Online Calibration of Extrinsic Parameters for Solid-State LIDAR Systems.

Area of Science:

• Robotics and Autonomous Systems
• Sensor Technology
• Computer Vision

Background:

• Calibrating multiple solid-state LIDAR systems is challenging due to differing hardware designs and scanning patterns.
• Establishing point correspondences between LIDAR-generated point clouds is crucial for accurate calibration.

Purpose of the Study:

• To develop and evaluate a robust method for calibrating multiple solid-state LIDAR sensors.
• To enable accurate point cloud registration and extrinsic parameter estimation for multi-LIDAR systems.

Main Methods:

• Data preprocessing to improve measurement quality.
• Feature extraction using the Fast Point Feature Histogram (FPFH) method.
• Extrinsic parameter computation via Fast Global Registration (FGR).

Main Results:

• Achieved a minimum root mean square error of 7 cm in a static indoor environment.
• Demonstrated the suitability of the method for online, real-time applications.
• Successfully established point correspondences between disparate solid-state LIDAR systems.

Conclusions:

• The proposed method effectively calibrates multiple solid-state LIDAR systems.
• This research advances multi-LIDAR fusion for improved perception and mapping.
• The approach has significant implications for autonomous driving, robotics, and environmental monitoring.