English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Overactuated omnidirectional flying vehicles are capable of generating forceand torque in any direction, which is important for applications such ascontact-based industrial inspection. This comes at the price of an increase inmodel complexity. These vehicles usually have non-negligible, repetitivedynamics that are hard to model, such as the aerodynamic interference betweenthe propellers. This makes it difficult for high-performance trajectorytracking using a model-based controller. This paper presents an approach thatcombines a data-driven and a first-principle model for the system actuation anduses it to improve the controller. In a first step, the first-principle modelerrors are learned offline using a Gaussian Process (GP) regressor. At runtime,the first-principle model and the GP regressor are used jointly to obtaincontrol commands. This is formulated as an optimization problem, which avoidsambiguous solutions present in a standard inverse model in overactuatedsystems, by only using forward models. The approach is validated using atilt-arm overactuated omnidirectional flying vehicle performing attitudetrajectory tracking. The results show that with our proposed method, theattitude trajectory error is reduced by 32% on average as compared to a nominalPID controller.

Learning dynamics for improving control of overactuated flying systems