We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Design and implementation of intelligent LiDAR SLAM for autonomous mobile robots using evolutionary normal distributions transform.
- Authors
Huang, Hsu-Chih; Xu, Sendren Sheng-Dong; Lin, Hsien-Chan; Xiao, Yuan-Sheng; Chen, Yu-Xiang
- Abstract
This paper presents a method that employs an evolutionary normal distributions transform (NDT) for simultaneous localization and mapping (SLAM) using light detection and ranging (LiDAR) for autonomous mobile robots. An adaptive inertia weight and two genetic operators were employed in the Taguchi-based whale optimization algorithm (WOA) to improve the search diversity and avoid local optima. NDT was used to model the environment of differential-drive mobile robots and WOA was applied to optimize the scan matching for the robotic SLAM problem. The NDT-WOA determines the SLAM pose estimation using sensed data from the physical world. A nonholonomic mobile robot was steered to achieve the NDT-WOA SLAM task with the derived robot kinematics and actuator dynamics. The proposed method was implemented in a TurtleBot3 Burger robotic development kit, which includes a single-board computer and an OpenCR control board. The Robot Operating System (ROS) was utilized to implement the evolutionary NDT-WOA SLAM system due to its flexibility, open source, and client library. Simulation and comparisons were conducted to illustrate the efficiency of the proposed NDT-WOA SLAM method compared to other SLAM paradigms. The experimental results show the effectiveness of the proposed evolutionary NDT-WOA SLAM for autonomous mobile robots. The results could have theoretical and practical significance for robotics research.
- Subjects
MOBILE robots; AUTONOMOUS robots; METAHEURISTIC algorithms; GAUSSIAN distribution; SLAM (Robotics); ROBOT kinematics
- Publication
Soft Computing - A Fusion of Foundations, Methodologies & Applications, 2024, Vol 28, Issue 6, p5321
- ISSN
1432-7643
- Publication type
Article
- DOI
10.1007/s00500-023-09219-0