Um ambiente de simulação de missões indoor para robôs terrestres de pneus diferenciais

Abstract

Nowdays, there are applications for mobile robots for the most diverse scenarios. In many cases, these applications involve indoor activities, where obstacles constantly change. For the robot to be able to navigate without getting lost or suffering any damage, it needs to knows where it is and where it should go. For this, a simulation environment was developed with a focus on grounded differential-drive robots. Among the implemented features are the simulation of reading from a distance rangefinder sensor to detect obstacles close to the robot. The use of these simulated readings to build a Local Map, and the navigation of the robot from one point to another on the map, avoiding collisions. The system was implemented in a modular way to allow the analysis of routes generated by different Path Planning algorithms. The analyzes consider metrics to evaluate the quality of the generated routes based on the performance of the robot in the navigation task. Finally, a comparison of the results in scenarios with different levels of complexity was presented, showing which algorithms did better in the tested scenarios.