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 target-driven navigation system to improve mapless visualnavigation in indoor scenes. Our method takes a multi-view observation of arobot and a target as inputs at each time step to provide a sequence of actionsthat move the robot to the target without relying on odometry or GPS atruntime. The system is learned by optimizing a combinational objectiveencompassing three key designs. First, we propose that an agent conceives thenext observation before making an action decision. This is achieved by learninga variational generative module from expert demonstrations. We then proposepredicting static collision in advance, as an auxiliary task to improve safetyduring navigation. Moreover, to alleviate the training data imbalance problemof termination action prediction, we also introduce a target checking module todifferentiate from augmenting navigation policy with a termination action. Thethree proposed designs all contribute to the improved training data efficiency,static collision avoidance, and navigation generalization performance,resulting in a novel target-driven mapless navigation system. Throughexperiments on a TurtleBot, we provide evidence that our model can beintegrated into a robotic system and navigate in the real world. Videos andmodels can be found in the supplementary material.

Towards Target-Driven Visual Navigation in Indoor Scenes via Generative Imitation Learning