Abstract:Traditional 2D Simultaneous Localization and Mapping (SLAM) algorithms commonly use occupancy grid map models, which are susceptible to Gaussian noise. The constraint information in the backend optimization process is limited. Sensor data utilization at various stages is also incomplete. To address these issues, this paper proposes a multi-sensor deep fusion SLAM method based on the Truncated Signed Distance Function (TSDF) map. Firstly, the inertial sensing unit (IMU) is pre-integrated, and then the distortion correction of the laser point cloud is corrected by using the posture obtained after pre-integration. In the front-end, the Unscented Kalman Filter (UKF) method is used to fuse odometry, IMU data, and LiDAR scan matching results to obtain the pose information of the robot. The backend uses IMU pre-integration factor, loopback detection, and laser point cloud registration to enhance constraints for global map pose optimization and achieve deep fusion of multi-sensor data. The map model uses a TSDF map, which constructs obstacle edges through weighted fusion and linear interpolation, and it truncates the grid cells around obstacles, thereby reducing the influence of Gaussian noise. The performance of Karto-SLAM, Cartographer, and the proposed algorithm is verified by comparing the public dataset and the dataset collected in the real environment. The results show that the proposed method effectively avoids the ghosting phenomenon of traditionally occupied raster maps and reduces Gaussian noise in terms of mapping. In terms of positioning accuracy, the effect of back-end optimization is enhanced by a multi-constraint relationship, which reduces the relative and absolute pose errors of the real trajectory. Our method improves localization accuracy by an average of 9% compared to Cartographer and by an average of 34% compared to Karto-SLAM.

Adaptive Square-Root Transformed Unscented FastSLAM with KLD-resampling

An Improved Transformed Unscented FastSLAM with Adaptive Genetic Resampling

Intelligent Filter-Based Slam For Mobile Robots With Improved Localization Performance

Robust SLAM using square-root cubature Kalman filter and Huber’ s GM-estimator

Algorithm for Mobile Robots' Simultaneous Localization and Mapping in Unknown Environments

LS-SLAM: SLAM with Lebesgue Sampling

Stereo Vision Based SLAM Using Rao-Blackwellised Particle Filter

An Improved FastSLAM Algorithm Based on Revised Genetic Resampling and SR-UPF

An Improved Adaptive Unscented FastSLAM with Genetic Resampling

A FastSLAM Algorithm Based on the Improved Auxiliary Particle Filter with Stirling Interpolation

Fast SLAM algorithm based on central difference Kalman filter

Local Optimized and Scalable Frame-to-model SLAM

A Robot SLAM Improved by Quantum-Behaved Particles Swarm Optimization

An Improved UFastSLAM With Generalized Correntropy Loss and Adaptive Genetic Resampling

Vision-based Unscented FastSLAM for Mobile Robot

The SLAM combining with visual odometry for small-scale aerial robots

A FastSLAM Algorithm Based on the Auxiliary Particle Filter with Stirling Interpolation

Enhanced Unscented Kalman Filter-Based SLAM in Dynamic Environments: Euclidean Approach

Novel SLAM algorithm for UGVs based on unscented Kalman filtering

Laser Simultaneous Localization and Mapping Algorithm Based on Adaptive Features and Closed-Loop Optimization

A Multi-Sensor Deep Fusion SLAM Algorithm Based on TSDF Map