Nowadays, handwriting still plays an important role in our lives, even though we live in a progressively more digital era. Therefore, losing the ability to write, whether because of physical or psychological complications, can become a huge handicap. The system described herein aims, in a first stage, to enable handwriting training through haptic guidance. The implementation was performed in Visual Studio and with the use of OpenHaptics toolkit. The second stage consists in character classification, written using a haptic joystick. The pre-eminent procedure for digit classification consists in the use of Histogram of Oriented Gradients coupled with a multiclass Support Vector Machine (HOG-SVM) whereas for letters the strategy involves using a Convolutional Neural Network (CNN).

Vertical Ground Reaction Forces (VGRF) reference profiles of human gait are important in medicine, for the recognition of gait disorders and in the assessment of rehabilitation treatments. A walking human is a dynamic varying system and in spite of the VGRF reference dependence on several patient's variables, doctors traditionally use the same static reference for all patients. The purpose of this study is to find out if an Extreme Learning Machine (ELM) is adequate to generate the dynamic VGRF reference profiles of healthy people depending on the subject's age, weight, height and stride duration. The ELM is compared with two other baseline Computational Intelligence (CI) methods, the Backpropagation Neural Network (BNN) and Multioutput Support Vector Regression (MSVR). Data from 28 healthy males walking at five different stride durations, collected using instrumented shoes, were used to train and test the CI models. The results showed that ELM is a well suited method to generate the dynamic VGRF reference profile for both dominant and non-dominant limbs, showing the lowest root mean square errors for the test set, 0.0201 and 0.0243 (fraction of body weight) for the dominant and non-dominant limbs respectively. This study reveals a promising methodology that can be implemented in real time gait analysis, allowing doctors to find a specific reference gait pattern for the gait analysis of an unhealthy person by specifying the age, weight, height and stride duration.

This paper introduces two new important issues to be considered in the design of the zero moment point (ZMP) trajectory reference for the sagittal plane balance control of an autonomous walking biped robot with an human-like gait.

The real-time balance control of an eight-link biped robot using a zero moment point (ZMP) dynamic model is difficult due to the processing time of the corresponding equations. To overcome this limitation, two alternative intelligent computing control techniques were compared: one based on support vector regression (SVR) and another based on a first-order Takagi-Sugeno-Kang (TSK)-type neural-fuzzy (NF) network. Both methods use the ZMP error and its variation as inputs and the output is the correction of the robot's torso necessary for its sagittal balance. The SVR and the NF were trained based on simulation data and their performance was verified with a real biped robot. Two performance indexes are proposed to evaluate and compare the online performance of the two control methods. The ZMP is calculated by reading four force sensors placed under each robot's foot. The gait implemented in this biped is similar to a human gait that was acquired and adapted to the robot's size. Some experiments are presented and the results show that the implemented gait combined either with the SVR controller or with the TSK NF network controller can be used to control this biped robot. The SVR and the NF controllers exhibit similar stability, but the SVR controller runs about 50 times faster.

Despite all the progress that has been made in Robotics and Artificial Intelligence, traditional approaches seem unsuitable to build truly intelligent robots, exhibiting human-like behaviours. Many authors agree that the source of intelligence is, to a great extent, the use of a huge memory, where sequences of events that guide our later behaviour are stored. Inspired by that idea, our approach is to navigate a robot using sequences of images stored in a Sparse Distributed Memory-a kind of associative memory based on the properties of high dimensional binary spaces, which, in theory, exhibits some human-like behaviours. The robot showed good ability to correctly follow most of the sequences learnt, with small errors and good immunity to the kidnapped robot problem.

This paper describes the control of an autonomous biped robot capable to be subjected to external pushing forces applied in the sagittal plane. A 1st order Takagi-Sugeno-Kang (TSK) type neuro fuzzy net, that was trained based on experimental and simulation data, is used. This net uses the zero moment point (ZMP) position and its variation as inputs, and the sagittal correction of the robotpsilas body is obtained as the output. The ZMP is obtained be acquiring use of the force distribution on four force sensors placed under each robotpsilas foot. The gait implemented on our biped is similar to a human gait. A human gait was acquired and adapted to the robot size. Some experiments were carried out and the results show that the used gait combined with the TSK neuro fuzzy net is appropriate to be used in the control of a biped robot, with and without an applied external force.

Genetic algorithms can provide good solutions for scheduling problems. In this paper we present a genetic algorithm to solve the single machine total weighted tardiness problem, a scheduling problem which is known to be NP-hard. First, we present a new concept of genetic operators for scheduling problems. Then, we present a developed software tool, called HybFlexGA, to examine the performance of various crossover and mutation operators by computing simulations of scheduling problems. Finally, the best genetic operators obtained from our computational tests are applied in the HybFlexGA. The computational results obtained with 40, 50 and 100 jobs show the good performance and the efficiency of the developed HybFlexGA.

Flexible manufacturing cells (FMC) are capable of producing a broad variety of products and changing their characteristics quickly and frequently. This flexibility provides for more efficient use of resources but makes the control of these systems more difficult. This paper presents several software applications developed for industrial robots and computer numerical control (CNC) machines. The objective of this software is to allow these equipments to be integrated, in an easy and efficient way, in modern FMC. For the industrial robots, we developed the "winRS232ROBOTcontrol" and "winEthernetROBOTcontrol" software. For the CNC machines, the software that was developed was "winMILLcontrol" for the mill and "winTURNcontrol" for the lathe. With this software, industrial robots and CNC machines can be integrated in modern FMC, in an easy and efficient way. An FMC was developed, with industrial characteristics, with the objective of showing the potentialities of the developed software. Thus, the proposed software in this paper was tested and validated in the developed FMC, and the experimental results show the viability and success of this software in the coordination of the different sectors in the FMC and in the improvement of their performance

In this paper it is presented an approach to improve the interpretability of a neuro-fuzzy system. This improvement is achieved through the modification of the Sugeno form of the consequent polynomials into corresponding triangular membership functions. The resulting neuro-fuzzy inference system has the same performance as the initial one and is an extension to our already published neuro-fuzzy architecture. This architecture has been used in the classification and control applications. In simulation, the proposed approach is applied after the corresponding neuro-fuzzy model of a non-linear function is obtained. A helicopter motion controller model was used as the non-linear function. The increase of interpretability of the controller shows the effectiveness of the proposed approach.

This work presents constraint programming (CP) approach applied to determine deadlock-free schedules in concurrent production processes. The rules of resource conflict resolution have been expressed by means of a set of predicates, which define constraints for the problem decisive variables. Predicate model has been defined in time and event domains and implemented in procedures of propagating and distributing the variable values. The predicates define sufficient conditions for existence of deadlock-free schedules of operations, which meet the prescribed qualitative and quantitative constraints

This paper presents a neuro-fuzzy controller to control a non-linear system such as the flight of a helicopter in the hover and forward flight mode positions. Hovering is a formidable stability problem, where helicopter pilots typically train for weeks before managing to do it manually. Hence automating this operation is in itself an impressive achievement. In this work, the neuro-fuzzy controller is based on the union of a fuzzy logic controller (FLC) and a general regression neural network (GRNN) for each control input of the helicopter. In the FLC's, the fuzzy inference used it is the Takagi and Sugeno type where the main difference to the other fuzzy inference types come from the specification of the consequent part. The GRNN provides estimates of continuous variables and is a one-pass learning algorithm with a highly parallel structure. Even with sparse data in a multidimensional measurement space, the algorithm provides smooth transitions from one observed value to another. A clustering algorithm was implemented to reduce the computation amount to obtain an estimate due to a large data gathered for training. The performance of the neuro-fuzzy controller is verified with simulation results.

This paper describes the application of the fuzzy logic control (FLC) theory to control the helicopter flight on two basic flight modes: hovering and forward. Hovering is a formidable stability problem, where helicopter pilots typically train for weeks before managing to do it manually. Hence automating this operation is in itself an impressive achievement. In this work, four FLC's execute control rules that represent the linguistic knowledge used by helicopter pilots and found as verbal description in pilot operating manuals. For each flight mode it is used two FLC's where the outputs corresponds to the pitch angle of longitudinal and lateral cyclic. For the hover flight mode the controllers variable input are the pitch attitude and pitch rate of the fuselage for the first controller and the roll angle and roll rate of fuselage for the second controller to control the side and forward velocity through the pitch angle of lateral and longitudinal cyclic. In the forward flight mode, it is used three variable input for each controller where four of them are the same used in the hover flight mode and it is also used the forward velocity variable in both controllers. Simulation results are presented, showing the effectiveness of the proposed FLC's for both flight modes.

A technique is proposed to detect and avoid obstacles for a mobile robot in an unknown environment. The usual problem of having too much sensorial information is dealt with by using several neural networks that cooperate in the guidance of the robot. Several unknown obstacle configurations were presented to the modular networks, proving that the MONODA architecture is very effective for obstacle avoidance when there is neither a priori nor a posteriori maps of the environment.