Trajectory-control using deep system identification and model predictive control for drone control under uncertain load

Abstract : Machine learning allows to create complex models if provided with enough data, hence challenging more traditional system identification methods. We compare the quality of neural networks and an ARX model when use in an model predictive control to command a drone in a simulated environment. The training of neural networks can be challenging when the data is scarce or datasets are unbalanced. We propose an adaptation of prioritized replay to system identification in order to mitigate these problems. We illustrate the advantages and limits of this training method on the control task of a simulated drone.
Complete list of metadatas

Cited literature [17 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01927035
Contributor : Antoine Mahé <>
Submitted on : Monday, November 19, 2018 - 3:41:08 PM
Last modification on : Wednesday, July 31, 2019 - 4:18:03 PM
Long-term archiving on : Wednesday, February 20, 2019 - 3:15:35 PM

File

root.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01927035, version 1

Citation

Antoine Mahé, Cédric Pradalier, Matthieu Geist. Trajectory-control using deep system identification and model predictive control for drone control under uncertain load. 22nd International Conference on System Theory, Control and Computing, Oct 2018, Sinaia, Romania. ⟨hal-01927035⟩

Share

Metrics

Record views

64

Files downloads

3