Lidar Point Cloud Time Synchronization for Ghosting Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for integrating laser radar data in driverless vehicles face issues such as ghosting due to unsynchronized data and reduced recognition accuracy from insufficient information, leading to potential safety threats and control errors.
Innovation Solution
A data acquiring method and apparatus that prioritize real-time acquisition and storage of point cloud data from laser radars based on importance levels and collection periods, aligning and storing data within time windows to ensure synchronized and comprehensive data integration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If point cloud data from multiple laser radars are integrated based on preprocessed data, then comprehensive information is obtained, but ghosting phenomenon occurs due to time domain offsets
Solution Approach 1:
The patent applies preliminary action by performing time synchronization and alignment of point cloud data from multiple laser radars before integration. The system calculates time offsets between different laser radars and adjusts the data timestamps accordingly, ensuring that data from all sensors corresponds to the same time window. This preliminary time-domain alignment prevents the ghosting phenomenon that would otherwise occur during integration, while still achieving comprehensive information coverage.
2Reliability
If multiple channels of point cloud data are calculated separately, then ghosting problem is avoided, but recognition accuracy decreases due to insufficient information
Solution Approach 1:
The patent applies merging by integrating point cloud data from multiple laser radars after performing time synchronization and alignment. The system combines the synchronized data sets to create a comprehensive point cloud that includes all features from all sensors. This merging approach maintains control reliability by ensuring proper temporal alignment while simultaneously improving recognition accuracy through the inclusion of complete information from all laser radars.
3Device complexity
If data from multiple laser radars are integrated without time synchronization, then data integration is simple, but serious control errors occur due to ghosting
Solution Approach 1:
The patent applies preliminary action by implementing time synchronization and alignment as a preliminary step before data integration. The system calculates time offsets between different laser radars, determines corresponding time windows, and adjusts data timestamps accordingly. This preliminary processing, while adding some computational steps, maintains relatively simple integration logic and prevents serious control errors by eliminating the ghosting phenomenon that would otherwise occur.
4Ease of operation
If point cloud data are collected at different time points, then each laser radar operates independently, but time domain offsets cause ghosting
Solution Approach 1:
The patent applies preliminary action by performing time synchronization and alignment on independently collected point cloud data before integration. The system calculates time offsets between laser radars that operate independently at different time points, then adjusts the data timestamps to establish corresponding time windows. This preliminary processing preserves the independence and operational flexibility of each laser radar while eliminating time domain offsets and preventing ghosting, thereby maintaining high data quality.
Data Source
AI summary
The present application discloses a data acquiring method and apparatus applied to a driverless vehicle. A specific implementation of the method includes: selecting, from at least one laser radar of the driverless vehicle, a laser radar having a highest importance level as a first laser radar; acquiring a start time of a current time window and executing following data processing steps: executing a real-time acquisition and storage operation on point cloud data packets collected by the each of the at least one laser radar after the start time of the current time window; and determining whether any condition in a group of conditions is satisfied; and setting the start time of the current time window to be the current time and continuing to execute the data processing steps, in response to determining that any of the group of conditions is satisfied. This implementation implements the alignment and storage of point cloud data packets collected by at least one laser radar of the driverless vehicle.