Adelardo Adelino Dantas de Medeiros

A survey of control architectures for autonomous mobile robots

This paper identifies attributes of intelligent robotic applications and surveys the different flavor in robot control architectures. Directions in robot control are discussed and the attributes and properties of different proposals are classified and compared. In the conclusion we present our point of view about the current state of designing robot control architectures.

Optical flow using color information: preliminary results

Optical flow cannot be completely determined only from brightness information of images, without introducing some assumptions about the nature of movements in the scene. Color is an additional natural source of information that facilitates the solution of this problem. This work aims to illustrate the improvement in the optical flow estimation by using color information through experimental results.

HYPERPRESENCE-an application environment for control of multi-user agents in mixed reality spaces

The HYPERPRESENCE system proposed in this work is a mix between hardware and software platforms developed for control of multi-user agents in a mixed reality environment. The hardware is basically composed of robot systems that manipulate objects and move in a closed, real environment and a video camera, imaging system. The environment can be any place that provides or needs interactions with virtual environments for showing results or else to allow manipulation in it via a virtual reality interface. The software part is composed of three main systems: an acquisition module for position control, a tower (hardware) communication module for manipulating the robots and a multi user VRML server for managing virtual spaces and to provide synchronism between the real places and the virtual ones. We present the design and implementation solution adopted for the HYPERPRESENCE architecture. We analyze problems with communication protocols, precision of positioning, how to relate physical and virtual objects and show our solutions for these problems.

An Approach for 2D Visual Occupancy Grid Map Using Monocular Vision

This paper presents an approach that uses planar information (homography matrix) to build a visual 2D occupancy grid map from monocular vision. Initially, a segmentation step is necessary to classify parts of the image in floor or non floor. From this classification is possible to determine which parts of the image are free and which parts of the image are obstacles. Practical results are presented to validate the proposal.

A Line-Based Approach to SLAM Using Monocular Vision

This work purposes a line-based approach to the SLAM problem using planar information. The solution presented is based on Extended Kalman Filter, in Hough transform and in use of a homographic matrix to perform the image-world mapping . The proposed model allows the use of line features directly in the Kalman equations without additional intermediate calculations and startup delayed initialization, same images being obtained from a single camera. Results with a real robot are presented to validate the proposal.

Identification of Quasi-linear Dynamic Model with Dead Zone for Mobile Robot with Differential Drive

This note deals with the identification of a mobile robot with differential drive. The robot system is modeled by two MISO Hammer stein systems with input dead zones. The robot dynamic model is based on the traveled distance increment instead of the robot coordinates, making the model linear and allowing the application of classical methods of identification. Both parameters of linear and nonlinear blocks are estimated simultaneously through application of recursive least squares. Our main contribution is to use simple identification techniques to recursively estimate the linear and non-linear parameters of the robot system, that may vary over the time. The feasibility of the method is demonstrated by examples using both simulated and real robot.

A CORBA-Based System for the Monitoring of Petroleum Fields

This paper presents the definition and implementation of a CORBA-based system for the distributed monitoring of automated oil wells located in onshore petroleum fields. Conducted by the automation of industrial processes in the petroleum and gas area, the development of computational distributed systems, based on emerging middleware technologies and designed to monitor and actuate in such particular domain, has been strongly worldwide stimulated. The monitoring of the oil wells’ behavior can serve as a powerful tool for the oil production analysis. Since the monitoring system takes part of a larger oil production system with heterogeneous and distributed modules, CORBA and its services provide support for transparent interaction among such modules. The implemented system has the following goals: to register characteristics associated with the operational process of oil extraction in automated wells of oil fields, and based on this collected information, to provide descriptive and behavioral results in order to make possible an efficient monitoring process and to also supply control procedures. To support multiple clients’ invocations, the system provides concurrent monitoring, control tasks and data distribution.

Straight-lines modelling using planar information for monocular SLAM

Straight-lines modelling using planar information for monocular SLAM This work proposes a SLAM (Simultaneous Localization And Mapping) solution based on an Extended Kalman Filter (EKF) in order to enable a robot to navigate along the environment using information from odometry and pre-existing lines on the floor. These lines are recognized by a Hough transform and are mapped into world measurements using a homography matrix. The prediction phase of the EKF is developed using an odometry model of the robot, and the updating makes use of the line parameters in Kalman equations without any intermediate stage for calculating the distance or the position. We show two experiments (indoor and outdoor) dealing with a real robot in order to validate the project.

Probabilistic Mapping by Fusion of Range-Finders Sensors and Odometry

One of the main challenges faced by robotics scientists is to provide autonomy to robots. That is, according to Medeiros (Medeiros, 1998) a robot to be considered autonomous must present a series of abilities as reaction to environment changes, intelligent behavior, integration of data provided by sensors (sensor fusion), ability for solving multiple tasks, robustness, operation without failings, programmability, modularity, flexibility, expandability, adaptability and global reasoning. Yet in the context of autonomy, the navigation problem appears. As described in Fig. 1, sense, plan and act capabilities have to be previously given to a robot in order to start thinking on autonomous navigation. These capabilities can be divided into sub-problems abstracted hierarchically in five levels of autonomy: Environment Mapping, Localization, Path Planning, Trajectory Generation, and Trajectory Execution (Alsina et. al., 2002).

Pose estimation of a humanoid robot using images from a mobile external camera

This paper describes an ongoing project to move a simple humanoid robot, without an advanced embedded electronics. We used another wheeled mobile robot which already has a locating system and is equipped with a camera. We will locate the humanoid in the environment based on images, and then take the necessary actions to move it. In the future, we will also move the robot with the camera, so it will take good images of the humanoid.