Extended Kalman Filter in Autonomous Mobile Robot Localization and Mapping
Апстракт
The implementation of neural extended Kalman filter is achieved in terms of the monocular SLAM problem: multi-layer perceptron neural network is coupled with EKF to improve the state transition model. The main advantage of NEKF is the ability of the neural network to learn a model of the system on-line. This article showed that the introduction of neural network has resulted in higher accuracy of NEKF than a “standard” extended Kalman filter implementation for monocular SLAM. Multiple repetitions of experiment are performed, and experimental results indicate that NEKF outperforms EKF and odometry in terms of accuracy. Future work could be extended through implementation of other Gaussian filters (Unscented Kalman Filter or Extended Information Filter) and different types of feedforward neural networks (Radial Basis Function or maybe even Hyper Basis Function). To achieve real time performance of 30 [Hz] additional hardware is necessary. Finally, through experiments with the mobile robo...t and simple USB web camera, we have showed that this approach can be applied but we are looking forward to implement it in the real manufacturing environment and assess its performance.
Кључне речи:
The neural extended Kalman filter (NEKF) / Monocular SLAM problem / Odometry / Unscented Kalman Filter or Extended Information Filter / Radial Basis Function / Hyper Basis Function Neural Networks / Feedforward neural networks / Mobile robot / USB web camera / Manufacturing environment / Autonomous robot behaviorИзвор:
Bulletin of the Transilvania University of Brasov (Selected paper of the 4th International Conference on Robotics – ROBOTICS 2008), 2008, 15, 50, 435-444Издавач:
- Published by Transilvania University Press
Финансирање / пројекти:
Колекције
Институција/група
Mašinski fakultetTY - CHAP AU - Vuković, Najdan AU - Miljković, Zoran PY - 2008 UR - https://machinery.mas.bg.ac.rs/handle/123456789/4552 AB - The implementation of neural extended Kalman filter is achieved in terms of the monocular SLAM problem: multi-layer perceptron neural network is coupled with EKF to improve the state transition model. The main advantage of NEKF is the ability of the neural network to learn a model of the system on-line. This article showed that the introduction of neural network has resulted in higher accuracy of NEKF than a “standard” extended Kalman filter implementation for monocular SLAM. Multiple repetitions of experiment are performed, and experimental results indicate that NEKF outperforms EKF and odometry in terms of accuracy. Future work could be extended through implementation of other Gaussian filters (Unscented Kalman Filter or Extended Information Filter) and different types of feedforward neural networks (Radial Basis Function or maybe even Hyper Basis Function). To achieve real time performance of 30 [Hz] additional hardware is necessary. Finally, through experiments with the mobile robot and simple USB web camera, we have showed that this approach can be applied but we are looking forward to implement it in the real manufacturing environment and assess its performance. PB - Published by Transilvania University Press T2 - Bulletin of the Transilvania University of Brasov (Selected paper of the 4th International Conference on Robotics – ROBOTICS 2008) T1 - Extended Kalman Filter in Autonomous Mobile Robot Localization and Mapping EP - 444 IS - 50 IS - Special Issue No. I Vol. 2 SP - 435 VL - 15 UR - https://hdl.handle.net/21.15107/rcub_machinery_4552 ER -
@inbook{ author = "Vuković, Najdan and Miljković, Zoran", year = "2008", abstract = "The implementation of neural extended Kalman filter is achieved in terms of the monocular SLAM problem: multi-layer perceptron neural network is coupled with EKF to improve the state transition model. The main advantage of NEKF is the ability of the neural network to learn a model of the system on-line. This article showed that the introduction of neural network has resulted in higher accuracy of NEKF than a “standard” extended Kalman filter implementation for monocular SLAM. Multiple repetitions of experiment are performed, and experimental results indicate that NEKF outperforms EKF and odometry in terms of accuracy. Future work could be extended through implementation of other Gaussian filters (Unscented Kalman Filter or Extended Information Filter) and different types of feedforward neural networks (Radial Basis Function or maybe even Hyper Basis Function). To achieve real time performance of 30 [Hz] additional hardware is necessary. Finally, through experiments with the mobile robot and simple USB web camera, we have showed that this approach can be applied but we are looking forward to implement it in the real manufacturing environment and assess its performance.", publisher = "Published by Transilvania University Press", journal = "Bulletin of the Transilvania University of Brasov (Selected paper of the 4th International Conference on Robotics – ROBOTICS 2008)", booktitle = "Extended Kalman Filter in Autonomous Mobile Robot Localization and Mapping", pages = "444-435", number = "50, Special Issue No. I Vol. 2", volume = "15", url = "https://hdl.handle.net/21.15107/rcub_machinery_4552" }
Vuković, N.,& Miljković, Z.. (2008). Extended Kalman Filter in Autonomous Mobile Robot Localization and Mapping. in Bulletin of the Transilvania University of Brasov (Selected paper of the 4th International Conference on Robotics – ROBOTICS 2008) Published by Transilvania University Press., 15(50), 435-444. https://hdl.handle.net/21.15107/rcub_machinery_4552
Vuković N, Miljković Z. Extended Kalman Filter in Autonomous Mobile Robot Localization and Mapping. in Bulletin of the Transilvania University of Brasov (Selected paper of the 4th International Conference on Robotics – ROBOTICS 2008). 2008;15(50):435-444. https://hdl.handle.net/21.15107/rcub_machinery_4552 .
Vuković, Najdan, Miljković, Zoran, "Extended Kalman Filter in Autonomous Mobile Robot Localization and Mapping" in Bulletin of the Transilvania University of Brasov (Selected paper of the 4th International Conference on Robotics – ROBOTICS 2008), 15, no. 50 (2008):435-444, https://hdl.handle.net/21.15107/rcub_machinery_4552 .