Luis María Cabás

ZMP Human Measure System

This paper describes a measuring system to calculate the ZMP in humans. The main goal is to do several tests with a hardware measure system in order to analyze whether it will be possible to implement it in the Rh-0 humanoid robot. The humanoid robot Rh-0 under development at the Robotics Lab of the University Carlos III of Madrid, will have a movement control system which will measure its ZMP on-line. First of all, several tests of these possible control systems on people have been developed, measuring their ZMP while they are walking on a plane floor.

CREATING A GESTURE RECOGNITION SYSTEM BASED ON SHIRT SHAPES

The present article describes the gesture recognition system that has been developed at the University Carlos I11 of Madrid. The system’s functionality is based on the assumption that the person that wants to interact with the robot should wear a shirt which characteristics must have been previously passed on to the program. The algorithm used for the recognition is presented in detail. The results of a number of tests are presented and discussed along with the suggestions for hture improvements.

Mechanical Design and Dynamic Analysis of the Humanoid Robot RH-0

This paper presents the design process and the dynamic analysis of the 21 DOF humanoid robot RH-0, developed in the Robotics Lab. at the University Carlos III of Madrid.

Advanced Motion Control System for the Humanoid Robot Rh-0

This paper presents an advanced motion control system for the Rh-0 humanoid robot. Rh-0 is a humanoid platform which we developed on the first phase of the Rh project, which was launched by the Robotics Lab at the University Carlos III of Madrid in 2002. The motion control system of Rh-0 is based on standard commercial available hardware components. It provides scalability, modularity and application of standardized interfaces. As a result, easy upgrade and further development of reduced weight humanoid robot is possible. In this paper the hardware system, control level software and experimental results are discussed.

DETECTING SOUND SOURCES WITH THE HUMANOID ROBOT RH-1

Rh-l is a humanoid robot under development at the University Carlos I11 of Madrid. Along with other interaction types, the robot is capable of recognizing voice commands. This article describes the sound localization system that has been added recently. It is independent of the voice recognition system and uses the information From several microphones attached to the robot’s head that is further processed in order to estimate the sound source direction. Results are presented and discussed and the ideas how to improve the performance are suggested.

A “T-shirt Based” Image Recognition System

This paper presents a simple image recognition system for the Rh-0 humanoid robot that has been developed in the University Carlos III of Madrid, Spain. This system extracts some basic features from the image that may be used in the robot’s navigation and path planning. The simplicity of this system is based on the assumption that the person interacting with the robot is wearing a red T-shirt as well as on some other assumptions. In the paper we describe the vision system’s structure and the steps followed in the development of the image recognition system. Experimental results and discussion complete the description of the system.

Development of the Light-Weight Human Size Humanoid Robot RH-0

This paper describes the dynamic and mechanical development of RH-0, an autonomous humanoid robot designed by the Robotics Lab of the University Carlos III of Madrid.

User Friendly Graphical Environment for Gait Optimization of the Humanoid Robot Rh-0

The objective of this paper is to present the user friendly graphical environment for gait optimization of the humanoid robot Rh-0 that is currently under development at the Robotics Lab of the University Carlos III of Madrid. The robot has 21 DOF and is 1.35m tall. The main objective of the developed environment is to solve the kinematics problem of high DOF humanoid robots: forward kinematics, inverse kinematics and path generation. This gait is computed taking into account that the minimal torques of the robot’s joints’ will permit to use the smallest actuators. The graphical environment permits to test different gaits in order to get smooth walking in any direction, satisfies the stability, movement ranges, speed and torque requirements. The first step consists of the selection of phases of robot’s gait. The classical Denavit-Hartenberg based method for solving the forward kinematics problem is used. To solve the inverse kinematics the 3D geometric methods considering the hip and foot position are developed. The generated robot’s path depends on the gait’s length and frequency, etc. The paths generation controls the stable ZMP position in every moment.