SLAM Algorithm Based on Heterogeneous Sensor Data Fusion

With the rapid development of artificial intelligence technology, simultaneous localization and mapping (SLAM), as a key technology, has attracted widespread attention. However, laser-based SLAM technology suffers from significant drift issues. A heterogeneous sensor data fusion algorithm aimed at improving the performance of SLAM is proposed in this paper. First, data from laser radar and the Inertial Measurement Unit (IMU) is collected. Then, the Cartographer algorithm itself employs the Extended Kalman Filter (EKF) method to fuse the data from both sources, ultimately constructing a 2D SLAM framework for environmental mapping. Laser radar data and IMU data from three different experimental scenarios are collected, and the mentioned algorithm is utilized to fuse the data for map construction. The error between the constructed map and the real map is compared, demonstrating that the heterogeneous sensor data fusion-based SLAM algorithm in this paper outperforms single lidar Cartographer. It overcame the issue of rotational drift and exhibited promising application potential.