• Safe Reinforcement Learning (RL) as Deep Deterministic Policy Gradient is used. • Deep Deterministic Policy Gradient (DDPG) is combined with metaheuristic SMA. • The approach mitigates the drawbacks of DDPG-based safe RL optimal control. • SMA initializes the parameters of the neural network-based controller. • State safety constraints are incorporated into the search process of SMA. This paper presents a novel optimal control approach resulting from the combination between the safe Reinforcement Learning (RL) framework represented by a Deep Deterministic Policy Gradient (DDPG) algorithm and a Slime Mould Algorithm (SMA) as a representative nature-inspired optimization algorithm. The main drawbacks of the traditional DDPG-based safe RL optimal control approach are the possible instability of the control system caused by randomly generated initial values of the controller parameters and the lack of state safety guarantees in the first iterations of the learning process due to (i) and (ii): (i) the safety constraints are considered only in the DDPG-based training process of the controller, which is usually implemented as a neural network (NN); (ii) the initial values of the weights and the biases of the NN-based controller are initialized with randomly generated values. The proposed approach mitigates these drawbacks by initializing the parameters of the NN-based controller using SMA. The fitness function of the SMA-based initialization process is designed to incorporate state safety constraints into the search process, resulting in an initial NN-based controller with embedded state safety constraints. The proposed approach is compared to the classical one using real-time experimental results and performance indices popular for optimal reference tracking control problems and based on a state safety score.

This paper proposes a conventional control structure with two Takagi-Sugeno Fuzzy Observers (TSFOs) for estimating the angular speed, the overall moment of inertia and the roller radius for an electromechanical system with variable parameters. The described TSFOs are used to estimate states in a complex and nonlinear mechanism, namely strip winding system, which has the capability to wrap a strip with invariable linear speed on a roller knowing that the angular speed and the overall moment of inertia are being modified by the variable roller radius. This paper derives the conditions for the stability of the control system and the observer development, which are represented as linear matrix inequalities. The effectiveness of TSFOs is evaluated with regard to achieving a specific rate of convergence. The conventional control structure employs, in conjunction with the two TSFOs a Proportional-Integral controller with parameters optimally tuned using a metaheuristic slime mould algorithm that solves the optimization problems with objective functions described as the integrals of squared control errors multiplied by time. The control system efficacy is demonstrated and validated through digital simulation results focusing on a comparative analysis of the two TSFOs with the optimally tuned parameters.

This paper proposes the African Vultures Optimization Algorithm (AVOA)-based tuning of low-cost fuzzy controllers for the payload position control of tower crane systems. The fuzzy controllers are built around first order discrete-time intelligent Proportional-Integral (iPI) controllers with Takagi-Sugeno-Kang Proportional-Derivative (PD) fuzzy terms. The parameters of the fuzzy controllers are optimally tuned using the recent metaheuristic AVOA, which solves an optimization problem with the cost function defined as the sum of squared control error multiplied by time, and its variables are the controller tuning parameters. The control system performance improvement is proved in terms of applying only five iterations of AVOA, and the comparison with other metaheuristic algorithms that solve the same optimization problem is carried out on the basis of real-time experimental results.

This paper derives several nonlinear models of a family of nonlinear tower crane systems. First, the state-space model is improved using six parameters, which are optimally tuned using a metaheuristic Grey Wolf Optimizer algorithm. Second, fuzzy models are obtained separately for the three system outputs using an incremental online identification algorithm that develops evolving Takagi-Sugeno-Kang fuzzy models. Third, the derivation of a Tensor Product (TP)-based model is conducted. The behaviors of the tower crane systems, the evolving fuzzy models, the TP-based model and the first principles model are tested in a different scenario to the parameter identification one, and the outputs are measured. The experimental results on tower crane laboratory equipment and the comparison show the good performance of the nonlinear models derived for this challenging process and their potential for model-based control.

This article presents a hybrid metaheuristic optimization algorithm that combines particle filter (PF) and particle swarm optimization (PSO) algorithms. The new PF–PSO algorithm consists of two steps: the first generates randomly the particle population;and the second zooms the search domain. An application of this algorithm to the optimal tuning of proportional-integral-fuzzy controllers for the position control of a family of integral-type servo systems is then presented as a second contribution. The reduction in PF–PSO algorithm's cost function allows for reduced energy consumption of the fuzzy control system. A comparison with other metaheuristic algorithms on canonical test functions and experimental results are presented at the end of this article.

The aim of this paper is to demonstrate the feasibility of Whale Optimization Algorithm (WOA) in solving complex control design and tuning problems of fuzzy control systems (FCSs) with a reduced parametric sensitivity. The sensitivity analysis of these FCSs implies the use of sensitivity models defined with respect to the parametric variations of the processes. The main goal is solving the optimization problems defined for servo system processes controlled by Takagi-Sugeno-Kang proportional-integral fuzzy controllers (TSK PI-FCs), through minimization of objective functions that include the output sensitivity functions. A design method is proposed in this regard, and it is validated through experimental results using a laboratory nonlinear servo system.

Keeping the safe distance from the leading vehicle is crucial for transportation companies with a fleet of old cars. While modern Advanced Driver Assistant Systems (ADAS) might be able to estimate the distance from the front-leading vehicle, traditional ADAS do not usually offer this feature. An alternative solution is to monitor the distance using smartphones that are attached to a place such as a sun visor. The basic idea behind this approach is to detect the front-leading vehicle using the smartphone camera and estimate its distance from the car. Although SSD can achieve real-time performance on powerful GPUs, it remains challenging to run this model in real-time on mobile devices. In this paper, we propose a monocular distance estimator for forward-leading vehicles using a smartphone which is faster and more accurate than the state-of-the-art SSD detector. Specifically, we propose a layer-wise method to generate more efficient default boxes for the SSD and develop a lightweight method for estimating the distance accurately. Our experiments show that the proposed method reduces the number of default boxes by an average of 38.4% while it improves the detection rate and the processing speed compared to the original SSD. Moreover, our monocular distance estimator provides a proper safety buffer zone when the distance is greater than 20 meters. A sample video is available at https://youtu.be/-ptvfabBZWA.

This paper suggests the merge of the positive features of two data-driven control techniques, namely second order Active Disturbance Rejection Control (ADRC) and Virtual Reference Feedback Tuning (VRFT). The efficiency of this combination, referred to as ADRC-VRFT, is tested by comparison with data-driven ADRC through experimental results on nonlinear twin rotor aerodynamic system (TRAS) laboratory equipment. The parameters of ADRC-VRFT algorithm are obtained in a model-free way and the parameters of ADRC algorithm are obtained in a model-based one making use of the nonlinear first principles mathematical model of TRAS.

This paper describes multi-component spatiotemporal attention mechanisms in application to object detection in videos. The detection of objects of interest relies on the analysis of feature-point areas (FPAs), which correspond to the object-relevant focus-of-attention (FoA) points extracted by the proposed spatiotemporal mechanisms of attention focusing. The attention mechanisms give detection priority to object-relevant FPAs with spatial saliency, spatiotemporal coherence, and area temporal change including motion. The preliminary test results of the proposed attention focusing mechanisms for object detection and tracking have confirmed its advantage in terms of robustness over existing visual attention-based detectors with comparable run-times.

This paper suggests two combinations of Tensor Product (TP)-based model transformation plus sliding mode control applied to the position control of nonlinear electromagnetic actuated clutch systems. Two cascade control system structures that consist of a TP-based controller in the inner control loop and a sliding mode-based controller in the outer control loop are presented. The proposed control system structures were tested on the nonlinear process model and validated by simulation results. A comparative analysis is given.

The objective of the current paper is to improve the performance of data-driven Model-Free Adaptive Control (MFAC) by adding a Proportional-Derivative Takagi-Sugeno fuzzy controller (PDTSFC) component. This MFAC improvement is called MFAC-Proportional-Derivative Takagi-Sugeno Fuzzy Controller (MFAC-PDTSFC). MFAC-PDTSFC is applied to a Multi Input-Multi Output control structure of a mathematical model that describes the behavior of a nonlinear tower crane system (TCS) laboratory equipment. The MFAC-PDTSFC parameters are tuned in a model-based manner using metaheuristic Grey Wolf optimizer algorithm. The performances of the new MFAC-PDTSFC are compared with those of MFAC and Proportional-Derivative Takagi-Sugeno Fuzzy Controller (PDTSFC).

In this paper, we provide a proof of concept on how to model and forecast average Vessel Service Time (VST)̅ using Long Short-Term Memory (LSTM) Recurrent Neural Networks (RNNs). The proposed model is learned from the Automatic Identification System (AIS) data by using machine learning. Geohash area (GeoArea) with a certain precision, convex hull area (ConvArea), and average vessel proximity (Δ) are mined for the port of Singapore every hour. These three metrics are used to calculate port spatial complexity (SpComplexity) and port spatial density (SpDensity) indicators. In addition, we propose an algorithm to mine the (VST)̅ and associate that with the mined GeoArea, ConvArea, and Δ and the calculated indicators (i.e., SpDensity and SpComplexity). Then, an LSTM model is trained and subsequently tested to forecast future (VST)̅, as Port Authorities are increasingly relying on data-driven insights for decision-making purposes. We trained and tested several LSTM models with four different time aggregation granularities (2, 4, 6, and 8 hours) and provided performance comparisons between them in terms of Mean Square Error (MSE). The experiments emphasized the feasibility of the proposed LSTM model to forecast (VST)̅.

A variety of near-optimal problem-solving methods inspired from operations research and computational intelligence has been proposed to solve the NP-hard multi-satellite collection scheduling problem (m-SatCSP). In particular, genetic algorithms, well-suited for large-scale problems due to their simplicity and low-cost implementation have been the most pervasive and proven to be very effective. However, most reported contributions mainly promote basic permutation-based genetic principles with limited variants, overlooking prior problem structure exploitation and, widely focus on a single satellite or simple trailing satellites constellation composition. In this paper, a novel graph-based hybrid genetic algorithm is introduced to tackle the static m-SatCSP, involving a mixture of any non-trailing satellites composing a virtual constellation. The proposed Genetic Algorithm-based collecTion scHedulER (GATHER) explicitly exploits problem structure governing feasible imaging opportunity transitions through collection graphs to better handle temporal constraints. Coupled to a low-cost task scheduling heuristic integrated to genetic operators it enhances directed search and speed-up the generation of feasible high-quality solutions. The scheduling approach introduces the notion of mutual satellite orbit compatibility for dissimilar platforms to efficiently explore promising regions of the search space in evolving collection path plans. Taking advantage of those problem domain knowledge instances, the hybrid strategy evolves a mixed population of feasible/infeasible individuals to maximize the expected collection value. A multi-objective implementation named MO-GATHER is finally proposed. Computational experiments confirm that GATHER is cost-effective and competitive in comparison to alternate baseline scheduling heuristics. The benefits of MO-GATHER to decision makers are also discussed.

In this paper, we introduce three maritime Port Congestion Indicators (PCIs) mined using Automatic Identification System (AIS) static and dynamic messages. The proposed indicators are spatial complexity, spatial density, and time criticality. To calculate the PCIs, we proposed three Big AIS Data mining algorithms to find the geohash area for certain precision, the convex hull area, and the average vessels proximity within the Port Area of Interest (AOI) and in the Period of Interest (POI). The indicators are calculated for the year of 2015 for three ports (Halifax, Hong Kong, and Singapore). The proposed PCIs capture the spatial complexity, spatial density, and time of service criticality. These indicators can be used by port authorities and other maritime stakeholders to alert for congestion levels that can be correlated to weather, high demand, or a sudden collapse in capacity due to strike, sabotage, or other disruptive events. We clustered the indicators for each port into three colour-coded (Green, Yellow, and Red) clusters corresponding to low, medium and high congestion levels. The centroids of these clusters can be used to predict future congestion levels of the port under consideration. To the best of our knowledge in published literature, this work is the first to introduce the application of AIS Big Data analytics to evaluate maritime port congestion levels.

The Automatic Identification System (AIS) is a vessel tracking system that automatically provides updates on a vessel's movement and other relevant voyage data to vessel traffic management centres and operators. Aside from assisting in real-time tracking and monitoring marine traffic, this system is used in the analysis of historical navigation patterns. In this work, we mined and aggregated vessel speeds from AIS messages within geohashed regions at different precision levels. This granulated, real-world information was brought into the formulation of a Speed-based Vessel Schedule Recovery Problem (S-VSRP). The goal is to mitigate disruptions in vessel schedule by adjusting the speeds while also conforming to the historical navigation patterns reflected in the AIS data. We introduce a new model for vessel schedule speed recovery problem by formulating it as a multi-objective optimization (MOO) problem called the Big-Data-enabled Granular S-VSRP (G-S-VSRP) and propose meta-heuristic optimization methods to find Pareto-optimal solutions. The three objectives are: (1) minimizing the total delay between origin and destination ports, (2) minimizing total financial loss, and (3) maximizing the average speed compliance with historical speed limits. Three evolutionary multi-objective optimizers (EMOO) were investigated and utilized to approximate the Pareto-optimal solutions providing vessel voyage speeds. The Pareto front gives the ability to inspect the tradeoff among the three conflicting objectives. To the best of our knowledge, this is the first time historical AIS data has been exploited in the published literature to mitigate disruptions in vessel schedules.

Supply chains in the current world require extensive use of transportation modes. In combination with ever growing supply chain streamlining efforts, supply chains are particularly vulnerable to disruptions that cause profound effects on downstream actors. Much information is needed to mitigate disruptions, information that could be gathered via a maritime Internet of Things (mIoT). In this paper, we put forth a methodology to detect potentially disruptive events in a maritime supply chain and generate candidate mitigating responses. The proposed framework places risk as the cornerstone of the data-driven analysis and uses a multi-criteria decision approach to propose appropriate actions. An application of the system to mitigate a disruption in a maritime segment of a supply chain is studied. Solutions are composed of combinations of actions to reduce the consequences of a disruption. This system is validated through a scenario in which a weather event causes a disruption in maritime transportation destined to fulfill a supply contract. Finally, conclusions on the system are provided.

Human computation (HC) is an active research field in which people play a notable role as computational elements in an automated system with the aim of arriving at a truly symbiotic human-computer interaction. Situational awareness (SA) and decision support systems (DSSs) are two domains where human computation is rapidly advancing, with the latter arising as an invaluable vehicle to achieve the former. Fuzzy systems and fuzzy logic are two commonly employed tools in these domains due to their inherent capabilities of representing and processing vague and imprecise information while conveying the analysis results in an interpretable fashion. In this paper, we elaborate on the human computation aspects of risk analysis within SA and DSS conducted with the aid of fuzzy sets. The study makes the following contributions: (1) we argue that risk analysis must be a highly automated yet still human-centric endeavour and highlight four manners in which the human component provides value to the underlying data/information fusion processes; (2) we illustrate this fuzzy/human risk analysis methodology through a multi-modular Risk Management Framework (RMF) architecture and its application to the maritime domain, particularly in hard-soft data fusion, automated response generation to maritime incidents, port anomaly filtering and dynamic risk management triggered by contextual knowledge and (3) the framework under discussion can be extrapolated to other domains with negligible effort.

This paper proposes structures of Takagi-Sugeno fuzzy (TSF) controllers along with approaches to design these structures dedicated to the azimuth and pitch position control of twin rotor aerodynamic systems (TRASs). The azimuth and the pitch positions are separately controlled using Single Input-Single Output (SISO) control system structures. Two Proportional-Integral-Derivative (PID) TSF controllers are suggested for azimuth position control, and they are build around linear PID controller structures. A PI and a PID TSF controller are suggested for pitch position control by fuzzifying the linear PI and PID controller structures. The validation of the new TFS controllers is carried out on nonlinear TRAS laboratory equipment. The performance of the SISO control systems with the new TFS controllers is compared with two linear controllers tuned by a metaheuristic Gravitational Search Algorithm optimizer.

The convergence of the proliferation of inexpensive high quality video cameras, the current global security-aware environment, and increasing capability of newer generation computer hardware enables the use of computational resources to aid human operators in the video surveillance task by not only alerting on scene change detection, but providing alerts based on object classification. Current techniques used in real-time classification of video objects tend to use fast classifiers, such as linear support vector machines and decision trees due to the historic limitation of computational resources. Furthermore, the features fed to these classifiers tend to be small in size or sparse, again, due to limited computational resources. The new computational devices, which are now readily available, provide the resources to enable non-linear classification techniques, such as complex neural networks operating on large dense feature vectors, and convolutional neural networks. In this paper, we compare and evaluate several neural network architectures and their corresponding features that are capable of classifying several video objects in real-time against the Pascal 2007 dataset.

Wireless sensor networks provide a way to monitor a region of interest. Incorporating a robot into the sensor network provides a basis for other types of functionality to be added. One possibility is the replacement of damaged sensors with excess sensors within the wireless sensor network. This scenario has been defined as the “Robot-Assisted ¡Sensor Relocation” (RASR) problem and focused only on minimizing the length of the trajectory taken by the robot. RASR has been recently expanded on as a multi-objective optimization (MOO) problem to examine a more realistic scenario by considering the reliability and placement location of the passive sensors used for replacement; this new problem is termed “Reliable Robot-Assisted Sensor Relocation (RRASR). In this paper, the possibility of multiple robots servicing the sensor network is considered and the RRASR problem formulation is modified accordingly. In addition, load balancing of robots by adding an objective function to the MOO representation is included. We refer to this multi-robot version as Reliable Multiple Robot-Assisted Sensor Relocation. The performance of six state-of-the-art evolutionary MOO algorithms using sensor networks of varying sizes and inflicted damage levels is examined.

This paper proposes a set of evolving Takagi-Sugeno-Kang (TSK) fuzzy models of the nonlinear dynamic mechanisms occurring in the myoelectric-based control of prosthetic hand fingers. The rule bases and the parameters of the TSK fuzzy models are evolved by an online identification algorithm. The experimental results prove the performance of the TSK fuzzy models by good output responses and root mean square error values. A performance comparison with two recurrent neural network architectures is included.

This paper presents details regarding the design of two state feedback control (SFC) solutions for the position control of a mechatronics application represented by the Model 220 Industrial Plant Emulator. Since SFC is not effective in terms of zero steady-state control error, the SFC structure of both solutions is inserted in a control loop that contains a PID controller with or without low-pass filter. This leads to the two SFC solutions proposed in this paper and dedicated to mechatronics applications with variable moment of inertia. The PID controllers are tuned by the Modulus Optimum method to ensure high control system performance expressed as increased phase margins and improved tracking performance. The performance of the proposed SFC solutions is illustrated by case studies that deal with experimentally identified parameters in two situations, rigid body dynamics and flexible drive dynamics. Simulation and experimental results obtained for the three significant values of the moment of inertia of the load disk are given.

This paper proposes a set of evolving Takagi-Sugeno-Kang (TSK) fuzzy models of the nonlinear dynamics specific to the azimuth and position of Multi Input-Multi Output (MIMO) twin rotor aerodynamic systems. Separate Multi Input-Single Output (MISO) fuzzy models of the azimuth and pitch positions are derived for this MIMO system. The rule bases and the parameters of the MISO TSK fuzzy models are evolved by an online identification algorithm (OIA) that belongs to the class of incremental OIAs, which implement adding mechanisms. The experimental results conducted on aerodynamic system laboratory equipment prove the performance of the TSK fuzzy models by good output responses and root mean square error values. A performance comparison with a set of MISO TSK fuzzy models evolved by another incremental OIA is included.

Summary The recent advances in technology and mobile computing led to the rapid growth of networked 3D streaming applications. The emerging services can involve augmented reality, virtual environment walkthrough, multiplayer gaming just to mention a few. Because of the limited network bandwidth of the client‐server approach, research works are now turning toward mobile ad hoc networks‐based streaming, where the resources of each peer are used during the streaming service. Peer‐to‐peer technologies are considering the solution to adapt for scalable applications. Yet, supplying partner selection and 3D data delivery are still significant challenges to face because of the dynamic wireless environment that causes link breakages, high packet loss, an adverse impact on the quality of the 3D media, and a low user satisfaction. In this paper, we propose a supplying partner selection technique coupled with a content delivery technique for peer‐to‐peer 3D streaming over thin mobile devices. Our proposed protocol, which we refer to as MULTIPLY, considers multihop suppliers in order to alleviate the load on the server and uses the signal strength measurement to analyze the wireless link when sending back the 3D data. Given the high dynamicity of the network due to the mobility of the users, the streaming can be greatly affected. We therefore study the impact of the mobility on MULTIPLY. The performance evaluation of our protocol obtained using an extensive set of simulation experiments is then reported. Copyright © 2013 John Wiley & Sons, Ltd.

Mobile robots can provide a wide range of value-added services to wireless sensor networks during their operational lifetime. One of them has to do with the replacement of damaged sensors with other functional, passive ones. This scenario has been recently studied as the “Robot-Assisted Sensor Relocation” (RASR) problem. RASR solutions traditionally focus only on one aspect of the problem: minimizing the total distance traveled by the robot.The existing centralized solutions do not take into account the reliability of the passive nodes that are selected as substitutes for the damaged nodes, for instance, their current battery level. With this in mind, we propose a multi-objective optimization (MOO) formulation of the problem, named Reliable Robot-Assisted Sensor Relocation (RRASR), where we consider two more objectives in addition to the trajectory length. These objectives result from the fact that a given passive sensor selected to replace a damaged sensor in the region may not be in perfect condition and that another passive sensor may be a better option. Due to the nature of MOO problems, we must present a diverse set of solutions that exhibit a trade-off among the different decision objectives to a network manager so they may take appropriate action. We evaluate the performance of four state-of-the-art evolutionary MOO algorithms with sensor networks of varying sizes and inflicted damage levels. To the best of our knowledge, this is the first time RASR is approached from an MOO angle.

This paper presents details on the implementation of evolving Takagi-Sugeno-Kang (TSK) fuzzy models of a nonlinear process represented by the pendulum dynamics in the framework of the representative pendulum-crane systems. The pendulum angle is the output variable of the TSK fuzzy models that are obtained by online identification. The rule bases and the parameters of the TSK fuzzy models are continuously evolved by an online identification algorithm (OIA) that adds new rules with more summarization power and modifies the existing rules and parameters. The OIA is associated with an input selection algorithm that guides the modelling in terms of ranking the inputs according to their importance factors. Three TSK fuzzy models evolved by the OIA are exemplified. The performance of the new evolving TSK fuzzy models is illustrated by experimental results conducted on pendulum-crane laboratory equipment.

The paper presents aspects related to the design and implementation of a Takagi-Sugeno (TS) proportionalderivative (PD) + integral (I) fuzzy controller for processes with variable moment of inertia. A two-step design method for the TS PD+I fuzzy controller applied to position control systems is proposed. The first step concerns the Extended Symmetrical Optimum method-based tuning of the parameters of linear PID controllers organized in a parallel scheme. The second step deals first with the fuzzification of the linear PD component in the PID controller scheme resulting in the TS PD fuzzy block (TS PD FB). The modal equivalence principle is next employed to tune the parameters of TS PD FB that operates as a bump-less interpolator between separately tuned PD controllers placed in the rule consequents. The presentation is focused on the position control of a representative mechatronics application with variable moment of inertia, namely the laboratory equipment built around the Model 220 Industrial Plant Emulator. Experimental results are given to validate the PID controllers and design method in several case studies. The comparison of TS PD+I fuzzy controller versus PID controllers is supported by experimental results.

This paper introduces a high efficient local spatiotemporal descriptor, called gradient boundary histograms (GBH). The proposed GBH descriptor is built on simple spatio-temporal gradients, which are fast to compute. We demonstrate that it can better represent local structure and motion than other gradient-based descriptors, and significantly outperforms them on large realistic datasets. A comprehensive evaluation shows that the recognition accuracy is preserved while the spatial resolution is greatly reduced, which yields both high efficiency and low memory usage.

This paper proposes a novel model-free trajectory tracking of multiple-input multiple-output (MIMO) systems by the combination of iterative learning control (ILC) and primitives. The optimal trajectory tracking solution is obtained in terms of previously learned solutions to simple tasks called primitives. The library of primitives that are stored in memory consists of pairs of reference input/controlled output signals. The reference input primitives are optimized in a model-free ILC framework without using knowledge of the controlled process. The guaranteed convergence of the learning scheme is built upon a model-free virtual reference feedback tuning design of the feedback decoupling controller. Each new complex trajectory to be tracked is decomposed into the output primitives regarded as basis functions. The optimal reference input for the control system to track the desired trajectory is next recomposed from the reference input primitives. This is advantageous because the optimal reference input is computed straightforward without the need to learn from repeated executions of the tracking task. In addition, the optimization problem specific to trajectory tracking of square MIMO systems is decomposed in a set of optimization problems assigned to each separate single-input single-output control channel that ensures a convenient model-free decoupling. The new model-free primitive-based ILC approach is capable of planning, reasoning, and learning. A case study dealing with the model-free control tuning for a nonlinear aerodynamic system is included to validate the new approach. The experimental results are given.

The convergence of a security aware environment with the proliferation of inexpensive high quality video imaging devices has led to the deployment of cameras at a high number of critical infrastructure sites. As many cameras are needed to keep all key access points under continuous observation, an operator of the surveillance system may become distracted from the many video feeds, possibly missing key events, such as suspicious individuals leaving an object behind or approaching a door. By providing an automated system for monitoring these types of events within a video feed, some of the burden placed on the operator is alleviated, thereby increasing the overall reliability and performance of the monitoring system, as well as providing archival capability for future investigations. In this paper, we propose a solution that uses a background subtraction based segmentation method to determine objects within the scene. An artificial neural network classifier is then employed to determine the class of each object detected in each frame, which is then temporally filtered using Bayesian inference to minimize the effect of occasional misclassifications. The behavior of the object is then determined based on its classification and spatio-temporal properties, and if the object is considered of interest, feedback is provided to the background subtraction segmentation technique for background fading prevention reasons.

This paper introduces an innovative adaptive hybrid Particle Swarm Optimization (PSO)-Gravitational Search Algorithm (GSA) dedicated to the optimal tuning of Takagi-Sugeno-Kang PI-fuzzy controllers (T-S-K PI-FCs). The adaptive hybrid PSO-GSA is comprised from five stages, which support the solving of optimization problems with objective functions that depend on the control error and on the output sensitivity function, and the variables of the objective functions are the fuzzy controller tuning parameters. The adaptive hybrid PSO-GSA is included in the controller tuning to offer control systems with T-S-K PI-FCs that ensure a reduced process parametric sensitivity. Digital simulation and experimental results are given to validate the fuzzy controller tuning in a laboratory nonlinear servo system application.

This paper suggests a new generation of optimal PI controllers for a class of servo systems characterized by saturation and dead zone static nonlinearities and second-order models with an integral component. The objective functions are expressed as the integral of time multiplied by absolute error plus the weighted sum of the integrals of output sensitivity functions of the state sensitivity models with respect to two process parametric variations. The PI controller tuning conditions applied to a simplified linear process model involve a single design parameter specific to the extended symmetrical optimum (ESO) method which offers the desired tradeoff to several control system performance indices. An original back-calculation and tracking anti-windup scheme is proposed in order to prevent the integrator wind-up and to compensate for the dead zone nonlinearity of the process. The minimization of the objective functions is carried out in the framework of optimization problems with inequality constraints which guarantee the robust stability with respect to the process parametric variations and the controller robustness. An adaptive gravitational search algorithm (GSA) solves the optimization problems focused on the optimal tuning of the design parameter specific to the ESO method and of the anti-windup tracking gain. A tuning method for PI controllers is proposed as an efficient approach to the design of resilient control systems. The tuning method and the PI controllers are experimentally validated by the adaptive GSA-based tuning of PI controllers for the angular position control of a laboratory servo system.

The protein docking problem refers to the task of predicting the appropriate matching of one protein molecule (the receptor) to another (the ligand), when attempting to bind them to form a stable complex. Research shows that matching the three-dimensional geometric structures of proteins plays a key role in determining a so-called docking pair. However, the active sites which are responsible for the binding do not always present a rigid-body shape matching problem. Rather, they may undergo deformations when docking occurs, which complicates the process. To address this issue, we present an isometry-invariant and topologically robust partial shape descriptor method for finding complementary protein sites. Our method employs Heat Kernel Signature shape descriptors which are based on the diffusion of heat on surfaces. Our experimental results against the Protein-Protein Benchmark 4.0 demonstrate the viability of our approach.

Evolutionary algorithms are a class of algorithms that try to mimic natural, biological evolution a la Darwinian natural selection, to compute solutions to a given problem. They are especially useful when no well known strategies for computing solutions to such a problem exist. Evolutionary algorithms begin by creating a collection (population) of candidate solutions to the problem at hand; and through repeated application of genetic operators such as crossover and mutation, they iterate over multiple generations of this population, until they eventually converge onto an attractive solution. One important problem facing code implementing Evolutionary Algorithms is that due to the dynamic nature of the individual chromosomes in a population, simple coding errors lead to complex bugs that are difficult to both diagnose and debug. This problem is only exacerbated when attempting to develop the algorithms in a dynamically typed language such as Python. This paper presents a novel Evolutionary Algorithm framework for the Python programming language that implements design-by-contract, a paradigm in which each function and class must follow a contractual set of pre-conditions and post-conditions. Failure to follow the contract causes an error condition identifying the violated clause, thereby catching bugs earlier in the development process and in a more descriptive manner.

Software testing is an important and time consuming part of the software development cycle. While automated testing frameworks do help in reducing the amount of programmer time that testing requires, the onus is still upon the programmer to provide such a framework with the inputs on which the software must be tested. This requires static analysis of the source code, which is more effective when performed as a peer review exercise and is highly dependent on the skills of the programmers performing the analysis. Thus, it demands the allocation of precious time of highly skilled programmers. An algorithm that automatically generates inputs to satisfy test coverage criteria for the software being tested would therefore be valuable, as it would imply that the programmer no longer needs to analyze code to generate the relevant test cases. This paper explores a hybrid evolutionary strategy with an evolutionary algorithm to discover such test case synthesis, in an improvement over previous methods which overly focus their search without maintaining the diversity required to cover the entire search space efficiently.

Local spatio-temporal features and bag-of-features representations have become popular for action recognition. A recent trend is to use dense sampling for better performance. While many methods claimed to use dense feature sets, most of them are just denser than approaches based on sparse interest point detectors. In this paper, we explore sampling with high density on action recognition. We also investigate the impact of random sampling over dense grid for computational efficiency. We present a real-time action recognition system which integrates fast random sampling method with local spatio-temporal features extracted from a Local Part Model. A new method based on histogram intersection kernel is proposed to combine multiple channels of different descriptors. Our technique shows high accuracy on the simple KTH dataset, and achieves state-of-the-art on two very challenging real-world datasets, namely, 93% on KTH, 83.3% on UCF50 and 47.6% on HMDB51.

This paper proposes the data-driven performance improvement of low-cost control systems (CSs) for vertical three-tank systems. The MIMO CSs dedicated to two tanks of the three-tank systems consist of two SISO control loops with separately tuned PI controllers. The Modulus Optimum method is applied to initially tune the PI controllers. Optimization problems are defined on the basis of an original objective function which depends on the controller tuning parameters and is expressed as the sum of squared output errors multiplied by variable weights. The performance improvement is achieved by a new convergent Iterative Feedback Tuning (IFT) algorithm which aims the parameter tuning of PI controllers by the experiment-based solving of the optimization problems. The convergence is ensured by the formulation of the parameter update laws in the IFT algorithm as a nonlinear dynamical feedback system in the parameter space and iteration domain and by setting the step sizes to fulfill inequality-type convergence conditions derived from Popov's hyperstability theory. The experimental results for a laboratory vertical three-tank system show the convincing CS performance improvement by few experiments.

This paper proposes a data-driven algorithm that solves an optimal control problem by iteratively tuning the controller. The data-driven algorithm solves the optimization problem for a nonlinear process with a linear controller, accounting for operational constraints and employing an interior-point barrier (IPB) algorithm. The search process in the IPB algorithm requires first-order information which is generated using identified models via neural networks in order to reduce the number of experiments. A case study which deals with the angular position control of a nonlinear aerodynamic system is included to validate the new algorithm by simulation results.

Summary form only given. After more than half of century of manufacturing-oriented robotics, robots are now evolving and expanding into new applications involving a human-style interaction with people in unstructured environments. Such an endeavour requires a different set of skills for the new generation of robots, which are expected to exhibit an increased ability to intelligently connect perception to action. Due to their expected omnipresence in the lives of people, the new-generation robots should be sensitive to the attitudes and expectations of normal people. They should be endowed with more efficient interaction capabilities allowing for a higher degree of flexibility and autonomy in order to make appropriate decisions in routine and emergency situations. In order to naturally blend within human society, the new-generation robots should not only look as humans, but should also behave as much as possible as humans. They are in a way expected to be, as initially imagined by Capek in his R.U.R. Rossum's Universal Robots play, anthropomorphic artefacts, androids, enabled to think on their own and governed by Asimov's laws of robotics hardwired into every robot's positronic brain. While for a long time, engineers have built upon mathematics, physics and chemistry in order to develop an ever growing variety of industrial artefacts and machines, this approach cannot anymore rise to the challenge of designing these androids. The time has now arrived to add biology and more specifically, human anatomy, physiology and psychology to the scientific sources of knowledge to develop a new, bio-inspired, generation of intelligent androids. Advocating this emergent trend, this presentation will discuss a number of relevant issues such as bioinspired robot sensors and neural networks, humanrobot interaction techniques for symbiotic partnership, as well as moral, ethical, theological, legal, and social challenges in a soon-to-be cyborgsociety world.

SUMMARY In recent years, three‐dimensional (3D) streaming over thin mobile devices has received very little attention from the research community. The 3D streaming‐based class of applications uses either a centralized or a distributed approach. whereas the former presents several drawbacks such as latency, server's bottleneck, and networks congestion; the latter, that is, peer‐to‐peer, has to deal with the supplying partner issue that consists in selecting the source that will be responsible for streaming the required 3D data. Given that frequent 3D streaming overuses the mobile device's resources and dissipates its energy, the energy factor should be taken into consideration during the selection of the supplier. In this paper, we propose two energy management control protocols for supplying partner selection in peer‐to‐peer 3D streaming that we refer to as one‐level energy‐based and two‐level energy‐based. We also propose a new source load estimator that takes into account two factors namely the source's residual energy and its number of served requests. In one‐level energy‐based, the requester uses the source's load information to prioritize the sources in an ascending order and distribute the requests starting with the least loaded sources. In two‐level energy‐based, both the requester and the supplier have important and complementary roles and participate in the energy management control. We then report on the performance evaluation of our energy management control protocols using an extensive set of simulation experiments with the NS2 tool. Copyright © 2011 John Wiley & Sons, Ltd.

This paper proposes a data-driven algorithm that solves a reference trajectory tracking problem defined as an optimization problem. The new data-driven reference trajectory tracking algorithm (DDRTTA) solves the optimization problem in the framework of iterative learning control (ILC). The DDRTTA updates the reference input sequence using an experiment-based approach which accounts for operational constraints and employs an interior point barrier algorithm. Therefore the DDRTTA combines the advantages of data-driven control and ILC. A case study which deals with the angular position control of a nonlinear servo system is included to validate the DDRTTA by experimental and simulation results.

The purpose of this research is to assess the sensitivity of humans to perceive asynchrony among media signals coming from a computer application. Particularly we examine haptic-to-video and haptic-to-audio skew. For this purpose we have designed an experimental setup, where users are exposed to a basic multimedia presentation resembling a ping-pong game. For every collision between a ball and a racket, the user is able to perceive auditory, visual, and haptic cues about the collision event. We artificially introduce negative and positive delay to the auditory and visual cues with respect to the haptic stream. We subjectively evaluate the perception of inter-stream asynchrony perceived by the users using two types of haptic devices. The statistical results of our evaluation show perception rates of around 100 ms regardless of modality and type of device.

Due to the important developments in the area of wireless networks and mobile computing, P2P networks-based 3D streaming over thin mobile devices is starting to gain the interest of many research works. Nevertheless, one of the main challenges that need to be resolved is the supplying partner selection issue that consists of selecting the source that holds the required 3D data. The selection process can be easy to apply, however several criteria need to be taken into consideration, in order to improve the system performance. In this paper, we propose a two-level energy-based supplying partner protocol for mobile P2P 3D streaming, which we refer to as 2L-EB. Our proposed protocol is applied at two levels: the requester's side and the supplier's side. We then report on the performance evaluation we have obtained to evaluate our 2L-EB protocol using an extensive set of simulation experiments with the NS2 tool.

Interference cartographs can be utilized to determine no-transmit zones and to set the interference-free transmit power for secondary users in cognitive radio networks. This paper provides a performance comparison of common interpolation techniques for generating interference cartographs. The performances of natural-neighbor, thin-plate spline and kriging interpolations are evaluated in terms of primary emitter localization accuracy and RF field strength estimation efficiency. The simulation results suggest that the natural-neighbor interpolation technique achieves the same level of accuracy and provides more desirable features suitable for cognitive radio networks.

This paper proposes an iterative Correlation-based Tuning (CbT) algorithm which guarantees the control system stability throughout the iterations. The new algorithm is based on a Robbins-Monro procedure which ensures the iterative tuning of controller parameters such that to minimize a cost function expressed as the squared sum of the cross-correlation function between the output error and the reference input. The control system stability is tested using the coprime factor uncertainty of the controller, and the small gain theorem for discrete-time systems is applied. Nonparametric frequency domain models are employed in the calculation of the bounds on systems' gains. A case study concerning the speed control of a nonlinear servo system is included to validate the new stable CbT algorithm, and experimental results are given.

The paper proposes low-cost adaptive control solutions dedicated to the position control of electromagnetic actuated clutch systems. The initial nonlinear model of the plant is simplified and next linearized to use it in the controller design procedures. A comparative analysis between five control solution (CS) — the classical PI and PID CS, the fuzzy CS, the adaptive CS with PI gain-scheduling controllers and fuzzy PID gain-scheduling CS — is carried out. The solutions were tested based on a nonlinear simplified model of the plant.

This paper presents a novel adaptive Gravitational Search Algorithm (GSA) for the optimal tuning of fuzzy controlled servo systems characterized by second-order models with an integral component and variable parameters. The objective functions consist of the output sensitivity functions of the sensitivity models defined with respect to the parametric variations of the processes. The proposed adaptive GSA solves the optimization problems resulting in a new generation of Takagi-Sugeno proportional-integral fuzzy controllers (T-S PI-FCs) with a reduced time constant sensitivity. A design method for T-S PI-FCs is then proposed and experimentally validated in the representative case study of the optimal tuning of T-S PI-FCs for the position control system of a servo system.

Childhood obesity is nowadays considered as one of the major health problems that many societies suffer from. The obesity epidemic leads to several life threatening conditions such as diabetes, heart disease, high blood pressure, and mental health problems like depression, anxiety and loneliness just to mention a few. Several approaches, including physical exercises, strict dietary, and exergames among others, have been adopted to address the obesity epidemic. Exergames are considered the innovative approach for fighting several health problem such as the obesity, where a combination of exercise and 3D gaming are proposed to incite kids to exercise as a team. Collaborative exergaming became even more popular given that it addresses the social side of the obesity epidemic, and it motivates kids to socialize with other kids. Traditional exergames are based on the client server approach where the server is responsible for streaming the 3D environment. However, this can lead to latency and server bottleneck if many clients participate in the exergame, which leads to the kids stopping exercising. Having an exergame application that does not suffer from networking problem such as delay, is very important given that it increases the exercise hours. In this work, we propose a new trend of mobile collaborative exergming applications that is based on the peer-to-peer (P2P) architecture, as well as two supplying partner selection protocols that aim at selecting the suitable source responsible for streaming the relevant 3D data. Our system, that we refer to as MOSAIC, is intended for mobile collaborative exergames that incite kids to move inside a large area, using thin mobile devices such as head mounted devices (HMD), have physical exercises, and collaborate with other kids which in consequence address several health problems such as the obesity epidemic on the physical and social plans. Our proposed mobile collaborative exergame aims at inciting the kids to exercise as a team for a longer time by improving the quality of the streaming and reducing the delay. This is accomplished by our proposed supplying partner selection protocols that provide a quick discovery of multiple supplying partners, by minimizing the time required the to acquire data. The performance evaluation we have obtained to evaluate our suite of protocols using a realistic set of exergame scenarios for obese kids is then presented and discussed.

This paper discusses the design and implementation of a framework that automatically extracts and monitors the shape deformations of soft objects from a video sequence and maps them with force measurements with the goal of providing the necessary information to the controller of a robotic hand to ensure safe model-based deformable object manipulation. Measurements corresponding to the interaction force at the level of the fingertips and to the position of the fingertips of a three-finger robotic hand are associated with the contours of a deformed object tracked in a series of images using neural-network approaches. The resulting model captures the behavior of the object and is able to predict its behavior for previously unseen interactions without any assumption on the object's material. The availability of such models can contribute to the improvement of a robotic hand controller, therefore allowing more accurate and stable grasp while providing more elaborate manipulation capabilities for deformable objects. Experiments performed for different objects, made of various materials, reveal that the method accurately captures and predicts the object's shape deformation while the object is submitted to external forces applied by the robot fingers. The proposed method is also fast and insensitive to severe contour deformations, as well as to smooth changes in lighting, contrast, and background.

This paper describes an application created on an Android mobile phone that recognizes gestures using the smartphone's orientation sensor. These gestures can be used to trigger events in another program running on a remote computer. We present a prototype application that generates different events for controlling a PowerPoint presentation, namely starting and stopping, as well as displaying the next and previous slide.

Due to the rapid improvements of wireless technologies and mobile devices'capabilities, we are witnessing an important growth in applications using thin mobile devices. Augmented reality took advantage of the enhancement of thin mobile devices, which led to the emergence of Mobile Augmented Reality (MAR). Given that thin mobile devices can not have the entire virtual environment (VE) stored locally, multimedia streaming has been designed and is widely used nowadays in MAR. Most of the MAR systems apply the conventional client-server approach, that may suffer from delay and congestion, which affects the system performance. P2P-based multimedia streaming is considered, therefore, an alternative approach. Although mobile P2P based video streaming received a great deal of attention, mobile P2P based 3D streaming was not the choice of designed systems, mainly due to the limited mobile device's capabilities. Moreover, the main challenge that needs to be taken care of when designing mobile P2P based 3D streaming systems, is the selection of a supplying partner. A few studies have proposed supplying partner protocols for mobile P2P-based 3D streaming. However, they did not take into account the device's energy that is considered an important constraint when dealing with 3D streaming. In this paper, we propose an efficient energy-aware supplying partner protocol for mobile P2P networks based 3D streaming. A new load estimator is defined that takes into consideration two factors: the node's remaining energy and its number of served requests. We therefore propose the integration of an energy-aware criterion into the supplying partner selection in order to reduce the generation of overloaded nodes, and their energy dissipation. We then report on the performance evaluation we have obtained to evaluate our protocol using an extensive set of simulation experiments.

This paper treats the application of two data-based model-free gradient-based stochastic optimization techniques, i.e., iterative feedback tuning (IFT) and simultaneous perturbation stochastic approximation (SPSA), to servo system control. The representative case of controlled processes modeled by second-order systems with an integral component is discussed. New IFT and SPSA algorithms are suggested to tune the parameters of the state feedback controllers with an integrator in the linear-quadratic-Gaussian (LQG) problem formulation. An implementation case study concerning the LQG-based design of an angular position controller for a direct current servo system laboratory equipment is included to highlight the pros and cons of IFT and SPSA from an application's point of view. The comparison of IFT and SPSA algorithms is focused on an insight into their implementation.

The rapid spread of wireless mobile devices and the advances of wireless communication have fueled the interest about streaming 3D graphics on mobile devices to be used in augmented reality based classes of applications. In these types of applications, a real world is mapped into the virtual world and thin mobile devices are employed to navigate in the virtual simulated environment (VE). In evidence, one of the prime difficulties in 3D streaming over thin mobile devices consists of the limited mobile resources and capabilities, i.e., low processing power, limited storage capacity, limited graphics' hardware and graphics' accelerator making it very difficult for mobile devices to render and process large and complex 3D scenes. So far, a significant body of work has been dedicated to the challenges of mobile networks-based 3D streaming such as streaming performance, and bandwidth limitation. On the downside, very few studies have been committed to the mobile supplying partner strategies aiming at determining the peer that owns the correct information and that possesses enough bandwidth to send the required data quickly and efficiently to other peers in need. In this paper, we propose MOSAIC, our supplying partner strategy protocol for mobile networks-based 3D streaming. MOSAIC is based on the quick discovery of multiple supplying partners, by optimizing the time required by peers to acquire data, avoiding unnecessary messages propagation and network congestion, and decreasing the network bandwidth over utilization.

This paper suggests a new stability analysis approach dedicated to a class of fuzzy control systems controlling multi input-multi output (MIMO) nonlinear processes by means of Takagi-Sugeno fuzzy logic controllers. The approach is based on LaSalle's global invariant set theorem, and an original stability theorem offers sufficient stability conditions. The applicability and efficiency of the theoretical results are illustrated by a MIMO case study dealing with the fuzzy control of a spherical three tank system.

When a certain control function is hard to achieve using a single intelligence technique, collaboration between different ones may succeed in performing such a complicated mission. This paper shows a mobile robot playing a significant role in a clean-room medical factory, where it is not recommended for the human to work. In that environment, neural networks and fuzzy logic were combined to form a suitable solution to perform the dedicated missions. In order to perform the speed control of a mobile robot, multi-layered neural networks designed for environmental recognition, and local navigation feed the fuzzy system with signals of change in direction with the nature of the sub-space of the working environment. To prove this concept, a computer based design and test of the computational intelligence system is performed. This system includes three neural controllers for local navigation, two neural networks for environmental recognition, and a fuzzy system for speed control. The system is fed off-line by a simulated model of a laser range-finder. These major components of the control system perform a global neural navigation and a fuzzy-neural speed control that guide a mobile robot to track its predefined path to arrive to its final goal through a set of sub-goals, or autonomously plan its path to arrive to the desired final goal, while avoiding obstacles that are found along the way.

The goal of the work presented in this paper is to develop a novel scheme for the measurement and representation of deformable objects without a priori knowledge on their shape or material. The proposed solution advantageously combines a neural gas network and feedforward neural network architectures to achieve diversified tasks as required for data collection on one side and the modeling of elastic characteristics on the other side. Data is collected for different objects using a joint sensing strategy that combines tactile probing and range imaging. The innovative object models, built as multi-resolution point-clouds associated with ¿tactile patches¿, present certain advantages over classical deformable 3D object models.

This paper introduces a haptic data reduction and transmission technique to reduce the packet rate in six degree-of-freedom (6-DoF) haptic-enabled telementoring systems. The presented method relies on the limitations of human haptic perception (i.e. the Just Noticeable Differences) with respect to a user's hand position and orientation in order to reduce the number of packets transmitted without compromising transparency. A haptic prediction model is exploited to further reduce the amount of haptic packets transmitted, and to improve the reconstruction of data samples on the receiver side. Several distance metrics are also discussed to evaluate the acuity of human haptic perception when data reduction is performed in 6-DoF settings. Psychophysical experiments validate the effectiveness of the suggested algorithm as great haptic data reduction is achieved (up to 96%), while preserving the overall quality of the telementoring environment.

In this paper, the importance of the user's primary holding pivot point on the end effector of a haptic interface is discussed. Both theoretical analysis and experimental results demonstrate that this holding pivot point is critical for the correct perception of the haptic properties assigned to the virtual objects. We also study the physical inertia effects of the end effector on the non-uniform stiffness perception of simulated virtual objects. To the best of our knowledge, no such combined consideration of holding pivot point and device structure related inertia has so far been made for works in 6-DOF haptic rendering. We have instrumented the end effector of a haptic interface with a membrane potentiometer to measure the user's primary holding pivot point in real-time. Accordingly, a preliminary adaptive feedback method is developed to render the appropriate forces/torques to compensate for the effects of the end effector's inertia on the haptic stiffness perception.

Mobile robots could play a significant role in places where it is impossible for the human to work. In such environments, neural networks, instead of traditional methods, are suitable solutions to locally navigate and recognize the environmentpsilas subspaces. In order to learn and perform two important functions ldquoenvironmental recognitionrdquo and ldquolocal navigationrdquo, multi-layered neural networks are trained to process distance measurements received from a laser range finder. This paper will focus on a computer based design and test of this neural system, that includes three neural controllers for local navigation, and two neural networks for environmental recognition, fed off-line by a simulated model of a laser range-finder. These neural networks are the major components of a control system that performs a global neural navigation of a mobile robot, which could be used to perform industrial missions within industrial environments. This control system can guide a mobile robot to track its predefined path to arrive to its final goal through a set of sub-goals, or autonomously plan its path to arrive to the desired final goal, and to avoid obstacles that are found along the way.

This article presents the results of some experiments in parallelizing the training phase of a feed-forward, artificial neural network. More specifically, we develop and analyze a parallelization strategy of the widely used neural net learning algorithm called back-propagation. We describe two strategies for parallelizing the back-propagation algorithm. We implemented these algorithms on several LANs, permitting us to evaluate and analyze their performances based on the results of actual runs. We were interested on the qualitative aspect of the analysis, in order to achieve a fair understanding of the factors determining the behavior of this parallel algorithms. We were interested in discovering and dealing with some of the specific circumstances that have to be considered when a parallelized neural net learning algorithm is to be implemented on a set of workstations in a LAN. Part of our purpose is to investigate whether it is possible to exploit the computational resources of such a set of workstations.

Most of the current laser pointing interfaces use a vision based approach, which requires camera calibration. This paper presents the use of a planar homography-based auto-calibrated technique to eliminate the camera calibration step, thus simplifying the setup process. The system performance of a user interface implemented with this technique is then analyzed in details.

In this paper, we describe an experimental performance-driven animation system for an avatar face using model-based video coding and audio-track driven lip animation.

This paper describes a flexible, coherent framework for organizing and operating a robotic assembly cell that has multiple robots. This framework, which is based on the concept of an agent as understood in the field of multiagent systems, supports the task-level programming and scheduling of assembly operations. In this framework, we decompose an assembly operation into two distinct phases: part fetching and part assembling. The agents of the cell correspond to these two assembly phases plus an agent to manage the shared physical space of the cell and an agent to properly schedule the assembly operations. The simulations and experiments with this framework are presented and discussed.

Delicate tasks for space robots such as the assembly of the International Space Station, require manipulator control systems which cope effectively with structural flexibility and oscillations typical for space robots. The use of vision systems offer new possibilities for precision control for space robots but also poses new challenges due to the noncollocated sensor/actuator configuration on the flexible robot structure. This paper investigates associated stability problems and presents a new control concept which is robust with respect to the variations, nonlinearities and unknowns in the behaviour of large space robots.

The reconstruction of three dimensional (3-D) objects is used in CAD/CAM, robotics, remote sensing, etc. The models (images) can be either directly acquired by using special devices such as range finders, CTR scanners, etc. or they can be recovered from a series of 2-D images of the object. The authors report on a new method for reconstructing 3-D objects from two 2-D images using CCD cameras. This method uses a pseudo-random encoded grid projected on the object. The grid nodes are used in matching left to right 2-D images. The set of matched points are further used to calculate the disparity of each point of the object surface. Experimental examples illustrate the performance of this simple and elegant technique.

This paper presents a new grid node indexing method based on pseudo-random binary array (PRBA) encoding. This method requires only one code bit per grid step, independent of the desired grid resolution. Applications are discussed for 3-D object recognition and for 2-D absolute position recovery of a free- ranging mobile robot.

Summarizes current development of a robotic sensing system to enhance the perceptive intelligence of manipulators working in relatively structured environments. The system consists of vision and tactile sensors which would be used to identify objects within a workspace and to determine their absolute position and orientation. A visual object recognition strategy is discussed, and a technique for fusing robot kinematics and tactile image data is described.>

An automated guided vehicle (AGV) having the ability to recover its absolute position anywhere on the guide-path is described. It uses an original guide-path encoding technique, based on the properties of pseudorandom binary sequences, resulting in a minimum code complexity of 1 bit per quantization step. An experimental AGV system was built to test the proposed absolute position measurement method and its application for AGV navigation when unexpected obstacles are encountered on the guide-path. The results recommend this technique for implementation as a stand-alone function to cost-effectively upgrade existent optically guided industrial AGVs.>