V2I-Calib: A Novel Calibration Approach for Collaborative Vehicle and Infrastructure LiDAR Systems
- URL: http://arxiv.org/abs/2407.10195v1
- Date: Sun, 14 Jul 2024 13:34:00 GMT
- Title: V2I-Calib: A Novel Calibration Approach for Collaborative Vehicle and Infrastructure LiDAR Systems
- Authors: Qianxin Qu, Yijin Xiong, Xin Wu, Hanyu Li, Shichun Guo,
- Abstract summary: This paper introduces a novel calibration method for cooperative vehicle and road infrastructure LiDAR systems.
We search for common matching boxes between vehicle and infrastructure nodes by constructing an affinity matrix.
Experiments on the DAIR-V2X dataset confirm the superiority of our method.
- Score: 13.089879956974979
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Cooperative vehicle and infrastructure LiDAR systems hold great potential, yet their implementation faces numerous challenges. Calibration of LiDAR systems across heterogeneous vehicle and infrastructure endpoints is a critical step to ensure the accuracy and consistency of perception system data, necessitating calibration methods that are real-time and stable. To this end, this paper introduces a novel calibration method for cooperative vehicle and road infrastructure LiDAR systems, which exploits spatial association information between detection boxes. The method centers around a novel Overall IoU metric that reflects the correlation of targets between vehicle and infrastructure, enabling real-time monitoring of calibration results. We search for common matching boxes between vehicle and infrastructure nodes by constructing an affinity matrix. Subsequently, these matching boxes undergo extrinsic parameter computation and optimization. Comparative and ablation experiments on the DAIR-V2X dataset confirm the superiority of our method. To better reflect the differences in calibration results, we have categorized the calibration tasks on the DAIR-V2X dataset based on their level of difficulty, enriching the dataset's utility for future research. Our project is available at https://github.com/MassimoQu/v2i-calib .
Related papers
- Galibr: Targetless LiDAR-Camera Extrinsic Calibration Method via Ground Plane Initialization [13.409482818102878]
Galibr is a fully automatic LiDAR-camera extrinsic calibration tool designed for ground vehicle platforms in any natural setting.
The method utilizes the ground planes and edge information from both LiDAR and camera inputs, streamlining the calibration process.
Our approach significantly enhances calibration performance, primarily attributed to our novel initial pose estimation method.
arXiv Detail & Related papers (2024-06-14T08:25:10Z) - Unsupervised Domain Adaptation for Self-Driving from Past Traversal
Features [69.47588461101925]
We propose a method to adapt 3D object detectors to new driving environments.
Our approach enhances LiDAR-based detection models using spatial quantized historical features.
Experiments on real-world datasets demonstrate significant improvements.
arXiv Detail & Related papers (2023-09-21T15:00:31Z) - Automated Automotive Radar Calibration With Intelligent Vehicles [73.15674960230625]
We present an approach for automated and geo-referenced calibration of automotive radar sensors.
Our method does not require external modifications of a vehicle and instead uses the location data obtained from automated vehicles.
Our evaluation on data from a real testing site shows that our method can correctly calibrate infrastructure sensors in an automated manner.
arXiv Detail & Related papers (2023-06-23T07:01:10Z) - Automated Static Camera Calibration with Intelligent Vehicles [58.908194559319405]
We present a robust calibration method for automated geo-referenced camera calibration.
Our method requires a calibration vehicle equipped with a combined filtering/RTK receiver and an inertial measurement unit (IMU) for self-localization.
Our method does not require any human interaction with the information recorded by both the infrastructure and the vehicle.
arXiv Detail & Related papers (2023-04-21T08:50:52Z) - TrajMatch: Towards Automatic Spatio-temporal Calibration for Roadside
LiDARs through Trajectory Matching [12.980324010888664]
We propose TrajMatch -- the first system that can automatically calibrate for roadside LiDARs in both time and space.
Experiment results show that TrajMatch can achieve a spatial calibration error of less than 10cm and a temporal calibration error of less than 1.5ms.
arXiv Detail & Related papers (2023-02-04T12:27:01Z) - Analyzing Infrastructure LiDAR Placement with Realistic LiDAR [14.163886343824064]
Vehicle-to-Everything(V2X) cooperative perception has attracted increasing attention.
How to find the optimal placement of infrastructure sensors is rarely studied.
We propose a pipeline that can efficiently and effectively find optimal installation positions for infrastructure sensors.
arXiv Detail & Related papers (2022-11-29T07:18:32Z) - Robust Self-Supervised LiDAR Odometry via Representative Structure
Discovery and 3D Inherent Error Modeling [67.75095378830694]
We develop a two-stage odometry estimation network, where we obtain the ego-motion by estimating a set of sub-region transformations.
In this paper, we aim to alleviate the influence of unreliable structures in training, inference and mapping phases.
Our two-frame odometry outperforms the previous state of the arts by 16%/12% in terms of translational/rotational errors.
arXiv Detail & Related papers (2022-02-27T12:52:27Z) - Efficient and Robust LiDAR-Based End-to-End Navigation [132.52661670308606]
We present an efficient and robust LiDAR-based end-to-end navigation framework.
We propose Fast-LiDARNet that is based on sparse convolution kernel optimization and hardware-aware model design.
We then propose Hybrid Evidential Fusion that directly estimates the uncertainty of the prediction from only a single forward pass.
arXiv Detail & Related papers (2021-05-20T17:52:37Z) - Automatic Extrinsic Calibration Method for LiDAR and Camera Sensor
Setups [68.8204255655161]
We present a method to calibrate the parameters of any pair of sensors involving LiDARs, monocular or stereo cameras.
The proposed approach can handle devices with very different resolutions and poses, as usually found in vehicle setups.
arXiv Detail & Related papers (2021-01-12T12:02:26Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.