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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this paper, we propose the problem of collaborative perception, whererobots can combine their local observations with those of neighboring agents ina learnable way to improve accuracy on a perception task. Unlike existing workin robotics and multi-agent reinforcement learning, we formulate the problem asone where learned information must be shared across a set of agents in abandwidth-sensitive manner to optimize for scene understanding tasks such assemantic segmentation. Inspired by networking communication protocols, wepropose a multi-stage handshake communication mechanism where the neuralnetwork can learn to compress relevant information needed for each stage.Specifically, a target agent with degraded sensor data sends a compressedrequest, the other agents respond with matching scores, and the target agentdetermines who to connect with (i.e., receive information from). Weadditionally develop the AirSim-CP dataset and metrics based on the AirSimsimulator where a group of aerial robots perceive diverse landscapes, such asroads, grasslands, buildings, etc. We show that for the semantic segmentationtask, our handshake communication method significantly improves accuracy byapproximately 20% over decentralized baselines, and is comparable tocentralized ones using a quarter of the bandwidth.

Who2com: Collaborative Perception via Learnable Handshake Communication