Player-Stage Based Simulator For Simultaneous Multi-Robot Exploration And Terrain Coverage Problem

Abstract

One of the possible ways of offering assistance without risking additional human lives during hazardous situations is by deploying a robot team, equipped with various sensors and actuators. Working with intelligent robotics requires a large investment in both money and time. There is a general purpose, open source simulator called Player/Stage, which provides a hardware abstraction layer to several popular robot platforms, and is commonly used by robotics community in research and university teaching, today. This simulator tends to be very simple and task-specific. Player, which is a distributed device server for robots, sensors and actuators, can control either a real or simulated robot thus allowing direct application of developed algorithms to real-life scenarios. Hence, we believe that Player/Stage, when coupled with robust hardware, is a viable paradigm for our Simultaneous MSTC(S-MSTC) algorithm. This paper gives details of our experience in running Player/Stage during the implementation of our online S-MSTC algorithm for multi-robot being implemented in C++ and we use Player/Stage middleware for validation and testing. In addition, the experience with Player/Stage can help us to do research in more complicated situations.

Pemanfaatan Middleware Robot Operating System (Ros) Dalam Menjawab Tantangan Revolusi Industri 4.0

“…Middleware ini terletak di antara sistem operasi dan perangkat lunak aplikasi [18], [19] serta berfungsi sebagai penghubung antara perangkat keras sensor, aktuator, dan komponen perangkat lunak [19]. Beberapa middleware telah dikembangkan dan digunakan oleh para peneliti untuk mengontrol sistem robotika seperti Player/Stage [20], Miro, MRDS, ASEBA, Orca, Pyro, ROS, dan ROS2 [21], [22]. Pada penelitian ini akan memanfaatkan middleware Robot Operating System (ROS) dan Raspberry Pi untuk mengontrol on/off perangkat elektronik yang dapat diaplikasikan pada teknik pengontrolan perangkat elektronik industri.…”

Comparison Between Genetic Fuzzy Methodology and Q-Learning for Collaborative Control Design

“…There has been a lot of research conducted in the field of collaborative robotics. These include development of controlled physical compliance for external contacts [1][2][3][4][5] that could be useful for human-robot collaborative tasks, swarm intelligence control algorithms [6][7][8][9] as well as multi-robot collaboration with minimal communication between the robots [10,11].Such intelligent collaborative robots can help in various applications such as material handling [12], mapping the interior of buildings [13], exploration [14], factory automation [15] etc., to name a few. This research focuses on a different type of problem where a team of independently controlled robots work together to achieve a common goal while they are also physically connected to an object through elastic cables.…”