J.R. Llata

Fuzzy expert system with double knowledge base for ultrasonic classification

Abstract This article shows a pattern recognition method for object classification using ultrasonic sensors and a dual knowledge base fuzzy expert system. The developed system uses a pair of ultrasonic sensors for obtaining information about the object shape from the ultrasonic echo signal envelope. In order to reduce the size of the database, a set of parameters is calculated for extracting knowledge about the object. However, the information provided by ultrasonic sensors contains a very high uncertainty level. This uncertainty is caused by several environmental effects, which are very difficult to eliminate in industrial applications. Among these environment factors are the air temperature and humidity, the air movement, etc. They create variations in the proprieties of the medium and disturbances during the acoustic propagation process. The presented system has been specially designed for industrial applications, where it is very difficult to reduce these disturbances and where it is necessary to use intelligent systems with high autonomy. The fuzzy expert system proposed has a dual knowledge base, that is, a statistical knowledge located on the memberships functions, and the standard rule-based knowledge. This expert system deals with the uncertainties in the information, and it is able to generate and modify the knowledge base and the decision rules in an automatic way. Furthermore, it is able to adapt the knowledge base to the slow changes produced by disturbing factors, such as humidity and temperature. On the other hand, because this system maintains a rule-based structure it is very easy to incorporate expert human knowledge.

Shape recognition and orientation detection for industrial applications using ultrasonic sensors

This paper deals with a method for recognizing the form and orientation of pieces. This system uses a single pair of ultrasonic sensors to distinguish different objects and their orientations, for a set of previously learned objects. This technique utilizes the feature that small variations of position produce small variations in the value of the echo envelope parameters characterizing the ultrasonic signal. Then, neural nets are applied to learn and retrieve the necessary data in order to obtain the real position of the object. Several NN structures have been tested in order to find those that provide the best results. This system has been evaluated with symmetrical geometrical figures. Subsequently, the application was utilized in a robotic system.

A similarity measure for 3D rigid registration of point clouds using image-based descriptors with low overlap

This paper introduces a novel similarity measure for 3D rigid registration algorithms that use comparison between image-based descriptors in order to find correspondences between two partial 3D point clouds belonging to the same object. Unlike the similarity measures based on correlation coefficient, joint entropy, mutual information or others that have been used by the most popular 3D registration algorithms this similarity measure is based on distance between pixels and takes into account the problems of clutter and occlusion that can appear in real situations that need 3D registration or object recognition.

Three Dimensional Ultrasonic Vision of Surfaces for Robotic Applications

Abstract This article deals with the utilisation of ultrasonic sensor airays as an artificial vision system for robotics applications. This array will be placed on the robot grip in such a way that it will be possible to detect the presence of an object, to direct the robot tool towards it and to locate the object position. Furthermore, it will provide visual information about the surface by means of the echo-signal technique. A possible structure for the vision system is presented.

Three-Dimensional Robotic Vision Using Ultrasonic Sensors

This article describes a three-dimensional artificial vision system for robotic applications using an ultrasonic sensor array. The array is placed on the robot grip so that it is possible to detect the presence of an object, to direct the robot tool towards it, and to locate the object position. It will provide visual information about the objects surface by means of superficial scanning and it permits the object shape reconstruction. The developed system uses an approximation of the ultrasonic radiation and reception beam shape for calculating the first contact points with the objects surface. On the other hand, the position of the arrays sensors has been selected in order to provide the sensorial head with other useful capabilities, such as edge detection and edge tracking. Furthermore, the article shows the structure of the sensorial head for avoiding successive rebounds between the head and the object surface, and for eliminating the mechanical vibrations among sensors.

Pattern recognition with ultrasonic sensors: a neural networks evaluation

This paper presents an evaluation of several types of neural networks for object recognition by means of ultrasonic sensors. Initially, in order to obtain information from the ultrasonic signal, a parametric method is proposed and a set of features is extracted from the ultrasonic echo envelope. Then, it is necessary to evaluate how much information is provided for each characteristic obtained. Therefore, it has been necessary to carry out an analysis in order to detect the most relevant features. Results about information provided for each feature are presented by order of preference. Subsequently, using these features extracted from the echo signal, an experimental set‐up has been carried out in order to highlight the capabilities of different types of neural networks with this information. Finally, results obtained from experimental tests are presented, and the pattern recognition capabilities of each neural network type, using the selected features, are shown.

Accurate Estimation of Airborne Ultrasonic Time-of-Flight for Overlapping Echoes

In this work, an analysis of the transmission of ultrasonic signals generated by piezoelectric sensors for air applications is presented. Based on this analysis, an ultrasonic response model is obtained for its application to the recognition of objects and structured environments for navigation by autonomous mobile robots. This model enables the analysis of the ultrasonic response that is generated using a pair of sensors in transmitter-receiver configuration using the pulse-echo technique. This is very interesting for recognizing surfaces that simultaneously generate a multiple echo response. This model takes into account the effect of the radiation pattern, the resonant frequency of the sensor, the number of cycles of the excitation pulse, the dynamics of the sensor and the attenuation with distance in the medium. This model has been developed, programmed and verified through a battery of experimental tests. Using this model a new procedure for obtaining accurate time of flight is proposed. This new method is compared with traditional ones, such as threshold or correlation, to highlight its advantages and drawbacks. Finally the advantages of this method are demonstrated for calculating multiple times of flight when the echo is formed by several overlapping echoes.

An approach for tracking oil slicks by using active contours on satellite images

This paper describes a process for achieving oil spill detection in satellite images acquired after tanker accidents. These images have been treated with image processing techniques, such us image equalization, image binarization and morphological operations, to obtain the residue segmentation. Once the oil slick is segmented and localized in the image, an active contour is generated around it and fitted to it. The active contour provides useful information about the shape and localization of the oil spill, even providing an estimation of the deformation and the displacement that it could suffer over the time. This information could allow the evolution of residues dumped at sea to be tracked. The validity of this process is demonstrated using several ENVISAT-ASAR images acquired over several regions (Spanish, Philippine and Korean coasts), in which a tanker accident has occurred and as a consequence oil spillage has taken place.

Working Together: A Review on Safe Human-Robot Collaboration in Industrial Environments

After many years of rigid conventional procedures of production, industrial manufacturing is going through a process of change toward flexible and intelligent manufacturing, the so-called Industry 4.0. In this paper, human–robot collaboration has an important role in smart factories since it contributes to the achievement of higher productivity and greater efficiency. However, this evolution means breaking with the established safety procedures as the separation of workspaces between robot and human is removed. These changes are reflected in safety standards related to industrial robotics since the last decade, and have led to the development of a wide field of research focusing on the prevention of human–robot impacts and/or the minimization of related risks or their consequences. This paper presents a review of the main safety systems that have been proposed and applied in industrial robotic environments that contribute to the achievement of safe collaborative human–robot work. Additionally, a review is provided of the current regulations along with new concepts that have been introduced in them. The discussion presented in this paper includes multi-disciplinary approaches, such as techniques for estimation and the evaluation of injuries in human–robot collisions, mechanical and software devices designed to minimize the consequences of human–robot impact, impact detection systems, and strategies to prevent collisions or minimize their consequences when they occur.

Evaluation of expert systems for automatic shape recognition by ultrasound

This paper presents an evaluation of several types of expert systems in automatic object recognition for robotic manipulators, using ultrasound. In fact, rule-based expert systems and probabilistic expert systems have been compared in a calculation of prismatic body orientation and object shape recognition. Furthermore, both types of expert systems were used to distinguish different piece shapes and to detect object position in a real scenario using a robotic manipulator. Information for the automatic recognition system is provided to the expert system by means of the ultrasonic signal coming back from the illuminated object, which is captured by only one receiver placed on the robot grip. Subsequently, in order to reduce the number of parameters to work with, a parametric method for characterisation of this signal is presented. This has been done by calculating several geometric parameters from the signal envelope. Afterwards, a study of the probability distribution function for each parameter provides the necessary information for the expert system to carry out the distinction between the different objects of interest. In this way, it permits the establishment of a comparison among different expert system types for automatic shape recognition using ultrasounds.