Benjamin Alt

Senior Team Lead Research at ArtiMinds Robotics //
Doctoral Candidate at the Institute for Artificial Intelligence at the University of Bremen

In my research, I teach industrial robots how to reason about their tasks, generate their own programs and optimize and refine them as they gain more experience - and how to deploy advanced AI models to solve real-world engineering problems. I lead the corporate research lab at ArtiMinds, collaborating with academic and industrial partners to bring about a future in which humans can intuitively program robots and ensure that their behavior is both task-oriented and safe.

My areas of interest include robot self-learning, unsupervised and meta-learning, differentiable programming, mechanistic interpretability and AI safety.

latest posts

selected publications

  1. RoboGrind: Intuitive and Interactive Surface Treatment with Industrial Robots
    Benjamin Alt, Florian Stöckl, Silvan Müller, and 10 more authors
    In 2024 IEEE International Conference on Robotics and Automation (ICRA), May 2024
  2. Robot Program Parameter Inference via Differentiable Shadow Program Inversion
    Benjamin Alt, Darko Katic, Rainer Jäkel, and 2 more authors
    In 2021 IEEE International Conference on Robotics and Automation (ICRA), May 2021
  3. Shadow Program Inversion with Differentiable Planning: A Framework for Unified Robot Program Parameter and Trajectory Optimization
    Benjamin Alt, Claudius Kienle, Darko Katic, and 2 more authors
    Sep 2024