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

Safe and efficient path planning is crucial for autonomous mobile robots. Aprerequisite for path planning is to have a comprehensive understanding of the3D structure of the robot's environment. On MAVs this is commonly achievedusing low-cost sensors, such as stereo or RGB-D cameras. These sensors may failto provide depth measurements in textureless or IR-absorbing areas and havelimited effective range. In path planning, this results in inefficienttrajectories or failure to recognize a feasible path to the goal, hencesignificantly impairing the robot's mobility. Recent advances in deep learningenables us to exploit prior experience about the shape of the world and henceto infer complete depth maps from color images and additional sparse depthmeasurements. In this work, we present an augmented planning system andinvestigate the effects of employing state-of-the-art depth completiontechniques, specifically trained to augment sparse depth maps originating fromRGB-D sensors, semi-dense methods and stereo matchers. We extensively evaluateour approach in online path planning experiments based on simulated data, aswell as global path planning experiments based on real world MAV data. We showthat our augmented system, provided with only sparse depth perception, canreach on-par performance to ground truth depth input in simulated onlineplanning experiments. On real world MAV data the augmented system demonstratessuperior performance compared to a planner based on very dense RGB-D depthmaps.

Predicting Unobserved Space For Planning via Depth Map Augmentation