• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.1
• /
• pp.16-21
• /
• 2003

The accuracy in maneuvering target tracking using multiple models is resulted in by the suitability of each target motion model to be used. The interacting multiple model (IMM) method and the adaptive IMM (AIMM) method require the predefined sub-models and the predetermined acceleration intervals, respectively, in consideration of the properties of maneuvers in order to construct multiple models. In this paper, to solve these problems, a genetic algorithm(GA) based-IMM method using fuzzy logic is proposed. In the proposed method, the acceleration input is regarded as an additive noise and a sub-model is represented as a set of fuzzy rules to calculate the time-varying variances of the process noises of a new piecewise constant white acceleration model. The proposed method is compared with the AIMM algorithm in simulation.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2011.06a
• /
• pp.335-336
• /
• 2011

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.6
• /
• pp.497-502
• /
• 2002

The problem of maneuvering target tracking has been studied in the field of the state estimation over decades. The Kalman filter has been widely used to estimate the state of the target, but in the presence of a maneuver, its performance may be seriously degraded. In this paper, to solve this problem and track a maneuvering target effectively, a DNA coding-based interacting multiple model (DNA coding-based W) method is proposed. The proposed method can overcome the mathematical limits of conventional methods by using the fuzzy logic based on DNA coding method. The tracking performance of the proposed method is compared with those of the adaptive IMM algorithm and the GA-based IMM method in computer simulations.

• Journal of Advanced Navigation Technology
• /
• v.15 no.4
• /
• pp.487-494
• /
• 2011

Data Fusion algorithm is essential in Target Detection using radar, and it has more reliability. In this paper, Multi Radar Fusion algorithm using IMM(Interacting Multiple Model) filter is suggested. This well-known IMM filter has better performance than Kalman filter has. In this simulation, Distributed Data Fusion process was applied, and three sub-filters and one main filter were employed. In addition, this simulation was evaluated by virtual radar data which include constant velocity, constant accelerate, turn rate. The result of an evaluation shows better performance in the maneuvering section of aircraft.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.8
• /
• pp.71-78
• /
• 2006

When a tracking filter may be designed in the Cartesian coordinate, the covariance of the measurement errors varies according to the range and the bearing of an interested target. In this paper, interacting multiple model based tracking filter is formulated in the Cartesian coordinate utilizing the analytic solution of the steady state Kalman filter, which can be able to consider the variation of the measurement error covariance. 100 Monte Carlo runs performed to verify the proposed method. The performance of the proposed method is compared with the conventional fixed gain and Kalman filter based IMM tracking filter in terms of the root mean square error. The simulation results show that the proposed approach meaningfully reduces the computation time and provides a similar tracking performance in comparison with the conventional Kalman filter based IMM tracking filter.

• 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.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.1
• /
• pp.93-100
• /
• 2003

Accuracy in maneuvering target tracking using multiple models relies upon the suit-ability of each target motion model to be used. To construct multiple models, the interacting multiple model (IMM) algorithm and the adaptive IMM (AIMM) algorithm require predefined sub-models and predetermined acceleration intervals, respectively, in consideration of the properties of maneuvers. To solve these problems, this paper proposes the GA-based IMM method as an intelligent tracking method for a maneuvering target. In the proposed method, the acceleration input is regarded as an additive process noise, a sub-model is represented as a fuzzy system to compute the time-varying variance of the overall process noise, and, to optimize the employed fuzzy system, the genetic algorithm (GA) is utilized. The simulation results show that the proposed method has a better tracking performance than the AIMM algorithm.

• Journal of Mechanical Science and Technology
• /
• v.19 no.9
• /
• pp.1742-1752
• /
• 2005

This paper presents a Multiple Target Tracking (MTT) Adaptive Cruise Control (ACC) system which consists of three parts; a multi-model-based multi-target state estimator, a primary vehicular target determination algorithm, and a single-target adaptive cruise control algorithm. Three motion models, which are validated using simulated and experimental data, are adopted to distinguish large lateral motions from longitudinally excited motions. The improvement in the state estimation performance when using three models is verified in target tracking simulations. However, the performance and safety benefits of a multi-model-based MTT-ACC system is investigated via simulations using real driving radar sensor data. The MTT-ACC system is tested under lane changing situations to examine how much the system performance is improved when multiple models are incorporated. Simulation results show system response that is more realistic and reflective of actual human driving behavior.

• Journal of Advanced Navigation Technology
• /
• v.24 no.3
• /
• pp.198-204
• /
• 2020

This paper designs a electromagnetic-log (EM-Log) aided navigation filter for maritime environment without global navigation satellite system (GNSS). When navigation is performed for a long time, Inertial navigation system (INS)'s error gradually diverges. Therefore, an integrated navigation method is used to solve this problem. EM-Log sensor measures the velocity of the vehicle. However, since the measured velocity from EM-Log contains the speed of the sea current, the aided navigation filter is required to estimate the sea current. This paper proposes a single model filter and interacting multiple (IMM) model filter methods to estimate the sea current and analyzes the influence of the sea current model on the filter. The performance of the designed aided navigation filter is verified using a simulation and the improvement rate of the filter compared to the pure navigation is analyzed. The performance of single model filter is improved when the sea current model is correct. However, when the sea current model is incorrect, the performance decreases. On the other hands, IMM model filter methods show the stable performance compared to the single model.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2004.04a
• /
• pp.459-464
• /
• 2004

In this paper, an autonomous navigation system for autonomous guided vehicles (AGVs) operated in an automated container terminal is designed. The navigation system is based on the sensors detecting the range and bearing. The navigation algorithm used is an interacting multiple model (IMM) algorithm to detect other AGVs and avoid other obstacles using informations obtained from multiple sensors. As models to detect other AGVs (or obstacles), two kinematic models are derived： Constant velocity model for linear motion and constant speed turn model for curvilinear motion. For constant speed turn model, an unscented Kalman filter (UKF) is used because of drawbacks of the extended Kalman filter (EKF) in nonlinear system. The suggested algorithm reduces the root mean squares error for linear motions, while it can rapidly detect possible turning motions.