Research Article A Bioinspired Neural Model Based Extended Kalman Filter for Robot SLAM Jianjun Ni, 1,2 Chu Wang, 1 Xinnan Fan, 1 and Simon X. Yang 3 1 College of IOT Engineering, Hohai University, Changzhou 213022, China 2 Changzhou Key Laboratory of Sensor Networks and Environmental Sensing, Hohai University, Changzhou 213022, China 3 Advanced Robotics and Intelligent Systems (ARIS) Laboratory, School of Engineering, University of Guelph, Guelph, ON, Canada N1G 2W1 Correspondence should be addressed to Jianjun Ni; njjhhuc@gmail.com Received 15 March 2014; Revised 22 July 2014; Accepted 22 July 2014; Published 13 August 2014 Academic Editor: Yi Chen Copyright © 2014 Jianjun Ni et al. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Robot simultaneous localization and mapping (SLAM) problem is a very important and challenging issue in the robotic feld. Te main tasks of SLAM include how to reduce the localization error and the estimated error of the landmarks and improve the robustness and accuracy of the algorithms. Te extended Kalman flter (EKF) based method is one of the most popular methods for SLAM. However, the accuracy of the EKF based SLAM algorithm will be reduced when the noise model is inaccurate. To solve this problem, a novel bioinspired neural model based SLAM approach is proposed in this paper. In the proposed approach, an adaptive EKF based SLAM structure is proposed, and a bioinspired neural model is used to adjust the weights of system noise and observation noise adaptively, which can guarantee the stability of the flter and the accuracy of the SLAM algorithm. Te proposed approach can deal with the SLAM problem in various situations, for example, the noise is in abnormal conditions. Finally, some simulation experiments are carried out to validate and demonstrate the efciency of the proposed approach. 1. Introduction Robot simultaneous localization and mapping (SLAM) is one of the most challenging problems in mobile robotic felds, which has important theory and application value in vari- ous robotic applications, such as the underwater detection, domestic service, and universe exploration [1–5]. Various approaches have been proposed to deal with the SLAM problem. Mullane et al. [6] proposed an integrated Bayesian framework for feature-based SLAM in the general case of uncertain feature number and data association. Lui and Jarvis [7] presented a pure vision-based topological SLAM system for mobile robot autonomous navigation in initially unknown environments. Chatterjee and Matsuno [8] proposed a new neurofuzzy based adaptive Kalman fltering algorithm for SLAM of mobile robots or vehicles. Zhou et al. [9] proposed an auxiliary marginal particle flter and FastSLAM based compositive SLAM algorithm to improve the performance of samples and increase the estimation accuracy. Kaess et al. [10] presented a novel approach to the simultaneous localization and mapping problem that is based on fast incre- mental matrix factorization. Benedettelli et al. [11] proposed a multirobot cooperative SLAM algorithm using M-Space representation of linear features, suitable for environments which can be represented in terms of lines and segments. Previous research on SLAM problem may be divided into two broad categories. One category consists of a number of mathematical probabilistic techniques, such as EKF based algorithm [12, 13], particle fltering based algorithm [14, 15], and Monte Carlo localization method [16, 17]. Te other cat- egory of research focuses on emulating the biological systems thought to be responsible for mapping and navigation in animals [18–20]. However, both of the category algorithms have some limitations. For example, the standard Kalman fltering is based on the minimum variance estimation; the flter will exhibit a divergence phenomenon with the increase in the number of flterings; namely, the error between the estimated value and the actual value will become bigger and bigger. Te biological emulating based algorithms are too Hindawi Publishing Corporation Mathematical Problems in Engineering Volume 2014, Article ID 905826, 11 pages http://dx.doi.org/10.1155/2014/905826