This article describes the system architecture of KUVE (KIST Unmanned Vehicle Electric) and unmanned autonomous navigation of it in KIST. KUVE, which is an electric light-duty vehicle, is equipped with two laser range finders, a vision camera, a differential GPS system, an inertial measurement unit, odometers, and control computers for autonomous navigation. KUVE estimates and tracks the boundary of road such as curb and line using a laser range finder and a vision camera. It follows predetermined trajectory if there is no detectable boundary of road using the DGPS, IMU, and odometers. KUVE has over 80% of success rate of autonomous navigation in KIST.

This paper proposes lane recognition algorithm using lane prominence technique to extract lane candidate. The lane prominence technique is combined with embossing effect, lane thickness check, and lane extraction using mask. The proposed lane recognition algorithm consists of preprocessing, lane candidate extraction and lane recognition. First, preprocessing is executed, which includes gray image acquisition, inverse perspective transform and gaussian blur. Second, lane candidate is extracted by using lane prominence technique. Finally, lane is recognized by using hough transform and least square method. To evaluate the proposed lane recognition algorithm, this algorithm was applied to the detection of lanes in the rainy and night day. The experiment results showed that the proposed algorithm can recognize lane in various environment. It means that the algorithm can be applied to lane recognition to drive unmanned vehicles.

This paper presents an optimum tire force distribution method for 6WD/6WS(6-Wheel-Drive and 6-Wheel-Steering) electric vehicles. Using an independent steering and driving system, the performance of 6WD/6WS vehicles can be improved, as, for example, with respect to their maneuverability under low speed and their stability at high speed. Therefore, there should be a control strategy for finding the optimum tire forces that satisfy the driver's command and minimize energy consumption. From the driver's commands (steering angle and accelerator/brake pedal stroke), the desired yaw moment, the desired lateral force, and the desired longitudinal force were obtained. These three values were distributed to each wheel as the torque and the steering angle, based on the optimum tire force distribution method. The optimum tire force distribution method finds the longitudinal/lateral tire forces of each wheel that minimize the cost function, which is the sum of the normalized tire forces. Next, the longitudinal/lateral tire forces of each wheel are converted into the reference torque inputs and the steering wheel angle inputs. The proposed method was tested through a simulation, and its effectiveness was verified.

This paper presents the sensor fusion algorithm that recognizes a primary vehicle by fusing radar and monocular vision data. In general, most of commercial radars may lose tracking of the primary vehicle, i.e., the closest preceding vehicle in the same lane, when it stops or goes with other preceding vehicles in the adjacent lane with similar velocity and range. In order to improve the performance degradation of radar, vehicle detection information from vision sensor and path prediction predicted by ego vehicle sensors will be combined for target classification. Then, the target classification will work with probabilistic association filters to track a primary vehicle. Finally the performance of the proposed sensor fusion algorithm is validated using field test data on highway.

This paper describes an integrated chassis control for a maneuverability, a lateral stability and a rollover prevention of a vehicle by the using of the ESC and AFS. The integrated chassis control system consists of a supervisor, control algorithms and a coordinator. From the measured and estimation signals, the supervisor determines the vehicle driving situation about the lateral stability and rollover prevention. The control algorithms determine a desired yaw moment for lateral stability and a desired longitudinal force for the rollover prevention. In order to apply the control inputs, the coordinator determines a brake and active front steering inputs optimally based on the current status of the subject vehicle. To improve the reliability and to reduce the operating load of the proposed control algorithms, a multi-core ECU platform is used in this system. For the evaluation of this system, a closed loop simulations with driver-vehicle-controller system were conducted to investigate the performance of the proposed control strategy.

As increasing the number of Electronic Control Units in a vehicle, the proportion for reliability and stability of the software is going increasingly. Accordingly, the traditional CAN network has occurred the situation that the requirement of developing vehicle software is not sufficient. To solve these problems, the FlexRay network which is ensured the high bandwidth and real-time is generated. However it is difficult to implement FlexRay based application software because of complex protocol than traditional CAN network. Accordingly the system for analysis and verification of network state is needed. Also vehicle vendor develops application software using Matlab/Simulink in order to increase productivity. But this development method is hard to solve the network problem of node to node. Therefore this paper implements Matlab/Simulink based FlexRay network system and verifies it through comparing with existing embedded system.

This paper presents a failure diagnosis and accommodation strategy which is capable of tolerating faulty actuators of a submersible vehicle. The proposed method is mainly based on a sliding mode control technique. The primary ideas include a performance index to describe the effectiveness of actuators, and a controller reconfiguration strategy using the actuator effectiveness index. The actuator effectiveness proposed in this work is defined as the relationship between the sliding surface and the controlled system behavior. The resulting actuator effectiveness is then used in reconfiguring the controller in order to counteract for the deteriorated control performance in the presence of a faulty actuator. The effectiveness of the proposed method is demonstrated by means of numerical simulations with a submersible vehicle.

Given a periodic reference signal or disturbance, repetitive control is a special control scheme to reduce a tracking error effectively by the periodic signal generator in the repetitive controller. In general, a repetitive controller is added on the existing feedback control system to improve the tracking performance. However, because the information used in the design of the feedback controller is not taken into account, the design problem of the repetitive controller is totally another problem irrespective of that of the feedback controller. In this paper, we present a more general method to design an add-on type repetitive controller using the information on the performance of the existing feedback control system. We first show that a robust stability condition of repetitive control systems is obtained using the well-known robust performance condition of general feedback control systems. It is also shown that we can obtain a steady-state tracking error described in a simple form without time-delay element if the robust stability condition is satisfied for the repetitive control system. From the obtained results, several design criterions for repetitive controller are provided. Through the simulation study, the feasibility of the proposed method is verified.

It is known that many insects and animals can return to their nest after exploration, with their own specific homing mechanisms. Their homing navigation methods have been applied to the robotic navigation. In this paper, we test the sector-based image matching method motivated by the honeybee's landmark navigation behaviour. Here, our robotic approach uses the reference compass to identify the current head direction and the relative angular position of landmarks for the navigation. The robot shows desirable homing behaviors if the robot is surrounded by landmarks. The result of robot experiment is in good agreement with that of simulation.

In this paper, an implementation of force control for a slave mobile robot in tele-operation environment is presented. A mobile robot is built to have a force control capability with a force sensor and tested for force tracking control performances. Both position and contact force are regulated by a PID based hybrid control method and the impedance force control method. To minimize accumulated errors due to the adaptive impedance force control method, the novel force control method with a weighted function is proposed. Experimental studies of regulating contact forces for different control algorithms are tested and their performances are compared.

This paper presents a gravity compensator for the manipulator of a service robot. The manipulator of a service robot is operated with low velocity for the safety reason in most cases. In this situation gravitational torques generated by the mass of links are often much greater than dynamic torques for motion. A gravity compensator can counterbalance the gravitational torques, thereby enabling to utilize relatively low power motors. In this paper the gravity compensation for the roll-pitch rotation is considered which is often used for the shoulder joints of the manipulator of a service robot or humanoid robot. A gimbals is implemented and two 1-dof gravity compensators are equipped at the base. One compensates the gravitational torque at the roll joint and another provides the compensational torque for the gimbals. Various analyses showed that the proposed compensator can counterbalance the gravitational torques of 87% at the pitch joint and 50% at the roll joint. It is verified from dynamic simulations that the proposed compensator effectively counterbalances the gravitational torques.

This paper presents an image alignment algorithm for application of AOI (Automatic Optical Inspection) based on SIFT. Since the correspondences result using SIFT descriptor have many wrong points for aligning, this paper modified and classified those points by five measures called the CCFMR (Cascade Classifier for False Matching Reduction) After reduced the false matching, rotation and translation are estimated by point selection method. Experimental results show that the proposed method has fewer fail matching in comparison to commercial software MIL 8.0, and specially, less than twice with the well-controlled environment’s data sets (such as AOI system). The rotation and translation accuracy is robust than MIL in the noise data sets, but the errors are higher than in a rotation variation data sets although that also meaningful result in the practical system. In addition to, the computational time consumed by the proposed method is four times shorter than that by MIL which increases linearly according to noise.

In this paper, an overload detecting algorithm for an excavator is presented. The proposed overload detecting algorithm is based on the time series analysis especially correlation function. The main purpose of this paper is to prevent damage or crack from the fatigue loaded on an excavator in advance. Generally, the larger data, the longer processing time, and the amount of the data used in this paper are also large, especially every sampling period, 1600 data are gathered and calculated. So this paper focuses on minimizing the number of required sensors by using the correlation function. From the cross correlation function, similar pattern sensors are eliminated and dissimilar pattern sensors are considered, and from the auto correlation function, the overload can be detected. To prove the efficiency of the proposed overload detecting algorithm, this paper shows the computer simulation results.

Heliostat, as a concentrator reflecting the incident solar energy to the receiver located at the tower, is the most important system in the tower-type solar thermal power plant, since it determines the efficiency and performance of solar thermal plower plant. Thus, a good sun tracking ability as well as its good optical property are required. In this paper, we propose a method to compensate the heliostat sun tracking error. We first model the sun tracking error, which could be measured using BCS (Beam Characterization System), by multilayered neural network. Then the extended Kalman filter was employed to train the neural network. Finally the model is used to compensate the sun tracking errors. Simulated result shows that the method proposed in this paper improve the heliostat sun tracking performance dramatically. It also shows that the training of neural network by the extended Kalman filter provides faster convergence property, more accurate estimation and higher measurement noise rejection ability compared with the other training methods like gradient descent method.

In AGPS method, user position can be obtained even in the shadow region by improving signal sensitivity. A hybrid long integration scheme employing both coherent and non-coherent integration method is commonly used in AGPS receivers. Because coherent loss increases as residual frequency become large, residual frequency should be minimized to maximize coherent integration gain. This paper presents performance analysis of residual frequency estimator using FFT in fine-time assistance AGPS method. Considering the hardware complexity and the estimation accuracy, optimal length of FFT is proposed for GPS L1 C/A signal. Signal sensitivity for estimating the residual frequency is also analysed. By field experimental results, it is found that the residual frequency can be successfully estimated using 1 second snap-shot data when GPS signal strength is larger than -150 dBm and its RMS error is 3Hz.