This paper deals with the design of robust DSMC (Discrete-Time Sliding Mode Control) scheme in order to overcome system uncertainty in steering system with mechanically joined structure. The proposed control scheme is one of robust control schemes based on system dynamics. Therefore, system dynamics required is not obtained from physical law but SCM (Signal Compression Method) through experiment in order to avoid complicate mathematical development and save time. However, SCM has a shortcoming that is the limitation of with $2^{nd}$ order linear model which does not include the dynamic of high-frequency band. Thus, considering system uncertainty, DSMC is designed. In addition, to reduce the chattering problem of DSMC, DSMC is derived from the reaching law and the Lyapunov stability condition. It is found that the proposed control scheme has robustness in spite of the perturbation of system uncertainty through computer simulation.

Real-time systems perform periodic tasks and real-time aperiodic tasks such as alarm processing. Especially the periodic tasks included in control systems such as robots have precedence relationships among them. This paper proposes a new scheduling algorithm based on topological sort and residual time. The precedence relationships among periodic tasks are translated to the priorities of the tasks using topological sort algorithm. During the execution of the system the proposed scheduling algorithm decides on whether or not a newly arrived real-time aperiodic task is accepted based on residual time whenever the aperiodic task such as alarm is arrived. The proposed algorithm is validated using examples.

Robust tuning rules of the motion profile are proposed to minimize the residual vibration. For asymmetric S-curve profile, tuning rules are analytically formulated using Laplace-domain approach. When the system modeling is known exactly, by placing a single zero of the motion profile on the pole of the system, the residual vibration can be perfectly eliminated under undamped system. However, if there are some amounts of the modeling errors, the residual vibration significantly increases. To track this issue, the robust tuning rules against modeling error are discussed. One of the proposed robust tuning rules is placing the multiple zeros of the motion profile on the pole of the system, and the other is placing the zeros of the motion profile around the pole of the system. Thanks to the proposed robust tuning rules, motion profile becomes more robust to modeling errors while minimizing the residual vibration. By simulation, the effectiveness of the proposed robust tuning rules is verified.

This paper proposes a new robust motion deblurring filter using the inertial sensor measurements for strapdown image IR applications. With taking the PSF measurement error into account, the motion blurred image is modeled by the linear uncertain state space equation with the noise corrupted measurement matrix and the stochastic parameter uncertainty. This motivates us to solve the motion deblurring problem based on the recently developed robust least squares estimation theory. In order to suppress the ringing effect on the deblurred image, the robust least squares estimator is slightly modified by adoping the ridge-regression concept. Through the computer simulations using the actual IR scenes, it is demonstrated that the proposed algorithm shows superior and reliable motion deblurring performance even in the presence of time-varying motion artifact.

This paper presents a comparative analysis on three different types of performance indices of obstacle detection for an overlapped ultrasonic sensor ring. Due to beam overlap, the entire sensing zone of each ultrasonic sensor can be divided into three smaller sensing subzones, which leads to significant reduction of positional uncertainty in obstacle detection. First, the positional uncertainty in obstacle detection is expressed in terms of the area of a sensing subzone, and type 1 performance index is then defined as the area ratio of side and center sensing subzones. Second, based on the area of a sensing subzone, type 2 performance index is defined taking into account the size of the entire range of obstacle detection as well as the degree of the positional uncertainty in obstacle detection. Third, the positional uncertainty in obstacle detection is now expressed in terms of the length of the uncertainty arc spanning a sensing subzone, and type 3 performance index is then defined as the average value of the uncertainty arc lengths over the entire range of obstacle detection. Fourth, using a commercial low directivity ultrasonic sensor, the changes of three different performance indices depending on the parameter of an overlapped ultrasonic sensor ring are examined and compared.

In this paper, we propose a robust fingertip extraction and extended Continuously Adaptive Mean Shift (CAMSHIFT) based robust hand gesture recognition for natural human-like HRI (Human-Robot Interaction). Firstly, for efficient and rapid hand detection, the hand candidate regions are segmented by the combination with robust $YC_bC_r$ skin color model and haar-like features based adaboost. Using the extracted hand candidate regions, we estimate the palm region and fingertip position from distance transformation based voting and geometrical feature of hands. From the hand orientation and palm center position, we find the optimal fingertip position and its orientation. Then using extended CAMSHIFT, we reliably track the 2D hand gesture trajectory with extracted fingertip. Finally, we applied the conditional density propagation (CONDENSATION) to recognize the pre-defined temporal motion trajectories. Experimental results show that the proposed algorithm not only rapidly extracts the hand region with accurately extracted fingertip and its angle but also robustly tracks the hand under different illumination, size and rotation conditions. Using these results, we successfully recognize the multiple hand gestures.

This paper presents performance evaluation of real-time mechanisms for real-time embedded linux. First, we presents process for implementing open-source real-time embedded linux namely RTAI and Xenomai. These are real-time extensions to linux kernel and we implemented real-time embedded linux over the latest linux kernel. Measurements of executions of real-time mechanisms for each distribution are performed to give a quantitative comparison. Performance evaluations are conducted in kernel space about repeatability of periodic task, response time of Semaphore, FIFO, Mailbox and Message queue in terms of inter-task communication for each distribution. These rules can be helpful for deciding which real-time linux extension should be used with respect to the requirements of the real-time applications.

To follow a sound source by a mobile robot, the relative position and orientation of the sound source from the mobile robot have been estimated using a microphone array. In this research, the difference among the traveling times of the sound source to each of three microphones has been used to calculate the distance and orientation of the sound source from the mobile robot which carries the microphone array. The cross-correlation between two signals has been applied for detecting the time difference between two signals, which provides reliable and precise value of the time difference comparing to the conventional methods. To generate the tracking direction to the sound source, fuzzy rules are applied and the results are used to control the mobile robot in a real-time. The efficiency of the proposed algorithm has been demonstrated through the real experiments comparing to the conventional approaches.

Human detection is a challenging task in many fields because it is difficult to detect humans due to their variable appearance and posture. Furthermore, it is also hard to track the detected human because of their dynamic and unpredictable behavior. The evaluation speed of method is also important as well as its accuracy. In this paper, we propose unified detection and tracking method for humans using Gaussian-PSO (Gaussian Particle Swarm Optimization) with the HOG (Histograms of Oriented Gradients) features to achieve a fast and accurate performance. Keeping the robustness of HOG features on human detection, we raise the process speed in detection and tracking so that it can be used for real-time applications. These advantages are given by a simple process which needs just one linear-SVM classifier with HOG features and Gaussian-PSO procedure for the both of detection and tracking.

Real-Time Traversability should be analyzed from the equiped sensors' data in real time for autonomous outdoor navigation. However, it is difficult to find out such traversability that considers the terrain roughness and the vehicle dynamics especially in case of skid type vehicle. The traversability based on real time dynamic analysis was proposed to solve such problem but in navigation with strait driving path. To adapt the method into the navigation with curved driving path, a path following controller should be incorporated into the dynamic model even though it cause the real time problem. In this paper, a dynamic model is proposed to solve the real time problem in the traversability analysis based on real time dynamic simualtion. The dynamic model contains the control dummy which is connected to the vehicle body with a universal joint to follow the curved path without controller. Simulation and experimental results on $6{\times}6$ articulated unmanned ground vehicle demonstrate the method's effectiveness and applicability into the traversability analysis on terrain with bumps.

Most path planning algorithms for a marine robot in the ocean environment have been developed without considering the robot's heading angle. As a result, the robot has a difficulty in following the path correctly. In this paper, we propose a limit-cycle circle set that applies to the $Theta^*$ algorithm. The minimum turning radius of a marine robot is calculated using a limit-cycle circle set, and circles of this radius is used to generate a configuration space of an occupancy grid map. After applying $Theta^*$ to this configuration space, the limit-cycle circle set is also applied to the start and end nodes to find the appropriate path with specified heading angles. The benefit of this algorithm is its fast computation time compared to other 3-D ($x,y,{\theta}$) path planning algorithms, along with the fact that it can be applied to the 3-D kinematic state of the robot. We simulate the proposed algorithm and compare it with 3-D $A^*$ and 3-D $A^*$ with post smoothing algorithms.

DC/DC switching power converters produce DC output voltages from different stable DC input sources regulated by a bi-polar transistor. The converters can be used in regenerative braking of DC motors to return energy back in the supply, resulting in energy savings for the systems containing frequent stops. The voltage mode DC/DC converter is composed of a PWM (Pulse Width Modulation) controller, a MOSFET (Metal Oxide Semiconductor Field Effect Transistor), an inductor, and capacitors, etc. PWM is applied to control and regulate the total output voltage. It is shown that the output of DC/DC converter depends on the variation of threshold voltage at MOSFET and the variation of pulse width. In the PWM operation, the missing pulses, the changes in pulse width, and a change in the period of the output waveform are studied by SPICE (Simulation Program with Integrated Circuit Emphasis) and experiments.

Together with the GPS-based approach, geolocation through mobile communication networks is a key technology for location-based service. Since the Mobile WiMAX system is considered as a candidate for fourth-generation mobile systems, it is important to investigate its location capability. The geolocation of Mobile WiMAX can be realized when the preamble symbols in the down-link channel are appropriately used for a TDOA (Time-Difference-of-Arrival) approach. However, the cellular structure of Mobile WiMAX inevitably generates co-channel interference, and it is difficult for the mobile terminal to acquire distance measurements from multiple base stations. Therefore, for geolocation via multilateration using the Mobile WiMAX network, it is very important to increase hearability. This paper proposes a geolocation method for Mobile WiMAX which employs interference cancellation and preamble signal overlapping for the enhancement of hearability. A novel interference cancellation strategy for complex-valued Mobile WiMAX signals is presented which has an iterative structure. Simulation results show that the proposed geolocation method provides the user's position with an accuracy of less than 20 m through the Mobile WiMAX cellular network if there is no multi-path or NLOS (None-Line-of-Sight).

In this paper, we analyzed the navigation error of TERCOM (TErrain COntour Matching), which is TRN (Terrain Referenced Navigation) batch processing, caused by scale factor error of radar altimeter and proved the possibility of false position fix when we use the TERCOM's feature matching algorithm. Based on these, we proposed the new TRN batch processing algorithm using the slope measurements of terrain. The proposed technique measures on periodic changes in the slope of the terrain elevation profile, and these measurements are used in the feature matching algorithm. By using the slope of terrain data, the impact of scale factor errors can be compensated. By simulation, we verified improved outcome using this approach compared to the result using the conventional method.

Multipath remains a dominant source of ranging errors in GNSS (Global Navigation Satellite System). And it is generally considered undesirable in the context of GNSS, since the reception of multipath can make significant distortion to the shape of the correlation function. In this paper, therefore, the model of the distorted shape of the correlation function is formulated and a MBME (Model-Based Multipath Estimation) technique for GPS L1/L5 receivers is proposed in order to estimate the parameters of the indirect signal such as the amplitude and the delay. The MBME technique does not require the any hardware modifications and it can estimate the parameters for both the short and long-delay multipath. Especially, it would be the very effective technique for the short-delay multipath if the L5 signal is available. Finally, the feasibility of the proposed technique has been confirmed by simulation results.