Key points are not available for this paper at this time.
Simultaneous Localization and Mapping (SLAM) is an increasingly important field in robotics. SLAM enables the processes of localization and mapping to run simultaneously by using methods like feature extraction, feature matching, extended Kalman Filter, and probability distribution models, allowing robots to have more autonomy and greater efficiency. SLAM has been applied in industry robots, autonomous vehicles, augmented reality, UAVs, humanoid robots, and planetary rovers. With an increasing number of applications of SLAM, the demand for higher-performance SLAM algorithms has driven innovation and advancement in the field every year. SLAM is a highly promising algorithm for the future of robotics, but problems like uncertainty, correspondence, and time complexity prevent the full use of SLAM in robotics applications. It is essential to analyze SLAM in all aspects to apply it to future work. This article provides a detailed insight into the SLAM process, considers previous advancements and current problems, and discusses the future of SLAM.
Ryan Xianqian Yuan (2024) studied this question.