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

We present a reward-predictive, model-based deep learning method featuringtrajectory-constrained visual attention for local planning in visual navigationtasks. Our method learns to place visual attention at locations in latent imagespace which follow trajectories caused by vehicle control actions to enhancepredictive accuracy during planning. The attention model is jointly optimizedby the task-specific loss and an additional trajectory-constraint loss,allowing adaptability yet encouraging a regularized structure for improvedgeneralization and reliability. Importantly, visual attention is applied inlatent feature map space instead of raw image space to promote efficientplanning. We validated our model in visual navigation tasks of planning lowturbulence, collision-free trajectories in off-road settings and hill climbingwith locking differentials in the presence of slippery terrain. Experimentsinvolved randomized procedural generated simulation and real-worldenvironments. We found our method improved generalization and learningefficiency when compared to no-attention and self-attention alternatives.

Trajectory-Constrained Deep Latent Visual Attention for Improved Local Planning in Presence of Heterogeneous Terrain