A mapping method based on EKF and point-segment matching for indoor environment

A map construction method based on point-segment (PS) feature with laser scanner is presented in this paper to accomplish the indoor mobile robot SLAM. The preprocessed sensor data is divided into objective set through dynamic threshold segmentation. To fit the feature, improved Hough-transform is adopted to pick out the short segments from the objective set and an incorporating method is applied to merge the segments belonging to the same feature set. Then, a segment phase matching method is used to establish the global map, during which the EKF along with the matching technique is employed to estimate the optimal pose. Finally, experiment is implemented on the robot ATR1 in the given environment. Though the robot deviated from its planning path up to -177.38mm in Y-direction and 124.56mm in X-direction, the constructed map still shows the rationality and effectiveness of the method.

[1]  Andreas Zell,et al.  A hybrid approach for vision-based outdoor robot localization using global and local image features , 2007, 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[2]  Peter Cheeseman,et al.  On the Representation and Estimation of Spatial Uncertainty , 1986 .

[3]  K. Gulez,et al.  Evaluations of different Simultaneous Localization and Mapping (SLAM) algorithms , 2012, IECON 2012 - 38th Annual Conference on IEEE Industrial Electronics Society.

[4]  Cai Zi-xing,et al.  Study on obstacles detection of mobile robot based on laser scanner , 2006 .

[5]  Bruno Steux,et al.  tinySLAM: A SLAM algorithm in less than 200 lines C-language program , 2010, 2010 11th International Conference on Control Automation Robotics & Vision.

[6]  Heiko Hirschmüller,et al.  Multisensor data fusion for robust pose estimation of a six-legged walking robot , 2011, 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[7]  Lining Sun,et al.  Vision-based mobile robot localization and mapping using the PLOT features , 2012, 2012 IEEE International Conference on Mechatronics and Automation.

[8]  Jean-Paul Laumond,et al.  Position referencing and consistent world modeling for mobile robots , 1985, Proceedings. 1985 IEEE International Conference on Robotics and Automation.

[9]  Zhang Wei-gong.  Lane Coordination Detection Based on Hough Transformation and Least Square Fitting , 2011 .

[10]  Martial Hebert,et al.  Moving object detection with laser scanners , 2013, J. Field Robotics.

[11]  Sankar K. Pal,et al.  A review on image segmentation techniques , 1993, Pattern Recognit..

[12]  P. Sivakumar,et al.  A REVIEW ON IMAGE SEGMENTATION TECHNIQUES , 2016 .

[13]  Jae-Bok Song,et al.  Monocular Vision-Based SLAM in Indoor Environment Using Corner, Lamp, and Door Features From Upward-Looking Camera , 2011, IEEE Transactions on Industrial Electronics.

[14]  Nando de Freitas,et al.  Rao-Blackwellised Particle Filtering for Dynamic Bayesian Networks , 2000, UAI.