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 work explores the feasibility of steering a drone with a (recurrent)neural network, based on input from a forward looking camera, in the context ofa high-level navigation task. We set up a generic framework for training anetwork to perform navigation tasks based on imitation learning. It can beapplied to both aerial and land vehicles. As a proof of concept we apply it toa UAV (Unmanned Aerial Vehicle) in a simulated environment, learning to cross aroom containing a number of obstacles. So far only feedforward neural networks(FNNs) have been used to train UAV control. To cope with more complex tasks, wepropose the use of recurrent neural networks (RNN) instead and successfullytrain an LSTM (Long-Short Term Memory) network for controlling UAVs. Visionbased control is a sequential prediction problem, known for its highlycorrelated input data. The correlation makes training a network hard,especially an RNN. To overcome this issue, we investigate an alternativesampling method during training, namely window-wise truncated backpropagationthrough time (WW-TBPTT). Further, end-to-end training requires a lot of datawhich often is not available. Therefore, we compare the performance ofretraining only the Fully Connected (FC) and LSTM control layers with networkswhich are trained end-to-end. Performing the relatively simple task of crossinga room already reveals important guidelines and good practices for trainingneural control networks. Different visualizations help to explain the behaviorlearned.

How hard is it to cross the room? -- Training (Recurrent) Neural Networks to steer a UAV