MiG

A Realistic Simulator for Humanoid Soccer Robot Using Particle Filter

Raúl Rojas, Fu Yao, Hamid Mobalegh, Longxu Jin, Miao Wang— 2013

This work presents a realistic simulator called Reality Sim for humanoid soccer robots especially in simulation of computer vision. As virtual training, testing and evaluating environment, simulation platforms have become one significant component in Soccer Robot projects. Nevertheless, the simulated environment in a simulation platform usually has a big gap with the realistic world. In order to solve this issue, we demonstrate a more realistic simulation system which is called Reality Sim with numerous real images. With this system, the computer vision code could be easily tested on the simulation platform. For this purpose, an image database with a large quantity of images recorded in various camera poses is built. Furthermore, if the camera pose of an image is not included in the database, an interpolation algorithm is used to reconstruct a brand-new realistic image of that pose such that a realistic image could be provided on every robot camera pose. Systematic empirical results illustrate the efficiency of the approach while it effectively simulates a more realistic environment for simulation so that it satisfies the requirement of humanoid soccer robot projects.

TitelA Realistic Simulator for Humanoid Soccer Robot Using Particle Filter
VerfasserRaúl Rojas, Fu Yao, Hamid Mobalegh, Longxu Jin, Miao Wang
VerlagSpringer-Verlag, Heidelberg
Datum2013
Quelle/n
Erschienen inRecent Advances in Robotics and Automation, Studies in Computational Intelligence, Vol. 480.
Größe oder Längepp. 275-288
RechteCopyright by Springer. When citing this work cite the original article.