A Framework for Visually Realistic Multi-robot Simulation in Natural Environment

Abstract

This paper presents a generalized framework for the simulation of multiple\nrobots and drones in highly realistic models of natural environments. The\nproposed simulation architecture uses the Unreal Engine4 for generating both\noptical and depth sensor outputs from any position and orientation within the\nenvironment and provides several key domain specific simulation capabilities.\nVarious components and functionalities of the system have been discussed in\ndetail. The simulation engine also allows users to test and validate a wide\nrange of computer vision algorithms involving different drone configurations\nunder many types of environmental effects such as wind gusts. The paper\ndemonstrates the effectiveness of the system by giving experimental results for\na test scenario where one drone tracks the simulated motion of another in a\ncomplex natural environment.\n