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/

Zendy Plus

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

Zendy Tools

Zendy Open

Simultaneous Localisation and Mapping (SLAM) algorithms are expensive to runon smaller robotic platforms such as Micro-Aerial Vehicles. Bug algorithms arean alternative that use relatively little processing power, and avoid highmemory consumption by not building an explicit map of the environment. BugAlgorithms achieve relatively good performance in simulated and robotic mazesolving domains. However, because they are hand-designed, a natural question iswhether they are globally optimal control policies. In this work we explore theperformance of Neuroevolution - specifically NEAT - at evolving controlpolicies for simulated differential drive robots carrying out generalised mazenavigation. We extend NEAT to include Gated Recurrent Units (GRUs) to help dealwith long term dependencies. We show that both NEAT and our NEAT-GRU canrepeatably generate controllers that outperform I-Bug (an algorithmparticularly well-suited for use in real robots) on a test set of 209 indoormaze like environments. We show that NEAT-GRU is superior to NEAT in this taskbut also that out of the 2 systems, only NEAT-GRU can continuously evolvesuccessful controllers for a much harder task in which no bearing informationabout the target is provided to the agent.

Evolving Indoor Navigational Strategies Using Gated Recurrent Units In NEAT