Evolutionary Learning of Dynamic Naive Bayesian Classifiers

Many problems such as voice recognition, speech recognition and many other tasks have been tackled with Hidden Markov Models (HMMs). These problems can also be dealt with an extension of the Naive Bayesian Classifier (NBC) known as Dynamic NBC (DNBC). From a dynamic Bayesian network (DBN) perspective, in a DNBC at each time there is a NBC. NBCs work well in data sets with independent attributes. However, they perform poorly when the attributes are dependent or when there are one or more irrelevant attributes which are dependent of some relevant ones. Therefore, to increase this classifier accuracy, we need a method to design network structures that can capture the dependencies and get rid of irrelevant attributes. Furthermore, when we deal with dynamical processes there are temporal relations that should be considered in the network design. In order to learn automatically these models from data and increase the classifier accuracy we propose an evolutionary optimization algorithm to solve this design problem. We introduce a new encoding scheme and new genetic operators which are natural extensions of previously proposed encoding and operators for grouping problems. The design methodology is applied to solve the recognition problem for nine hand gestures. Experimental results show that the evolved network has higher average classification accuracy than the basic DNBC and a HMM.     

SENFIS: a Sensor Node File System for increasing the scalability and reliability of Wireless Sensor Networks applications

In the last years the Wireless Sensor Networks’ (WSN) technology has been increasingly employed in various application domains. The extensive use of WSN posed new challenges in terms of both scalability and reliability. This paper proposes Sensor Node File System (SENFIS), a novel file system for sensor nodes, which addresses both scalability and reliability concerns. SENFIS can be mainly used in two broad scenarios. First, it can transparently be employed as a permanent storage for distributed TinyDB queries, in order to increase the reliability and scalability. Second, it can be directly used by a WSN application for permanent storage of data on the WSN nodes. The experimental section shows that SENFIS implementation makes an efficient use of resources in terms of energy consumption, memory footprint, flash wear levelling, while achieving execution times similarly with existing WSN file systems.     

Simultaneous learning of perception and action in mobile robots

One of the main problems of robots is the lack of adaptability and the need for adjustment every time the robot changes its working place. To solve this, we propose a learning approach for mobile robots using a reinforcement-based strategy and a dynamic sensor-state mapping. This strategy, practically parameterless, minimises the adjustments needed when the robot operates in a different environment or performs a different task.Our system will simultaneously learn the state space and the action to execute on each state. The learning algorithm will attempt to maximise the time before a robot failure in order to obtain a control policy suited to the desired behaviour, thus providing a more interpretable learning process. The state representation will be created dynamically, starting with an empty state space and adding new states as the robot finds new situations that has not seen before. A dynamic creation of the state representation will avoid the classic, error-prone and cyclic process of designing and testing an ad hoc representation. We performed an exhaustive study of our approach, comparing it with other classic strategies. Unexpectedly, learning both perception and action does not increase the learning time.     

A hybrid solution to the multi-robot integrated exploration problem

In this paper we present a hybrid reactive/deliberative approach to the multi-robot integrated exploration problem. In contrast to other works, the design of the reactive and deliberative processes is exclusively oriented to the exploration having both the same importance level. The approach is based on the concepts of expected safe zone and gateway cell. The reactive exploration of the expected safe zone of the robot by means of basic behaviours avoids the presence of local minima. Simultaneously, a planner builds up a decision tree in order to decide between exploring the current expected safe zone or changing to other zone by means of travelling to a gateway cell. Furthermore, the model takes into account the degree of localization of the robots to return to previously explored areas when it is necessary to recover the certainty in the position of the robots. Several simulations demonstrate the validity of the approach.     

Nogood-FC for solving partitionable constraint satisfaction problems

Many real problems can be naturally modelled as constraint satisfaction problems (CSPs). However, some of these problems are of a distributed nature, which requires problems of this kind to be modelled as distributed constraint satisfaction problems (DCSPs). In this work, we present a distributed model for solving CSPs. Our technique carries out a partition over the constraint network using a graph partitioning software; after partitioning, each sub-CSP is arranged into a DFS-tree CSP structure that is used as a hierarchy of communication by our distributed algorithm. We show that our distributed algorithm outperforms well-known centralized algorithms solving partitionable CSPs.     

Detecting skin in face recognition systems: A colour spaces study

Skin colour detection is a technique very used in most of face detectors to find faces in images or videos. However, there is not a common opinion about which colour space is the best choice to do this task. Therefore, the motivation for our study is to discover which colour model is the best option to build an efficient face detector which can be embedded in a functional face recognition system. We have studied 10 of the most common and used colour spaces doing different comparisons among them, in order to know which one is the best option for human skin colour detection. In concrete, we have studied the models: RGB, CMY, YUV, YIQ, YPbPr, YCbCr, YCgCr, YDbDr, HSV—or HSI—and CIE-XYZ. To make the comparison among them, we have used 15 truth images where the skin colour of a face is clearly separated from the rest of the image (background, eyes, lips, hair, etc.). Thus we can compare at level pixel each colour model, doing a detailed study of each format. We present the final conclusions comparing different results, such as: right detections, false positives and false negatives for each colour space. According to the obtained results, the most appropriate colour spaces for skin colour detection are HSV model (the winner in our study), and the models YCgCr and YDbDr.     

Adaptive decoding for brain-machine interfaces through Bayesian parameter updates

Brain-machine interfaces (BMIs) transform the activity of neurons recorded in motor areas of the brain into movements of external actuators. Representation of movements by neuronal populations varies over time, during both voluntary limb movements and movements controlled through BMIs, due to motor learning, neuronal plasticity, and instability in recordings. To ensure accurate BMI performance over long time spans, BMI decoders must adapt to these changes. We propose the Bayesian regression self-training method for updating the parameters of an unscented Kalman filter decoder. This novel paradigm uses the decoder's output to periodically update its neuronal tuning model in a Bayesian linear regression. We use two previously known statistical formulations of Bayesian linear regression: a joint formulation, which allows fast and exact inference, and a factorized formulation, which allows the addition and temporary omission of neurons from updates but requires approximate variational inference. To evaluate these methods, we performed offline reconstructions and closed-loop experiments with rhesus monkeys implanted cortically with microwire electrodes. Offline reconstructions used data recorded in areas M1, S1, PMd, SMA, and PP of three monkeys while they controlled a cursor using a handheld joystick. The Bayesian regression self-training updates significantly improved the accuracy of offline reconstructions compared to the same decoder without updates. We performed 11 sessions of real-time, closed-loop experiments with a monkey implanted in areas M1 and S1. These sessions spanned 29 days. The monkey controlled the cursor using the decoder with and without updates. The updates maintained control accuracy and did not require information about monkey hand movements, assumptions about desired movements, or knowledge of the intended movement goals as training signals. These results indicate that Bayesian regression self-training can maintain BMI control accuracy over long periods, making clinical neuroprosthetics more viable.     

Simultaneous Localization and Mapping Using Rao-Blackwellized Particle Filters in Multi Robot Systems

In this paper we investigate the problem of Simultaneous Localization and Mapping (SLAM) for a multi robot system. Relaxing some assumptions that characterize related work we propose an application of Rao-Blackwellized Particle Filters (RBPF) for the purpose of cooperatively estimating SLAM posterior. We consider a realistic setup in which the robots start from unknown initial poses (relative locations are unknown too), and travel in the environment in order to build a shared representation of the latter. The robots are required to exchange a small amount of information only when a rendezvous event occurs and to measure relative poses during the meeting. As a consequence the approach also applies when using an unreliable wireless channel or short range communication technologies (bluetooth, RFId, etc.). Moreover it allows to take into account the uncertainty in relative pose measurements. The proposed technique, which constitutes a distributed solution to the multi robot SLAM problem, is further validated through simulations and experimental tests.     

Aerodynamic and aeroelastic analyses on the CAARC standard tall building model using numerical simulation

The simulation of the wind action over the CAARC (Commonwealth Advisory Aeronautical Council) standard tall building model is performed in the present work. Aerodynamic and aeroelastic analyses are reproduced numerically in order to demonstrate the applicability of CFD techniques in the field of wind engineering. A major topic in this paper is referred to one of the first attempts to simulate the aeroelastic behavior of a tall building employing complex CFD techniques. Numerical results obtained in this work are compared with numerical and wind tunnel measurements and some important concluding remarks about the present simulation are also reported.     

Correct behavior identification system in a Tagged World

This paper presents a system that is able to process the information provided by a Tagged World to identify user’s behavior and to produce alarms in dangerous situations. The system inputs are signals from sensors, which are used to recognize correct behavior (action sequences) by Inductive Learning, using Data Mining techniques. The inference engine is a reasoning device that is implemented by means of Regular Grammars. It permits us to control user’s behavior. As output, the system produces and sends alarms when a user action sequence is wrong, indicating the erroneous actions, forgotten future, and so on. To test our system, the Tagged World is supposed to be at a house, where we have used RFID technology to control the objects inside it.