OdeMHPlanner provides a framework for uncertainty-aware learning and planning in dynamical systems with unknown dynamics and infrequent output measurements.

It implements the method described in:

Learning Dynamics from Infrequent Output Measurements for Uncertainty-Aware Optimal Control Robert Lefringhausen, Theodor Springer, Sandra Hirche arXiv:2512.08013 (2025) https://arxiv.org/abs/2512.08013

Overview

The package targets control problems in which the system dynamics are unknown and the state is only partially observed through infrequent output measurements. Instead of relying on a single identified model, OdeMHPlanner explicitly represents uncertainty over both the dynamics and the latent state trajectory.

The approach combines:

1. Bayesian learning via a Metropolis–Hastings (MH) sampler equipped with a numerical ODE solver to sample from the posterior distribution over dynamics and latent states.
2. Scenario-based optimal control, where posterior samples are propagated to compute control inputs that are robust to model uncertainty.

This enables principled uncertainty quantification and safer decision-making compared to point-estimate-based methods.

For detailed instructions and examples, please refer to the Documentation.

Installation

This package is not registered. Clone and instantiate it locally:

git clone https://github.com/TUM-ITR/ode-mh-planner.git
cd ode-mh-planner
julia --project=. -e 'using Pkg; Pkg.instantiate()'

Solvers

The results reported in the paper were obtained using proprietary HSL linear solvers (in particular ma57) for improved numerical performance. HSL solvers are available from https://www.hsl.rl.ac.uk/ and can be obtained under a free academic license; installation instructions are provided on their website.

If HSL solvers are not available, the code automatically falls back to standard open-source solvers (e.g. MUMPS). While functional, this configuration may be slower and has not been exhaustively tested.

Citation

If you use this code in your research, please cite our paper:

@article{lefringhausen2025learning,
  title={Learning Dynamics from Infrequent Output Measurements for Uncertainty-Aware Optimal Control},
  author={Lefringhausen, Robert and Springer, Theodor and Hirche, Sandra},
  journal={arXiv preprint arXiv:2512.08013},
  year={2025}
}