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 focuses on inverse reinforcement learning for autonomousnavigation using distance and semantic category observations. The objective isto infer a cost function that explains demonstrated behavior while relying onlyon the expert's observations and state-control trajectory. We develop a mapencoder, that infers semantic category probabilities from the observationsequence, and a cost encoder, defined as a deep neural network over thesemantic features. Since the expert cost is not directly observable, the modelparameters can only be optimized by differentiating the error betweendemonstrated controls and a control policy computed from the cost estimate. Wepropose a new model of expert behavior that enables error minimization using aclosed-form subgradient computed only over a subset of promising states via amotion planning algorithm. Our approach allows generalizing the learnedbehavior to new environments with new spatial configurations of the semanticcategories. We analyze the different components of our model in a minigridenvironment. We also demonstrate that our approach learns to follow trafficrules in the autonomous driving CARLA simulator by relying on semanticobservations of buildings, sidewalks, and road lanes.

Inverse reinforcement learning for autonomous navigation via differentiable semantic mapping and planning