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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Animal and robot social interactions are interesting both for ethologicalstudies and robotics. On the one hand, the robots can be tools and models toanalyse animal collective behaviours, on the other hand, the robots and theirartificial intelligence are directly confronted and compared to the naturalanimal collective intelligence. The first step is to design robots and theirbehavioural controllers that are capable of socially interact with animals.Designing such behavioural bio-mimetic controllers remains an importantchallenge as they have to reproduce the animal behaviours and have to becalibrated on experimental data. Most animal collective behavioural models aredesigned by modellers based on experimental data. This process is long andcostly because it is difficult to identify the relevant behavioural featuresthat are then used as a priori knowledge in model building. Here, we want tomodel the fish individual and collective behaviours in order to develop robotcontrollers. We explore the use of optimised black-box models based onartificial neural networks (ANN) to model fish behaviours. While the ANN maynot be biomimetic but rather bio-inspired, they can be used to link perceptionto motor responses. These models are designed to be implementable as robotcontrollers to form mixed-groups of fish and robots, using few a prioriknowledge of the fish behaviours. We present a methodology with multilayerperceptron or echo state networks that are optimised through evolutionaryalgorithms to model accurately the fish individual and collective behaviours ina bounded rectangular arena. We assess the biomimetism of the generated modelsand compare them to the fish experimental behaviours.

Evolutionary optimisation of neural network models for fish collective behaviours in mixed groups of robots and zebrafish