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 describe a policy learning approach to map visual inputs to drivingcontrols conditioned on turning command that leverages side tasks on semanticsand object affordances via a learned representation trained for driving. Tolearn this representation, we train a squeeze network to drive usingannotations for the side task as input. This representation encodes thedriving-relevant information associated with the side task while ideallythrowing out side task-relevant but driving-irrelevant nuisances. We then traina mimic network to drive using only images as input and use the squeezenetwork's latent representation to supervise the mimic network via a mimickingloss. Notably, we do not aim to achieve the side task nor to learn features forit; instead, we aim to learn, via the mimicking loss, a representation of theside task annotations directly useful for driving. We test our approach usingthe CARLA simulator. In addition, we introduce a more challenging but realisticevaluation protocol that considers a run that reaches the destinationsuccessful only if it does not violate common traffic rules. A videosummarizing this work is available at https://youtu.be/ipKAMzmJpMs , and codeis available at https://github.com/twsq/sam-driving .

SAM: Squeeze-and-Mimic Networks for Conditional Visual Driving Policy Learning