citations · 2012
4. 4
   A Human–Robot Interaction Perspective on Assistive and Rehabilitation Robotics

   132 citations · 2017
5. 5
   Projecting robot intentions into human environments

   125 citations · 2016
6. 6
   Estimation of perturbations in robotic behavior using dynamic mode decomposition

   124 citations · 2015
7. 7
   Open X-Embodiment: Robotic Learning Datasets and RT-X Models : Open X-Embodiment Collaboration⁰

   119 citations · 2024
8. 8
   Open X-Embodiment: Robotic Learning Datasets and RT-X Models

   101 citations · 2023
9. 9
   Learning multiple collaborative tasks with a mixture of Interaction Primitives

   96 citations · 2015
10. 10
    Learning interaction for collaborative tasks with probabilistic movement primitives

    92 citations · 2014

Key Collaborators

Contact & Links