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

This paper presents a novel trajectory tracker for autonomous quadrotornavigation in dynamic and complex environments. The proposed frameworkintegrates a distributional Reinforcement Learning (RL) estimator for unknownaerodynamic effects into a Stochastic Model Predictive Controller (SMPC) fortrajectory tracking. Aerodynamic effects derived from drag forces and momentvariations are difficult to model directly and accurately. Most currentquadrotor tracking systems therefore treat them as simple `disturbances' inconventional control approaches. We propose Quantile-approximation-basedDistributional Reinforced-disturbance-estimator, an aerodynamic disturbanceestimator, to accurately identify disturbances, i.e., uncertainties between thetrue and estimated values of aerodynamic effects. Simplified Affine DisturbanceFeedback is employed for control parameterization to guarantee convexity, whichwe then integrate with a SMPC to achieve sufficient and non-conservativecontrol signals. We demonstrate our system to improve the cumulative trackingerrors by at least 66% with unknown and diverse aerodynamic forces comparedwith recent state-of-the-art. Concerning traditional Reinforcement Learning'snon-interpretability, we provide convergence and stability guarantees ofDistributional RL and SMPC, respectively, with non-zero mean disturbances.

Interpretable Stochastic Model Predictive Control using Distributional Reinforced Estimation for Quadrotor Tracking Systems