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

Imitation learning (IL) can train computationally-efficient sensorimotorpolicies from a resource-intensive Model Predictive Controller (MPC), but itoften requires many samples, leading to long training times or limitedrobustness. To address these issues, we combine IL with a variant of robust MPCthat accounts for process and sensing uncertainties, and we design a dataaugmentation (DA) strategy that enables efficient learning of vision-basedpolicies. The proposed DA method, named Tube-NeRF, leverages Neural RadianceFields (NeRFs) to generate novel synthetic images, and uses properties of therobust MPC (the tube) to select relevant views and to efficiently compute thecorresponding actions. We tailor our approach to the task of localization andtrajectory tracking on a multirotor, by learning a visuomotor policy thatgenerates control actions using images from the onboard camera as only sourceof horizontal position. Our evaluations numerically demonstrate learning of arobust visuomotor policy with an 80-fold increase in demonstration efficiencyand a 50% reduction in training time over current IL methods. Additionally, ourpolicies successfully transfer to a real multirotor, achieving accuratelocalization and low tracking errors despite large disturbances, with anonboard inference time of only 1.5 ms.

Tube-NeRF: Efficient Imitation Learning of Visuomotor Policies from MPC using Tube-Guided Data Augmentation and NeRFs