Fortunately, Kennedy and Eberhart present- ed the ParticleSwarm Optimization in 1995 [14, 15]. PSO is one of evolutionary computation tech- nique to find the best solution by acting like social behavior of groups such as fish schooling or bird flocking. There are several benefits of the PSO as compared to other evolutionary computation me- thods. The PSO is not difficult to be implemented and is computationally reasonable since its memo- ry and CPU speed requirements are low. Additio- nally, the PSO requires only a few process should be completed and parameters to be adjusted. On the other side, the PSO has quick convergence ab-ility to find optimum or near optimum solution. Generally, PSO has demonstrated to be an effect- tive method for numerous wide ranging optimiza- tion problems. Moreover, in some cases it does not suffer from the problems encountered by other evolutionary computation [11-13].
An office-room scenario that has 10 by 10 meters spaces was also used for testing the mobilerobot movement in basic behaviours combination and in complex environment. Several simulation fields as shown in Fig. 5 were designed to test the performance of the mobilerobot in basic behaviours combination. Generally, each field contains obstacles, a start and a target position. Several conflict scenarios between obstacles, walls and target point, corridor-like environment and dead end condition were included in the fields to test the ability of the proposed algorithm. The fitness values are used to analyse the performance of this proposed algorithm. A single fuzzy context rules (SFCR) applied by Hagras et al.  is used as a comparison. The fitness values of each simulation field are listed in Table II.
Since Fuzzy Logic Controller ( FLC ) can mimic human behavior,many reasearchers applied FLC to control either wheel mobilerobot .A thorough literature overview was done on the usage of FLC as applied to the various DC motor system. Ho-Hoon Lee and Sung-Kun cho proposed a new fuzzy logic anti swing control for industrial three dimensional overhead cranes.How ever, PID controller still approached for position control which is based on model controller . Yodyium and Mo-Yuen work (2000) implement Dc Motor speed control by usingfuzzy logic controller.The design of the motor is same to other motor and result are shown both loaded and unloaded condition.The controller is cheap as it required only small amount of components and easily improved to adaptive fuzzy controller.The controller provide good performance and compact size and low cost . Abdullah I. Al-Odienat and Ayman A. Al- Lawama (2008) proposed Fuzzy logic controllers (FLC’s) have the following advantages over the conventional controllers: they are cheaper to develop, they cover a wider range of operating conditions, and they are more readily customizable in natural language terms. A self-organizing fuzzycontroller can automatically refine an initial approximate set of fuzzy rules. Application of PI-type fuzzycontroller increases the quality factor .
For example, if the robot has "Back-up" behavior in case of a collision, and "Wallfollowing", "Obstacle avoidance", "Wander" behaviors allowing it to safely move, the subsumption hierarchy would be as mentioned. But if a new behavior, "Go to goal" needs to be added to the architecture, a conflict occurs (Figure 2-29). If the "Go to goal" behavior is added at top of "wallfollowing", it will subsume all the behaviors below, including "Obstacle avoidance". If it is placed below "Obstacle avoidance", it will not be able to leave a wall that it is following, and will not be able to go to the goal unless the goal is on the wall.
Where n is positive integer and T is the sampling period. y(nT), e(nT), r(nT) and a(nT) denote process output, error, rate and acc at sampling time nT, respectively. GE (gain for error) is the input scalar for rate, GA (gain for acc) the input scalar for acc and GU (gain for controller output) the output scalar of the FLC. F(.) describes the fuzzification of the scaled output of the FLC at sampling time nT.dUi(nT) (i=1,2) designate the incremental output of the fuzzy control block i from the defuzzification of the fuzzy set ‘output i’ ui~(nT) at sampling time nT. Thus the FLC includes the following components.
Some hardware implementations have been explored. Tunstel and Jamshidi described a fuzzy logic controller that provided a mobilerobot equipped with a MC68HC11 microcontroller with the capability to exhibit a wall-followingbehavior . Saffiotti et al. presented a mobilerobot that can avoid obstacles on the way and seek for the goal. It was implemented on the mobilerobot, named Flakey by SRI International . Buschka et al. uses fuzzy logic to account for errors and imprecision in visual revognition, and for extreme uncertainty in the estimate of the robot’s motion. It only requires an approximate model of the sensor system and a qualitative estimate of the robot’s displacement, and it has a moderate computational cost. The method was demonstrated on a Sony AIBO legged robot in the RoboCup demain .
The use of the fuzzy logic method in the navigation task has been analyzed by a lot of previous studies. In overcoming the obstacle avoidance and stabilization of the position of mobilerobot wheel problem Faisal et al  has designed sensor-basedfuzzy sensor wireless for mobilerobot navigation tasks between static and moving barriers. While in the paper , it can be seen that there were design and implemen- tation of the fuzzy hybrid architecture for intelligent navigation systems and mobilerobot control in avoiding obstacles in static and dynamic environments. Just as in the case of robot football, fuzzy logic is very important, applied to individual robot be- haviors and actions, especially for obstacle avoidance and achieving targets . In research , Algabri et al. have designed two fuzzy logic behaviors for mobilerobot navigation i.e., behavior to achieve targets and avoid obstacles with different scenarios. However, it is important to pay attention to the development of this archi- tecture, that is for the same path-planning problem.
where I is the number of iterations corresponding to the number of target positions, Time is the percentage of the number simulation steps performed from the total time provided, Way is the percentage of the distance left from the start position to the target position in the current stage, Coll is the number of the mobilerobot collisions with obstacles or walls, and DeltaWallSq is the sum of square of difference between the left distance and the right distance. The fitness function thus defined tries to take into account the different aspects relevant to a good robot performance: rewarding low execution times (Time) and the degree of completion of the task (Way), punishing collisions with the obstacles or walls (Coll), and maintaining the mobilerobot movement in centre of the corridor (DeltaWallSq).
This paper presents the design of an autonomous robot as a basic development of an intelligent wheeled mobilerobot for air duct or corridor cleaning. The robot navigation is based on wallfollowing algorithm. The robot is controlled us- ing fuzzy incremental controller (FIC) and embedded in PIC18F4550 microcontroller. FIC guides the robot to move along a wall in a desired direction by maintaining a constant distance to the wall. Two ultrasonic sensors are installed in the left side of the robot to sense the wall distance. The signals from these sensors are fed to FIC that then used to de- termine the speed control of two DC motors. The robot movement is obtained through differentiating the speed of these two motors. The experimental results show that FIC is successfully controlling the robot to follow the wall as a guid- ance line and has good performance compare with PID controller.
We have assigned a particular value of the action variable for each combination of condition variable based on intuition. An obstacle is very near and straight ahead, the robot deviated towards ahead left. However, when the obstacle is very near but on the left of the MobileRobot, the MobileRobot goes ahead. As the critical obstacle is away from the MobileRobot, it has a tendency to move ahead. This set of rule is pretty good and we shall see later that an FLC with this rule base can navigate well in certain scenarios. However, currently we are also extending the –fuzzy approach to adaptively find the best action variable for a particular combination of condition variable, thereby eliminating the need for such a user – defined rule base.
Starting from a randomly distributed set of particles (potential solutions), the algorithms try to improve the solutions according to a quality measure (fitness function). The improvisation is performed through moving the particles around the search space by means of a set of simple mathematical expressions which model some interparticle communications. These mathematical expressions, in their simplest and most basic form, suggest the movement of each particle toward its own best experienced position and the swarm’s best position so far, along with some random perturbations. There is an abundance of different variants using different updating rules . The general structure of a canonical PSO algorithm is shown in figure (1).
In cluster analysis, some of the features of a given data set may fall in higher relevance than others. For avoid this issue, Feature-Weighted Fuzzy C-Means (FWFCM) approaches are helpful in recent years. An Improved FWFCM (IFWFCM) 5 is overcome the certain deficiencies in the existing FWFCMs, e.g., the elements in a feature-weight vector cannot be adaptively adjusted during the training phase and also the update formulas of a feature-weight vector cannot be derived analytically.
In practical applications, the PF algorithm is easy to be affected by the dilution of samples, and a large number of samples are needed to obtain a better positioning effect, resulting in an increase in the complexity of the algorithm. Aiming at the problems of PF algorithm, this paper introduces the selection, crossover and mutation operation in evolutionary computation into PF algorithm to improve the performance of the algorithm. The evolutionary particle filter algorithm is applied to robot pose tracking and global localization. The simulation results show that the application of evolutionary strategy not only improves the diversity of samples, reduces the effect of sample dilution, and realizes the optimization of sample, reduce the number of required samples, the evolutionary particle filter algorithm has higher positioning accuracy and robustness. The further research work is to compare the performance of the proposed particle filter algorithm with the resampling algorithm and the simulated annealing particle filter algorithm.
Swarm robotics is the study of a field of multi-robotics in which a large number of robots that are coordinated in a distributed ways. It is based on the use of local rules, and simple robots compared to the complexity of the task to achieve, and inspired by social insects. Large number of mobile robots can perform complex tasks in a more efficient way if compare with a single robot, giving robustness and flexibility to the group. In this project, an overview of swarm robotics is given to introduce the main properties, characteristics and comparing it to general multi- robotic systems by researching and investigating from experimental results.
So far, it is very difficult to find an effective common image segmentation method to make various images reach the optimal segmentation quality. Many image segmentation algorithms are meant for certain type of image or certain specific segmentation. From the type of image, image segmentation includes gray image segmentation, color image segmentation and texture image segmentation. According to the definition of image segmentation, the image segmentation algorithms are divided into two types: one is based on region and it uses the regional similarity, namely assuming that the neighborhood pixels in the same region have similar characteristics such as gray, color or texture while the other is based on boundary and it uses the discontinuity between regions .
Classic fuzzy modeling and controlling structures are based on extensive expertise of the designer and some heuristic pre- knowledge, in order to avoid these shortcomings fuzzy logic and neural network structure are operated together. This approach involves merging or fusing fuzzy systems and neural networks into an integrated system to reap the benefits of both. The most important neuro-fuzzy model is the Mamdani model. The Mamdani Model incorporates an idea that local dynamics of a non linear system can be represented by different linear dynamic models. In applications of fuzzy-neural networks the learning capability of the neural networks is used for determining optimum values of fuzzy antecedent (membership) and consequent (rule) parameters.
This is to certify that the thesis entitled “Navigation of Mobil RobotusingFuzzy Logic” is the bona fide work of Krushna Shankar Sethi and SanjeevPothen Jacob under the Guidance of Dr. D.R.K.Parhi for the requirement of the award of the degree of BACHELOR OF TECHNOLOGY specialization “Mechanical Engineering” andsubmitted in the Department of Mechanical Engineeringat National Institute of Technology Rourkela, During the period 2012-2013 .
Position control system of an Electro-Hydraulic Actuator System (EHAS) is investigated in this paper The EHAS is developed by taking into consideration the nonlinearities of the system: the friction and the internal leakage. A variable load that simulates a realistic load in robotic excavator is applied as the trajectory reference. A method of control strategy that is implemented by employing a Fuzzy Logic Controller (FLC) whose parameters are optimized usingParticleSwarm Optimization (PSO) is proposed. The scaling factors of the fuzzy inference system are tuned to obtain the optimal values which yield the best system performance. The simulation results show that the FLC is able to track the trajectory reference perfectly for orifice opening. Orifice opening more than introduces chattering, where the FLC alone is not sufficient to overcome this. The PSO optimized FLC reduces the chattering significantly. This result suggests the implementation of the proposed method in position control of EHAS.
is used as a leader to update particles position. However, in the case of multi-objective optimization problems, each particle might have a set of different leaders (each non-dominated solution could be selected as a leader) which only one of them can be selected in order to update its position. In this paper, we describe a leader selection technique that is based on the density measures. For this purpose, a neighborhood radius R neighborhood is defined for leaders. Two leaders are neighbors if their Euclidean distance (measured in the objective domain) is less than R neighborhood . Using this definition, the number of neighbors of each leader is calculated in the objective function domain. The particle which has fewer neighbors is preferred as leader. However, after several iterations, the leader position and its density will change. Thus, leader selection operation should be repeated and a new leader must be identified. Therefore, the maximum iteration is divided into several equal periods, and each period has the same iteration T . The relationship of maximum iteration, number of periods and T satisfies the following equation.
In the last decade, the main developments in the area of robotics have come through technological breakthroughs in the areas of computing telecommunications, software, and electronic devices. These technologies have facilitated improvements in intelligent sensors, actuators, and planning and decision making units which have significantly increased the capabilities of mobile robots. The latest trend in robotic intelligence is toward imitating life, for instance in evolutionary robots and emotional control robots. Another area of technological challenge for the next decade is the development of micro-robots and nano-robots for medical applications. In order to drive the most suitable feedback controllers for each control system, it is convenient to classify the possible motion tasks: Point-to-point motion, Path following and Trajectory tracking.