RoCKInChallengeToulouse

Sponsors and Acknowledgments

First of all we would like to thank the Spanish Ministry of Economy and Competitiveness for the partial support to this work under grant DPI2013-40534-R.

We want to show our appreciation to these companies for their support to watermelon team:

RoCKIn Project Description

RoCKIn logo.

This challenge focuses on domestic service robots. The project aims to create robots with enhanced networking and cognitive abilities. They should be able to perform useful tasks such as helping the impaired and the elderly (one of the main goals of our group).

In the initial stages of the competition, individual robots will begin by overcoming basic individual tasks, such as navigating through the rooms of a house, manipulating objects or recognizing faces, and then coordinating to handle house-keeping tasks simultaneously, some of them under natural interaction with humans.

• Official Web Page

• Official Wiki

Watermelon Project: Team Description

• Project Codename

Watermelon Project

• Team Coordinator:

Vicente Matellán Olivera

• Team Members:

Technical- Manipulation/Grasping, Simulation: Fernando Casado
Technical- SW Integration, Middleware, 2D Perception: Francisco Martín Rico
Technical- SW Integration, HRI Dialogue, Team Leader: Francisco Lera
Technical- Hardware: Carlos Rodríguez
Technical- 3D Perception: Víctor Rodríguez 

• Other Information:

Academic Year: 2014-2015
Repositories: https://github.com/Robotica-ule/MYRABot
Tags: Robotics competitions, Feasible Assistive Robots
Technology: ROS, PCL, c++, svn, OpenCV, cmake, OpenGL, Qt, Aruco, 
State: Development

Project Summary

Our challenge is to create a feasible platform able to take part in Robotics Competitions. We are working for integrating different hardware solutions in a DIY robot.

Robot

We want to take part in RoCKIn with the platform developed during the last two years in the Catedra Telefónica-ULE.

MYRABot Competition Robot.

Robot Hardware

Component	Model	Description
Frame	n/a	Poplar laminated wood
Computer 1	Mountain Graphite 20	web
Computer 2	Mountain Prototype	Touch Display, 360º open, intel i5.
Controllers	(a)Arduino 2560, (b)USB2serial	(a)arm, range sensors, (b)create
Base	Create (iRobot)	web
Ultrasound Sensors	Maxsensor mb1220 (x5)	Range: 7 meters
RGB sensor	Logitech	Webcam
RGBD sensors	Kinect, Asus Xtion
Battery	standard	12V, 7A
Arm (Actuator)	Dinamixel AX12 servos (x5)	Joints and servomotors from Bioloid

Robot Software

Option	Control Software	Version	Description
Robot (Simulator available)	Gazebo and ROS	Gazebo 1.19, ROS - HYDRO

Competition Benchmarks

It is possible to take part in 6 different benchmarks. We have focused our work in 4 of them described in the next sections.

FBM1 - Object Perception

The goal is to analyse the capabilities of a robot in processing perception sensor data in order to extract information about observed objects.

FBM2 - Object Manipulation

This functional benchmark is focused at assessing the capabilities of a robot to correctly operate manual commands of the types that are commonly found on domestic appliances operated by humans as light switches.

Modelling

Model 1

Model 2

Final Prototype

Lab Test

<videoflash>hDI7Yruv-u4</videoflash>

FBM3 - Speech Understanding

The goal of this functional benchmark is to evaluate the ability of a robot to understand speech commands that a user gives in a home environment.

Mountain Prototype

MountainBot Robot Prototype.

Thanks to our sponsor MOUNTAIN, we have developed a light version of MYRABot. This robot runs using the Mountain prototype which has a touch screen. Also the display is able to open 360º.