Point Cloud Acquisition Without GNSS Using Simulated PPS and Inertial Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for acquiring point cloud data in environments without GNSS signals, such as urban subways and tunnels, face limitations in accuracy and efficiency due to reliance on satellite signals, requiring numerous stations and artificial feature points, which restricts data acquisition over long distances.
Innovation Solution
A method and device that re-sample line data from topographic maps to generate discrete data, simulate a GNSS satellite protocol using a PPS generator and distance measuring instrument, and optimize POS data to control LiDAR for accurate point cloud data acquisition, eliminating the need for extensive station setup and artificial features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If single point measurement method is used to acquire point cloud data in subway tunnel, then point cloud data can be acquired, but a large number of stations need to be set up and the distance between stations cannot exceed the effective scanning distance
Solution Approach 1:
The patent replaces the mechanical station-based measurement system with an inertial navigation system that uses accelerometers and gyroscopes to calculate position and attitude continuously during movement, eliminating the need for discrete station setups while maintaining measurement accuracy
Solution Approach 2:
The patent implements continuous point cloud data acquisition and positioning during movement through the tunnel using inertial sensors, rather than performing discrete measurements at stationary points, allowing uninterrupted data collection over long distances
2Productivity
If SLAM technique is used for point cloud data acquisition, then continuous point cloud data can be obtained during movement, but matching accuracy depends on homonymous feature points which are not obvious in single-feature tunnels
Solution Approach 1:
The patent replaces the vision-based SLAM feature matching system with a physical measurement system using inertial sensors and distance measuring instruments that directly measure position and distance without relying on visual feature recognition, ensuring accurate positioning in feature-poor environments
Solution Approach 2:
The patent introduces inertial measurement units and distance measuring instruments as intermediary devices that provide direct physical measurements of position and distance, serving as a reliable mediator between movement and positioning when visual feature matching fails
3Measurement precision
If artificial feature points are placed in tunnel to improve SLAM matching accuracy, then matching accuracy improves, but the process becomes more complex and time-consuming
Solution Approach 1:
The patent eliminates the need for artificial feature point placement by using inertial navigation and direct distance measurement technologies that determine position through physical sensing rather than visual feature recognition, simplifying the data preparation process
4Measurement precision
If moving speed is limited and acquisition distance is limited to ensure high precision, then point cloud data accuracy is maintained, but it is difficult to acquire data for long-distance tunnels
Solution Approach 1:
The patent enables continuous inertial navigation and point cloud acquisition over long distances by continuously integrating accelerometer and gyroscope data to track position and attitude changes throughout the entire tunnel traversal, rather than limiting to short segmented acquisitions
Solution Approach 2:
The patent replaces the limited-range SLAM system with an inertial navigation system that can accumulate positioning data continuously over long distances without requiring frequent re-initialization or feature point re-detection, extending the effective acquisition range
Data Source
AI summary
The present invention provides a point cloud data acquisition method and device under a situation of no GNSS signal. The method comprises: re-sampling line data acquired from a topographic map to obtain discrete line data; generating a full-second PPS pulse in a simulating manner; counting by using a distance measuring instrument, sampling the count, when it is detected that the full-second PPS pulse is received, calculating position information at a current moment; simulating a GNSS satellite protocol according to the position information at the current moment; parsing the GNSS satellite protocol by using a point cloud data acquisition module to complete time synchronization, and controlling a LiDAR to acquire point cloud data; parsing the GNSS satellite protocol by using an inertial measurement module, and recording attitude determination positioning data in real time to generate POS data; and optimizing the POS data to obtain accurate point cloud data.