English

Learning Observation Models with Incremental Non-Differentiable Graph Optimizers in the Loop for Robotics State Estimation

Robotics 2023-09-07 v1

Abstract

We consider the problem of learning observation models for robot state estimation with incremental non-differentiable optimizers in the loop. Convergence to the correct belief over the robot state is heavily dependent on a proper tuning of observation models which serve as input to the optimizer. We propose a gradient-based learning method which converges much quicker to model estimates that lead to solutions of much better quality compared to an existing state-of-the-art method as measured by the tracking accuracy over unseen robot test trajectories.

Keywords

Cite

@article{arxiv.2309.02525,
  title  = {Learning Observation Models with Incremental Non-Differentiable Graph Optimizers in the Loop for Robotics State Estimation},
  author = {Mohamad Qadri and Michael Kaess},
  journal= {arXiv preprint arXiv:2309.02525},
  year   = {2023}
}

Comments

6 pages, 4 figures. Published at the Differentiable Almost Everything Workshop of the 40th International Conference on Machine Learning

R2 v1 2026-06-28T12:13:34.767Z