Luc Jaulin

A reliable quantification of the worst-case influence of model uncertainty and external disturbances is crucial for the localization of vessels in marine applications. This is especially true if uncertain GPS-based position measurements are used to update predicted vessel locations that are obtained from the evaluation of a ship’s state equation. To reflect real-life working conditions, these state equations need to account for uncertainty in the system model, such as imperfect actuation and external disturbances due to effects such as wind and currents. As an application scenario, the GPS-based localization of autonomous DDboat robots is considered in this paper. Using experimental data, the efficiency of an ellipsoidal approach, which exploits a bounded-error representation of disturbances and uncertainties, is demonstrated.

@article{rauh-gourret-lagattu-hummes-jaulin-reuter-wirtensohn-hoher-2022.,
  title = {Experimental Validation of Ellipsoidal Techniques for State Estimation in Marine Applications},
  author = {Rauh, Andreas and Gourret, Yohann and Lagattu, Katell and Hummes, Bernardo and Jaulin, Luc and Reuter, Johannes and Wirtensohn, Stefan and Hoher, Patrick}},
  year = {2022},
  journal = {Algorithms},
  volume = {15},
  number = {5},
  article-number = {162},
}

For a mobile robot to operate in its environment it is crucial to determine its position with respect to an external reference frame using noisy sensor readings. A scenario in which the robot is moved to another position during its operation without being told, known as the kidnapped robot problem, complicates global localisation. In addition to that, sensor malfunction and external influences of the environment can cause unexpected errors, called outliers, that negatively affect the localisation process. This paper proposes a method based on the fusion of a particle filter with bounded-error localisation, which is able to deal with outliers in the measurement data. The application of our algorithm to solve the kidnapped robot problem using simulated data shows an improvement over conventional probabilistic filtering methods.

@article{neuland-mantelli-hummes-jaulin-maffei-prestes-kolberg-2021,
  title = {Robust Hybrid Interval-Probabilistic Approach for the Kidnapped Robot Problem},
  author = {Neuland, Renata and Mantelli, Mathias and Hummes, Bernardo and Jaulin, Luc and Maffei, Renan and Prestes, Edson and Kolberg, Mariana},
  year = {2021},
  journal = {International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems},
  volume = {29},
  number = {02},
  pages = {313-331},
  month = february,
}