Journal Article•
Tracking Algorithm for Maneuvering Target Based on Adaptive Kalman Filter
TL;DR: An improved method for tracking a maneuver target is proposed, where the maneuver detector provides the estimate of time instant at which a target starts to maneuver and when a target maneuver is determined, the kalman filter model will be adjusted with varied target motion state.
Abstract: The application of kalman filter in tracking the maneuver target is not available as it is used in tracking the target of uniform motion.Therefore,an improved method for tracking a maneuver target is proposed.By the proposed method,the maneuver detector provides the estimate of time instant at which a target starts to maneuver;and when a target maneuver is determined,the kalman filter model will be adjusted with varied target motion state.The maneuver,modeled as acceleration,is estimated recursively.Finally,the performance of the proposed approach proves to be superior to that of the kalman filters by simulation.
Citations
More filters
01 Nov 2018
TL;DR: A method of combining the MEC + CEM detection algorithm and the Kalman filter assistant tracking algorithm that can reach 94.06% for the detection of golf ball, which is better than the external disturbance factor and much faster than the Hough transform algorithm.
Abstract: In order to reduce the recognition complexity and improve the detection accuracy and tracking efficiency, so that the robot can quickly and efficiently hit a golf ball into the hole, this paper proposes a method of combining the MEC + CEM detection algorithm and the Kalman filter assistant tracking algorithm. First, the acquired RGB image is transformed into HSV color space and the optimal threshold is taken. Then, the MEC + CEM detection algorithm is used to obtain the exact position of the golf ball. The matching information is used to correct the position information and the credibility judgment. Finally, the mathematical model of robot hitting is established. The experimental results show that the average accuracy of the algorithm can reach 94.06% for the detection of golf ball, which is better than the external disturbance factor and much faster than the Hough transform algorithm. In the success rate of hitting the hole, the robot using this algorithm is higher than human and meet the requirements of the golf tournament.
2 citations
References
More filters
01 Nov 2018
TL;DR: A method of combining the MEC + CEM detection algorithm and the Kalman filter assistant tracking algorithm that can reach 94.06% for the detection of golf ball, which is better than the external disturbance factor and much faster than the Hough transform algorithm.
Abstract: In order to reduce the recognition complexity and improve the detection accuracy and tracking efficiency, so that the robot can quickly and efficiently hit a golf ball into the hole, this paper proposes a method of combining the MEC + CEM detection algorithm and the Kalman filter assistant tracking algorithm. First, the acquired RGB image is transformed into HSV color space and the optimal threshold is taken. Then, the MEC + CEM detection algorithm is used to obtain the exact position of the golf ball. The matching information is used to correct the position information and the credibility judgment. Finally, the mathematical model of robot hitting is established. The experimental results show that the average accuracy of the algorithm can reach 94.06% for the detection of golf ball, which is better than the external disturbance factor and much faster than the Hough transform algorithm. In the success rate of hitting the hole, the robot using this algorithm is higher than human and meet the requirements of the golf tournament.
2 citations