• Journal of Institute of Control, Robotics and Systems
• /
• v.19 no.8
• /
• pp.718-724
• /
• 2013

In today's automotive industry, there exist several systems that help drivers reduce the possibility of accidents, such as the ADAS (Advanced Driver Assistance System). The ADAS helps drivers make correct and quick decisions during dangerous situations. This study analyzed the performance of the IMM (Interacting Multiple Model) method based on multiple Kalman filters using the data acquired from a driving simulator. An IMM algorithm is developed to identify the current discrete state of neighboring vehicles using the sensor data and the vehicle dynamics. In particular, the driving modes of the neighboring vehicles are classified by the cruising and maneuvering modes, and the transition between the states is modeled using a Markovian switching coefficient. The performance of the IMM algorithm is analyzed through realistic simulations where a target vehicle executes sudden lane change or acceleration maneuver.

• Journal of Advanced Navigation Technology
• /
• v.3 no.2
• /
• pp.156-166
• /
• 1999

In multi-vehicle tracking systems Kalman filter is generally used for tracking vehicles. Despite well known advantages of Kalman filter that presents optimality with constraints, it is difficult to track several vehicles in real time simultaneously due to a large number of computations. In this paper, we propose a multi-vehicle tracking system with an adaptive predictor that employs recursive least square algorithm which can be easily implemented for real time application on a transversal filter. The performance of the proposed tracking system is compared to one with Kalman filter using a synthetic sequential image generated by computer graphics and real sequential image taken at intersections. Simulation results show that the proposed tracking system can be applied to track vehicles in real sequential image at the rate of 30 frame/sec on a PC environments without any special hardwares.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.8
• /
• pp.633-642
• /
• 2016

It is important for advanced active safety systems and autonomous driving cars to get the accurate estimates of the nearby vehicles in order to increase their safety and performance. This paper proposes a sensor fusion method for radar and vision sensors to accurately estimate the state of the preceding vehicles. In particular, we performed a study on compensating for the lateral state error on automotive radar sensors by using a vision sensor. The proposed method is based on the Interactive Multiple Model(IMM) algorithm, which stochastically integrates the multiple Kalman Filters with the multiple models depending on lateral-compensation mode and radar-single sensor mode. In addition, a Probabilistic Data Association Filter(PDAF) is utilized as a data association method to improve the reliability of the estimates under a cluttered radar environment. A two-step correction method is used in the Kalman filter, which efficiently associates both the radar and vision measurements into single state estimates. Finally, the proposed method is validated through off-line simulations using measurements obtained from a field test in an actual road environment.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.54 no.12
• /
• pp.689-697
• /
• 2005

In this paper, an application of a decentralized $H_{\infty}$ controller(DHC) to multiple controlled-permanent magnet(CMAG) magnetic levitation(Maglev) systems is presented. The designed DHC using two Riccati equations iteratively has simpler structure and needs less computational loads than conventional centralized $H_{\infty}$ controller. A target plant is a hybrid-type CMAG system with permanent magnet and coil, and its mathematical model is firstly derived to design the DHC. To implement the designed algorithm, a real Maglev vehicle system including digital controller, chopper, sensor, etc., is manufactured. To compare the performances of the DHC method with an observer-based state feedback control(OSFC), the input tracking and disturbance rejection characteristics are experimentally tested. As performance indices(PI), integral of squared error(ISE), integral of absolute error(IAE), integral of time multiplied by absolute error(ITAE) and integral of time multiplied by squared error(ITSE) are used. From the experimental results, it can be seen that the input tracking and disturbance rejection performances of the DHC are better than those of the conventional controller.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.2 no.2 s.3
• /
• pp.85-95
• /
• 2003

This paper proposes a multiple moving objects tracking system which is adaptable itself to circumstances. Snake model is sensitive to the start position value because it does not accurately express contours of objects in complex image. It can be improved as the proposed system gets background images by using difference images, segments objects using neighborhood pixels and assesses the position feature values acquired on the start position value to deformable Snake model. And also the system can simplify complex background images and reduce search regions by the constituent points of a Snake laid in Positions of object. It is showed that the proposed system can be appBied to multiple moving vehicle racking systems by the experimental results of 30fps AVI file.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.1
• /
• pp.11-20
• /
• 2000

This paper presents the method of extracting the contour and shape parameters for moving object tracking in traffic scenes. The contour is extracted by applying difference image method in reduction image and the features are extracted from original image to grow the accuracy of tracking. We used features such as circle distribution, center moment, and maximum and minimum ratio. Data association problem is solved by these features. Kalman filters are used for moving object tracking on real time. The simulation results indicate that the proposed algorithm appears to generate feature vectors good enough for multiple vehicle tracking.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.5
• /
• pp.641-647
• /
• 2009

In this paper, a new filtering scheme for adaptive cruise control (ACC) system is presented. In the proposed scheme, the identification of the mode of the preceding vehicle is considered as a classification problem and it is done by a neural network classifier. The neural network classifier outputs a posterior probability of the mode of the preceding vehicle and the probability is directly used in the IMM framework. Finally, ten scenarios are made and the proposed NIMM is tested on them to show its validity.

• Transactions of The Korea Fluid Power Systems Society
• /
• v.5 no.3
• /
• pp.15-22
• /
• 2008

The road simulators have become common tools within the automotive industry for evaluation of vehicle and vehicle system durability performance. These simulators need appropriate input signal generation algorithms to realize the actual driving conditions due to non-linear vehicle and test rig behaviour. Although somewhat unconventional from a control standpoint, the iteration approach has proven to be a very effective method for control of complex, multiple degree-of-freedom systems where the tracking parameter is known a priori. In this paper, the road profile replication algorithm is verified by applying Belgian road to the developed road simulator. The simulation and experimental results are included to evaluate the performance of this simulator. This road simulator provides considerable savings in cost, development time, and testing risk during developing automotive components.

• Journal of Internet Computing and Services
• /
• v.12 no.6
• /
• pp.73-82
• /
• 2011

The location-based service is emerging again to the public attention. The location recognition environment up-to-now has been studied with its focus only on a person, an object or a moving object. However, this study proposes a location recognition model that serves to recognize and track, in real time, multiple passengers in a moving vehicle. Identifying the locations of passengers can be classified into two classes: one is to use the high price terminal with GPS function, and the other is to use the economic price compact terminal without GPS function. Our model enables the simple compact terminal to provide effective location recognition under the on-boarding situation by transmitting messages through an interface device and sensor networks for a vehicle equipped with GPS. This technology reduces transmission traffic after detecting the condition of a vehicle (being parked or running), because it does not require transmission/receiving of information on the locations of passengers who are confined in a vehicle when the vehicle is running. Also it extends battery life by saving power consumption of the compact terminal. Hence, we carried out experiments to verify its serviceability by materializing the efficient tracking system for passengers with the detection algorithm of a stopping vehicle proposed in this study. Moreover, about 200 experiments using the system designed with this technology proved successful recognition on on-boarding and alighting of passengers with the maximum transmission distance of 12 km. In addition to this, the running recognition tests showed the test with the detection algorithm of a stopping vehicle has reduced transmission traffic by 41.6% compared to the algorithm without our model.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.22 no.3
• /
• pp.362-370
• /
• 2011

This paper presents a K-band radar system installed on the ground moving vehicle to detect and track a high-speed target. The presented radar is separated into three search regions to satisfy a wide area detection and a limitation of the installing space of the radar, and each region performs detecting the target independently and tracking the detected target automatically. The presented radar radiating K-band FMCW waveform acquires range and velocity information of the target at the every dwell and receiving antenna of the radar is applied the multiple baseline interferometer to extract the precise angle information of the target. 3-dimensional tracking accuracy of the radar is 0.25 m RMSE measured actually through a fire experiment of an imitation target.