Decision Behavioral Approach for Self Organizing Multi Agent Robots Based on Deep Neural Network

Abstract

Swarm interacts locally without any centralized control to formulate the predefined shape. The algorithm is based on bio inspired behaviors occur in animal flocks which means the algorithm depends on the current flocks distribution and the predefined shape is detected through a main controller to estimate the shape. Shapes are previously trained through a deep neural network on the controller to detect the geometric shape. Deep Neural network’s input is a given current robots distributions in the map after eliminating un potential pixels in the map according to obstacles and map borders. Simulation based test are done to validate self-organizing algorithm approaching square pattern formation test with varying numbers of robots.