Design and Implementation of a New Hybrid Localization Algorithm Based on SA-MCL for Autonomous Robots

Abstract

Localization is a position estimation problem and is one of the most critical areas of research in the field of autonomous mobile robots. Particularly, the difficulty of the problem increases when the robot lacks knowledge of its initial position. This situation is referred to as the global localization problem, and particle filter-based algorithms have been proposed in the literature to address this issue. In this study, we investigate an energy-based Self Adaptive Monte Carlo Localization (SA-MCL) algorithm, which is a global localization algorithm, and propose a method to enhance the determination of similar energy regions more optimally on the map during the initialization phase. Furthermore, we demonstrate through simulations and real-world experiments that when the SA-MCL algorithm is used in a hybrid manner with the standard MCL algorithm, which employs relatively fewer particles, it provides more accurate and reliable position estimates compared to the original SA-MCL algorithm.

Keywords

• Autonomous robots
• Localization
• SA-MCL
• Standard MCL

Otonom Robotlar İçin KU-MCL Tabanlı Yeni Bir Hibrit Konum Belirleme Algoritması Tasarımı ve Uygulaması

Abstract

Lokalizasyon, bir konum tahmin problemidir ve otonom mobil robotlar üzerine yapılan çalışmalar arasında en kritik öneme sahip alanlardan biridir. Özellikle başlangıç anında robot kendi konumunu bilmiyorsa problemin zorluğu daha da artmaktadır. Robotun başlangıç anında kendi konumunu bilmemesi problemine global lokalizasyon problemi denilmiştir ve bu problemi çözmek için literatürde parçacık filtre tabanlı lokalizasyon algoritmalar mevcuttur. Bu çalışmada ise bir global lokalizasyon algoritması olan ve başlangıç anında parçacıkların daha akıllı ve efektif bir şekilde harita üzerine atanmasını sağlayan enerji tabanlı Kendinden Uyarlamalı Monte Carlo Lokalizasyon (KU-MCL) algoritması incelenerek benzer enerji bölgelerinin daha optimal bir şekilde belirlenebilmesi için bir yöntem önerilmiştir. Bunun yanında KU-MCL algoritması nispeten daha az parçacık kullanan standart MCL algoritması ile hibrit olarak çalıştığında, orijinal KU-MCL algoritmasına göre daha doğru ve güvenilir konum tahminlerinin yapıldığı simülasyon ve gerçek ortam deneyleri ile gösterilmiştir.

Keywords

• Otonom robotlar
• Lokalizasyon
• KU-MCL
• Standart MCL

References

1. F. O. Coelho, J. P. Carvalho, M. F. Pinto, and A. L. Marcato, “Ekf and computer vision for mobile robot localization,” in 2018 13th APCA International Conference on Automatic Control and Soft Computing (CONTROLO), 2018, pp. 148–153.
2. A. Yılmaz and H. Temeltaş, “An Improvement on SA-MCL Algorithm: Ellipse Based Energy Grids,” in 2018 6th International Conference on Control Engineering & Information Technology (CEIT), 2018, pp. 1–6.
3. Y. Wang et al., “An improved adaptive Monte Carlo localization algorithm fused with ultra wideband sensor,” in 2019 IEEE International Conference on Advanced Robotics and its Social Impacts (ARSO), 2019, pp. 421–426.
4. O. V. Altinpinar, E. C. Contarli, A. Kağizman, U. Uguzlar, E. Cansu, and V. Sezer, “Comparison of Autonomous Robot’s Mapping Performance Based on Number of Lidars And Number of Tours,” in 2022 Innovations in Intelligent Systems and Applications Conference (ASYU), 2022, pp. 1–6.
5. F. Gu, S. Valaee, K. Khoshelham, J. Shang, and R. Zhang, “Landmark graph-based indoor localization,” IEEE Internet Things J., vol. 7, no. 9, pp. 8343–8355, 2020.
6. L. Zhang, R. Zapata, and P. Lepinay, “Self-adaptive Monte Carlo localization for mobile robots using range finders,” Robotica, vol. 30, no. 2, pp. 229–244, 2012.
7. O. V. Altinpinar and V. Sezer, “A novel indoor localization algorithm based on a modified EKF using virtual dynamic point landmarks for 2D grid maps,” Robotics and Autonomous Systems, vol. 170, (2023): 104546.
8. A. W. Li and G. S. Bastos, “A hybrid self-adaptive particle filter through KLD-sampling and SAMCL,” in 2017 18th International Conference on Advanced Robotics (ICAR), 2017, pp. 106–111.

9. F. Dellaert, D. Fox, W. Burgard, and S. Thrun, “Monte carlo localization for mobile robots,” in Proceedings 1999 IEEE international conference on robotics and automation (Cat. No. 99CH36288C), 1999, vol. 2, pp. 1322–1328.
10. S. Thrun, W. Burgard, and D. Fox, “Probalistic Robotics,” Kybernetes, vol. 35, no. 7/8, pp. 1299–1300, Jan. 2006, doi: 10.1108/03684920610675292.
11. D. Fox, “Kld-sampling: Adaptive particle filters and mobile robot localization,” Adv. Neural Inf. Process. Syst., vol. 151, p. 152, 2002.
12. M. Quigley et al., “ROS: an open-source Robot Operating System,” in ICRA workshop on open source software, 2009, vol. 3, no. 3.2, p. 5.
13. R. Amsters and P. Slaets, “Turtlebot 3 as a robotics education platform,” in Robotics in Education: Current Research and Innovations 10, 2020, pp. 170–181.