Abstract:Accurate and reliable state estimation and mapping are the foundation of most autonomous driving systems. In recent years, researchers have focused on pose estimation through geometric feature matching. However, most of the works in the literature assume a static scenario. Moreover, a registration based on a geometric feature is vulnerable to the interference of a dynamic object, resulting in a decline of accuracy. With the development of a deep semantic segmentation network, we can conveniently obtain the semantic information from the point cloud in addition to geometric information. Semantic features can be used as an accessory to geometric features that can improve the performance of odometry and loop closure detection. In a more realistic environment, semantic information can filter out dynamic objects in the data, such as pedestrians and vehicles, which lead to information redundancy in generated map and map-based localization failure. In this paper, we propose a method called LiDAR inertial odometry (LIO) with loop closure combined with semantic information (LIO-CSI), which integrates semantic information to facilitate the front-end process as well as loop closure detection. First, we made a local optimization on the semantic labels provided by the Sparse Point-Voxel Neural Architecture Search (SPVNAS) network. The optimized semantic information is combined into the front-end process of tightly-coupled light detection and ranging (LiDAR) inertial odometry via smoothing and mapping (LIO-SAM), which allows us to filter dynamic objects and improve the accuracy of the point cloud registration. Then, we proposed a semantic assisted scan-context method to improve the accuracy and robustness of loop closure detection. The experiments were conducted on an extensively used dataset KITTI and a self-collected dataset on the Jilin University (JLU) campus. The experimental results demonstrate that our method is better than the purely geometric method, especially in dynamic scenarios, and it has a good generalization ability.

Semantic Assisted LiDAR Odometry with Loop Closure in Large Scale Urban Environment

SA-LOAM: Semantic-aided LiDAR SLAM with Loop Closure

Efficient 3D LIDAR Based Loop Closing Using Deep Neural Network

LIO-CSI: LiDAR inertial odometry with loop closure combined with semantic information

Semantic scan context: a novel semantic-based loop-closure method for LiDAR SLAM

Semantic Assisted Loop Closure Detection for Automated Driving

Efficient LiDAR Odometry for Autonomous Driving.

A High-Precision 3D Lidar Odometry Based on Image Semantic Constraints

Semantic-Assisted LIDAR Tightly Coupled SLAM for Dynamic Environments

LiDAR Odometry and Mapping Based on Semantic Information for Outdoor Environment

When-to-Loop: Enhanced Loop Closure for LiDAR SLAM in Urban Environments Based on SCAN CONTEXT

LLOAM: LiDAR Odometry and Mapping with Loop-closure Detection Based Correction

Monocular SLAM for Large Scale Scenes

Robust Real-Time Optimized LiDAR Odometry with Loop Closure Detection

ASL-SLAM: A LiDAR SLAM with Activity Semantics-Based Loop Closure

RS-SLAM: Real time semantic slam with driverless car using LiDAR-Camera-IMU sensing

SELVO: A Semantic-Enhanced Lidar-Visual Odometry.

SD-SLAM: A Semantic SLAM Approach for Dynamic Scenes Based on LiDAR Point Clouds

A Semantic-Assisted LiDAR SLAM Based on Attention Feature Points