An efficient LiDAR-based localization method for self-driving cars in dynamic environments

Zhang, Yihuan; Wang, Liang; Jiang, Xuhui; Zeng, Yong; Dai, Yifan

doi:10.1017/s0263574721000369

Cited by 27 publications

(15 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although these methods usually have good performance in dense 3D scans, they depend on exact point or surface matching. However, it is rare in sparse and nonrepetitive point clouds of lidar [17]. To reduce computational cost, a natural idea is adopting the voxelization strategy [18].…”

Section: Related Workmentioning

confidence: 99%

BA-LIOM: tightly coupled laser-inertial odometry and mapping with bundle adjustment

Li,

Zhang,

Zhang

et al. 2024

Robotica

View full text Add to dashboard Cite

We design a scheme for laser-inertial odometry and mapping with bundle adjustment (BA-LIOM), which can greatly mitigate the problem of undesired ground warping due to sparsity of laser scans and significantly reduce odometry drift. Specifically, an Inertial measurement unit (IMU)-assisted adaptive voxel map initialization algorithm is proposed and elaborately integrated with the existing framework LIO-SAM, allowing for accurate registration in the beginning of the localization and mapping process. In addition, to accommodate to fast-moving and structure-less scenarios, we design a tightly coupled odometry, which jointly optimizes both the IMU preintegration constraints and scan matching with adaptive voxel maps. The voxels (edge and plane, respectively) are updated with BA optimization. And then the accurate mapping result is obtained by performing local BA. The proposed BA-LIOM is thoroughly assessed using datasets collected from multiple platforms over a variety of environments. Experimental results show the superiority of BA-LIOM over the state-of-the-art methods in robustness and precision, especially for large-scale scenarios. BA-LIOM improves the accuracy of localization by $61\%$ and $73\%$ on the buildings and lawn datasets, respectively, and has a $29\%$ accuracy improvement over LIO-SAM on the KITTI datasets. A supplementary video can be accessed at https://youtu.be/5l4ZFhTc2sw.

show abstract

Section: Related Workmentioning

confidence: 99%

BA-LIOM: tightly coupled laser-inertial odometry and mapping with bundle adjustment

Li,

Zhang,

Zhang

et al. 2024

Robotica

View full text Add to dashboard Cite

show abstract

“…Researchers proposed many localization algorithms to obtain the accurate position of the object by matching the data from onboard sensors and the digital map. In [6], a localization method is proposed to estimate the pose of self-driving cars using a 3D-LiDAR sensor. With a map-matching method proposed to match the features to the map, a robust iterative closest point algorithm is utilized to deal with curb features, and a probability search method deals with intensity features.…”

Section: Related Workmentioning

confidence: 99%

“…Ongoing research in object detection and tracking employs various sensing approaches and algorithms. Researchers typically use sensor technologies such as passive infrared sensors (PIR), ultra-wideband radar [4,5], LIDAR [6] and digital cameras [7,8]. However, all these technologies have challenges in terms of accuracy, privacy, and environmental robustness [9].…”

Section: Introductionmentioning

confidence: 99%

mmWave Radar Sensors Fusion for Indoor Object Detection and Tracking

2022

View full text Add to dashboard Cite

Indoor object detection and tracking using millimeter-wave (mmWave) radar sensors have received much attention recently due to the emergence of applications of energy assignment, privacy, health, and safety. Increasing the valid field of view of the system and accuracy through multi-sensors is critical to achieving an efficient tracking system. This paper uses two mmWave radar sensors for accurate object detection and tracking: two noise reduction stages to reduce noise and distinguish cluster groups. The presented data fusion method effectively estimates the transformation of the data alignment and synchronizes the result that can allow us to visualize the objects’ information acquired by one radar on another one. An efficient density-based clustering algorithm to provide high clustering accuracy is presented. The Unscented Kalman Filter tracking algorithm with data association tracks multiple objects simultaneously in terms of accuracy and timing. Furthermore, an indoor object tracking system is developed based on our proposed method. Finally, the proposed method is validated by comparing it with our previous system and a commercial system. The experimental results demonstrate that the proposed method’s advantage is of positive significance for handling the effect of occlusions at higher numbers of weak data and for detecting and tracking each object more accurately.

show abstract

“…Zhang et al proposed a real-time localization method that estimates the location of an autonomous driving vehicle using a 3D-LiDAR system [ 8 ]. Point cloud data of a curb beside the road are extracted based on the laser intensity features and matched to a high-precision curb map generated offline.…”

Section: Related Workmentioning

confidence: 99%

An Entropy Analysis-Based Window Size Optimization Scheme for Merging LiDAR Data Frames

Kim

Jung

Hong

et al. 2022

Sensors

View full text Add to dashboard Cite

LiDAR is a useful technology for gathering point cloud data from its environment and has been adapted to many applications. We use a cost-efficient LiDAR system attached to a moving object to estimate the location of the moving object using referenced linear structures. In the stationary state, the accuracy of extracting linear structures is low given the low-cost LiDAR. We propose a merging scheme for the LiDAR data frames to improve the accuracy by using the movement of the moving object. The proposed scheme tries to find the optimal window size by means of an entropy analysis. The optimal window size is determined by finding the minimum point between the entropy indicator of the ideal result and the entropy indicator of the actual result of each window size. The proposed indicator can describe the accuracy of the entire path of the moving object at each window size using a simple single value. The experimental results show that the proposed scheme can improve the linear structure extraction accuracy.

show abstract

An efficient LiDAR-based localization method for self-driving cars in dynamic environments

Cited by 27 publications

References 40 publications

BA-LIOM: tightly coupled laser-inertial odometry and mapping with bundle adjustment

BA-LIOM: tightly coupled laser-inertial odometry and mapping with bundle adjustment

mmWave Radar Sensors Fusion for Indoor Object Detection and Tracking

An Entropy Analysis-Based Window Size Optimization Scheme for Merging LiDAR Data Frames

Contact Info

Product

Resources

About