Carmadi Machbub

Design and implementation of adaptive neural networks algorithm for DC motor speed control system using simple microcontroller

This paper shows the design and implementation of adaptive neural networks controller for DC motor speed control system. The DC motor is actuated by a pulse width modulation (PWM)-based H-bridge actuator. By using certain learning methods to the networks, changing a plants parameter can be estimated and used to produce an appropriate control action. The experiment shows that the controller conforms to the adaptive control scheme, although there are flaws which relate to feedback resolution. The most interesting result is that the implementation circuit is based on a very simple 8-bit microcontroller system.

Optimizing Voice over Internet Protocol (VoIP) networks based-on Extended E-model

In this paper, we optimize low speed networks access for designing an efficient Voice over Internet Protocol network, especially in 64 kbps, 128 kbps, and 256 kbps bandwidth capacities. We draw some analytical approach based on Extended E-model to quantify some levels of quality of service. Using the model, those levels are determined by network parameters such as delay, jitter, packet loss level, network utilization, and bandwidth capacity and by implementation configuration such as voice coder, packetization scheme, and size of jitter buffer. Our objective is to find maximum number of calls in some given bandwidth capacities while maintaining a certain level of the quality of service. Using numeric estimation, we found that the optimum solution for some given bandwidth capacities can be achieved by applying G.723.1 5.3 kbps voice coder, packet loss level less than 1%, jitter less than 80 ms, and network utilization less than 80%.

Integrated battery-electric vehicle model and its optimal speed control application

Energy-efficient system is very desirable in an electric vehicle. Having energy-efficient control systems in the vehicle should be advantageous in this regard. This paper presents an integrated battery-electric vehicle model to capture battery dynamics. In order to evaluate the usefulness of the model for having energy-efficient system, a linear quadratic integral (LQI) control of the vehicles speed has been designed. Simulations have been performed with observations on an energy consumption formulation. The simulations show that the use of the integrated battery model in the control system design may result in more energy-efficient system in an electric vehicle application.

White : Controlling The Internet Congestion With Integral Controller

Congestion occurs when resource demands exceeds the capacity. At this moment, congestion are solved by the involvement of two controlling mechanism, they are flow control in the TCP source and active queue management (AQM) in the link. AQM will provide feedback to the source a kind of indication for the occurrence of the congestion in the link, whereas the source will adapt the sending rate appropriate with the feedback. The indication of the feedback congestion is influenced by the number of the sources, time delay, and its sending rate. Consequently, adaptation of the sending rate based on the feedback indirectly observes behavior of other sources. This control system structure is much known with distributed control system, because the control mechanism is distributed in each source. On this structure, the source mechanism is more important than the link. This paper will explain a new algorithm, which focus more on the link as an allocation capacity controller. Then the source will adapt the sending rate based on set point the link provides

Comparison of prediction methods for moving objects in 3D coordinates using Kalman Filter and Least Square

This paper proposed for comparison of two methods used to predict the trajectory of a moving object in 3D fields using Kalman Filter and Levenberg-Marquardt Least Squares. The Kalman Filter needs dynamic model of the object to predict trajectory, different from The Levenberg-Marquardt method requires previous movements of the object, and the system can determine the next move, to perform this experiment a simulation of the motion was created that an object moving in 3D fields using MATLAB, the movement is also added random noise using normal probability to give real effect, where in fact, there are noise sensors that affect the performance of a system. An experiment proposed to test the speed of each method achieving convergent, and the endurance of each method to receive noise.

Comparison of Indonesian speaker recognition using Vector Quantization and Hidden Markov Model for unclear pronunciation problem

This paper presents a comparison of two classifier methods based on accuracy level in Indonesian speaker recognition for unclear pronunciation problem in a word, simple sentences, and complete sentences. The first method is Vector Quantization (VQ) based on distortion distance and the second method is Hidden Markov Model (HMM) based on the probability value of the data is observed. Based on the experiments, It can be concluded that HMM method have better accuracy than VQ method especially for pronunciation of simple sentences.

Design and implementation of walking pattern and trajectory compensator of NAO humanoid robot

Humanoid robot is a dynamic robot which can walk with two legs like a human. Many researchers are interested in this class of robot due to its flexibility in reaching a variety of terrain and also its similarity to humans. The main problem often faced by developers is the motion control system which is rather nontrivial. Having 25 DOF with only two legs, it is difficult to model the robot movement that achievesstatic and dynamic balance. To facilitate the developer, there needs to be a model that can be used as a foundation for the development of robot motion system. This model allows us to obtain the information which needed in the formation of the robot posture. Information such as joint position, position of the center of mass, ZMP position, support polygon and load torque of each joint can affect the balance of the formed posture. In addition, with the establishment of some proper posture, it is possible to determine several sequential postures that generate a walking pattern. The walking pattern enablesthe humanoid robot to move to another place. However, in reality, this robot movement often has a shifting trajectory from the target location due to various reasons. A Neuro-Fuzzy system has been designed and implemented to compensate error in trajectory tracking. This system processes the trajectory shift by fuzzy logic (linguistic) combined with learning algorithms that allows the system to change its own structures to produce better control output. Smaller shift on trajectory is achieved as the number of learning iteration increases.

Scenes categorization based on appears objects probability

Automation of visual perception on the machine is the automation of the interpretation of the data generated by the camera sensor like humans. Some automated systems such as the robot and intelligent vehicle relies heavily on an understanding of the environment. The most fundamental understanding of the visual environment is the Machines ability to identify categories of scenery is still a lot of uncertainty. The uncertainty of the categorization of the scene appears because of the difficulty categories indicated in complex environments, the layout of objects and different scene in the same category. We present a model approach Distribution of the chance appearance of objects and classify them to obtain the semantic meaning of a scene. From the experiments conducted, the model shows high accuracy and using a dataset SUN908 this approach is an effective approach to explore the knowledge-based on the scenes label using large dataset.

Opponent zigzag movement model capture and prediction in robotic soccer

Robotic soccer research as a standard problem to foster Artificial Intelligent research in robotics and related fields still needs improvement in game play modeling. Action anticipation decision could be based on opponent game play model. Artificial Neural Network has been used to capture game play model, but still has problems of slow learning convergence and inaccuracy. Artificial Neural Network algorithm need to be a small, fast, and robust application for small devices of robot soccer. In this paper, experiments of model capturing simulated zigzag soccer movement has been done using Radial Basis Function Neural Network, Inverse Function Neural Network, pseudo-inverse Singular Value Decomposition, existing and modified Simurosot robot soccer prediction algorithm. Model capture results show, despite higher error in zigzag movement model capture, the Inverse Function Neural Network and the modified Simurosot prediction algorithm yield lower error in movement prediction.

Educational game design using the 7 steps for designing serious games method (Case study: Mathematical subject on comparison and scale material for 7th grade junior high school)

National educational system was riddled with many problems both in terms of internal and external to date. Internal issues referring to the constraints of some national education standards among which the standard process, teachers and education personnel standards, and competency standards. The standard process has a problem that improvement mindset has not done well in the learning process of the learning method centered on the teacher (teacher centered) into a student-centered teaching methods (student oriented)[17]. Standard teachers and education personnel have a problem of low professionalism of teachers in Indonesia proved in 2010 the number of teachers who meet the qualification standards nationally only 43%[2]-Standardized competence of graduates have problems one of which is evidenced by the results of the 2012 survey data by the Program for International Student Assessment (PISA), which mentions the mathematical ability of students in Indonesia was ranked 64 out of 65 countries[1]. In addition to internal problems are also external problems in the organization of education as the result of research that says that as many as 600 students in grade eight and nine on average spend 9 hours per week playing games keseluruhan. Playing games have proven to deliver negative correlation with academic achievement of students, if played in a long period of time[5]. In this digital era, it is appropriate that educational games have a more comprehensive capabilities for responding the internal and external challenges of the world of education. Therefore the authors undertake the design and construction of educational games vidyanusa mission 19th to study the materials of scale and comparison using The 7 Steps for Designing Serious Games method to resolve some of the problems of national educational standards. This method begins with the analysis specifications pedagogical objectives to be achieved, the selection of model type game that will be used, the adjustment of pedagogical scenario with games scenario are fun, the search of software components that can be used, description detail of scenarios, the quality control of pedagogical and determination of specifications for subcontractors and game design tools[10].