Ján Vaščák

Use and Perspectives of Fuzzy Cognitive Maps in Robotics

Fuzzy Cognitive Maps (FCM) started in the last decade to penetrate to areas as decision-making and control systems including robotics, which is characterized by its distributiveness, need for parallelism and heterogeneity of used means. This chapter deals with specification of needs for a robot control system and divides defined tasks into three basic decision levels dependent on their specification of use as well as applied means. Concretely, examples of several FCMs applications from the low and middle decision levels are described, mainly in the area of navigation, movement stabilization, action selection and path cost evaluation. Finally, some outlooks for future development of FCMs are outlined.

Vehicle navigation by fuzzy cognitive maps using sonar and RFID technologies

Emerging concept of the so-called intelligent space (IS) offers means for use of mobile autonomous devices like vehicles or robots in a very broad area without necessity for these devices to own all necessary sensors. From this reason also new navigation methods are developing, which utilize IS means, with the aim to offer maybe not so accurate but first of all cheep and reliable solutions for a wide variety of devices. Our paper deals with the examination of possibility to interconnect sparsely deployed RFID tags with sonars. As signals produced by these two technologies are often affected by uncertainty and incompleteness we use fuzzy logic for their processing as well as control of the entire navigation process. For this purpose a special type of a fuzzy cognitive map was proposed. The paper describes real navigation experiments with a simple vehicle and evaluates them by selected criteria. Based on obtained results their explanations and conclusions for potential future research are sketched.

Interactive evolutionary optimization of fuzzy cognitive maps

Abstract Modeling dynamic systems with Fuzzy Cognitive Maps (FCMs) is characterized by the simplicity of the model representation and its execution. Furthermore, FCMs can easily incorporate human knowledge from the given domain. Despite the many advantages of FCMs, there are some drawbacks, too. The quality of knowledge obtained from the domain experts, and any differences and uncertainties in their opinions, has to be improved by different methods. We propose a new approach for handling incompleteness and natural uncertainty in expert evaluation of the connection matrix of a particular FCM. It is based on partial expert estimations and evolutionary algorithms in the role of an expert-driven optimization and outside of the FCM optimization (adaptation) research area known as Interactive Evolutionary Computing (IEC). In the present paper, a modification of IEC for the purposes of FCM optimization is presented, referred to as the IEO-FCM method, i.e., the Interactive Evolutionary Optimization of Fuzzy Cognitive Maps. Experimental results on two control problems suggest that the IEO-FCM method can improve the quality of an FCM even in situations without any measured data necessary for other known learning algorithms.

Automatic Design and Optimization of Fuzzy Inference Systems

Fuzzy inference systems have found a very spread application field, especially in areas, which interact with humans. However, they lack any self-learning capabilities for design of their knowledge bases. Beside such means as neural networks and interpolation methods also genetic algorithms are used in this area. First of all the conventional approaches of genetic algorithms have found use in rule-based fuzzy inference systems. In addition, other approaches, as parts of a broader group of evolutionary algorithms, like particle swarm optimization and simulated annealing were applied for this area. Finally, various other promising approaches like fuzzy cognitive maps were adapted for fuzzy logic, too. Therefore, the structure of this chapter has three basic parts and it deals at first with adaptation and knowledge acquisition possibilities of fuzzy inference systems in general. Consecutively, methods of using genetic algorithms for the design of rule-based fuzzy inference systems are described. In the last part the scope of fuzzy cognitive maps is analysed and some adaptation approaches based on evolutionary algorithms are introduced.

Learning of Fuzzy Cognitive Maps by a PSO Algorithm for Movement Adjustment of Robots

Motional stability and robustness play a very important role mainly in bipedal robots, especially if it is connected with a dynamic environment, where many motion changes are necessary to be done. Here, this problem is shown on kicking a ball in robotic soccer. The movement control leads to constructing movement trajectories which should secure stable behaviour. Some control approaches are oriented in creating smooth trajectories instead of complicated stability analyses. For such purposes the so-called Bezier curves are used. In this paper we use Fuzzy Cognitive Maps (FCMs) for determining parameters of Bezier curves as well as a Particle Swarm Optimization (PSO) algorithm for learning FCMs. The main advantages of PSO consist in their speed and necessity of a relatively small training set. Two types of a kicking system for generating smooth movement trajectories are proposed and compared in the paper, which is documented by performed experiments.

Local weather prediction system for a heating plant using cognitive approaches

Graphical abstractDisplay Omitted HighlightsA structure of a weather prediction system is proposed.Realized measuring points for data collection and transfer of weather variables.Designed and tested prediction model based on chained neural networks.Realized means for data modification based on fuzzy logic.A prediction system is implemented in a heating plant. Present-day requirements emphasize the need of saving energy. It relates mainly to industrial companies, where the minimization of energy consumption is one of their most important tasks they face. In our paper, we deal with the design of the so-called weather prediction system (WPS) for the needs of a heating plant. The primary task of such a WPS is timely predicting expected heat consumption to prepare the technology characterized by long delays in advance. Heat prediction depends primarily on weather so the crucial part of WPS is the weather, especially temperature, prediction. However, a prediction system needs a variety of further data, too. Therefore, WPS must be regarded as a complex system, including data collection, its processing, own prediction and eventual decision support. This paper gives the overview about existing data processing systems and prediction methods and then it describes a concrete design of a WPS with distributed data measuring points (stations), which are processed using a structure of neural networks based on multilayer perceptrons (MLP) with a combination of fuzzy logic. Based on real experiments we show that also such simple means as MLPs are able to solve complex problems. The paper contains a basic methodology for designing similar WPS, too.

Three-term relation neuro-fuzzy cognitive maps

In this paper, we propose a novel approach to modeling using fuzzy cognitive maps, which we refer to as the Three-Term Relation Neuro-Fuzzy Cognitive Map or simply the TTR NFCM. The proposed method is mostly suited to model complex nonlinear technical systems with dynamic internal characteristics. With this method we aim to solve some of the most critical problems of the conventional fuzzy cognitive maps. We target two of these problems by hybridization with artificial neural networks. First of them is a linear nature of relations between the concepts. The second is a lack of mutual dependence between the relations connecting to the same concept. Finally, we tackle a problem of relation dynamics using an inspiration from the control engineering. While focusing on bringing these advanced additional methods to the design of cognitive maps, we also aim to keep the degree of dependency on expert knowledge on the same level as with the conventional fuzzy cognitive maps. We achieve this by utilizing the machine learning methods. However, since the proposed method is heavily dependent on automated data-driven learning, it is suitable mainly for systems which are well observable and can produce sufficient training datasets.

Adaptive Fuzzy Cognitive Maps Using Interactive Evolution: A Robust Solution for Navigation of Robots

Fuzzy cognitive maps belong to emerging approaches used for various tasks in artificial intelligence. They are especially useful for solving the problem of navigation of vehicles as fuzzy systems are very robust in general. Therefore, they are suitable for the real world applications. One of disadvantages of fuzzy systems is their inability to learn. In this paper, we propose the use of fuzzy cognitive maps for navigation of a humanoid robot Nao and also an adaptive mechanism based on interactive evolution. To get data about the surrounding world, we are using the robot’s camera. Depending on the situation in the arena, the best direction is selected with the use of membership functions for target and obstacles. Parameters of these functions can be set manually from a program interface or the optimal parameters can be found using interactive evolution. The interactive evolution was selected to obtain the best results in the shortest time. Two approaches to the interactive evolution were tested. The first type was a simple interactive evolution, the second type used thresholds to find the most promising individuals to hold the ideal parameters and only these were presented to a human for evaluation. Experiments were made using manual setting of the parameters as well as using the adaptation mechanism of the first and the second type, where the second type was able to find the right set of parameters in a shorter time than the first one.

Design of a Fuzzy Adaptive Autopilot

Aircraft behaviour can be described by sets of parameters that characterize their aerodynamic properties and flight conditions. They are obtained by measuring in aerodynamic tunnels under special laboratory conditions however not fully describing all possible flight situations. Their accuracy is also limited. As considerable changes of the description occur during a flight, the use of non-adaptive autopilots, especially in a combat aircraft, is considerably limited. The removal of this obstacle is possible only by means of a continuous on-line adaptation of an aircraft model and by a consecutive adaptation of the autopilot. This paper deals with the design of a performance- adaptive fuzzy controller as an autopilot. The structure of such a contoller, as well as the adaptation principle, are described with the aim to implement it to an autopilot of a combat aircraft.

Agent-Based Cloud Computing Systems for Traffic Management

The need for a safe and economically efficient traffic represents a challenge for creating traffic management systems. This paper deals with merging three basic concepts, namely cloud-based technologies, agent-based approaches and fuzzy cognitive maps to solve the localisation and path planing for several vehicles of different types in the form of independent agents, which communicate with a supervisory system located on the cloud. The proposed system was directly tested on a playground and obtained results were analysed using several selected criteria. Finally, some further possibilities of potential utilization and future research are mentioned.