A Soft Time Synchronization Framework for Multi-Sensors in Autonomous Localization and Navigation

This paper proposes a soft time synchronization framework for synchronizing multiple sensors in autonomous localization and navigation. The main contribution is that we propose to synchronize all sensors through ROS nodes without hardware trigger based synchronization. Firstly, a ROS node is established for each sensor to acquire sensor messages and publish sensor messages on ROS topics. The sensors include color camera, 3D LIDAR and high-precision RTK-GPS devices in our test. Then, ROS synchronizer is used to synchronize messages on ROS topics. The time stamp of synchronized data set and generation time of the sensor messages can be used to validate the synchronization performance. From experimental results, it is shown that the proposed soft time synchronization framework can achieve low latency and low synchronization error compared to other datasets, proving that the soft synchronization framework is effective and applicable to various sensors.