Autonomous Vehicle Positioning with Sensor Data Buffering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in real-time positioning due to inconsistencies in delay times between different sensors such as SINS, GNSS, and LiDAR, leading to disordered measurement information and reduced accuracy.
Innovation Solution
An integrated positioning method and system using a first and second Kalman filter to process IMU and measurement data, buffering data to synchronize system state variables and error state variables, and performing time and measurement updates to correct system state variables, thereby improving real-time and accuracy of positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sensors (SINS, GNSS, LiDAR) are integrated for positioning, then positioning accuracy is improved, but measurement information becomes disordered due to inconsistent delay times
Solution Approach 1:
The system performs preliminary actions by buffering measurement data from multiple sensors before processing. The buffering mechanism stores measurement information temporarily, allowing the system to reorder and synchronize data from sensors with different delay characteristics (SINS, GNSS, LiDAR) before feeding them into the fusion filter, thus preventing disorder while maintaining integration benefits
Solution Approach 2:
The patent introduces a buffering mechanism as an intermediary component between data acquisition and fusion processing. This intermediary layer decouples the asynchronous arrival of measurement data from different sensors, enabling proper sequencing and synchronization without affecting the core fusion algorithm or sensor operations
2Stability of the object's composition
If measurement data is buffered and processed in order, then measurement information order is maintained, but real-time positioning performance deteriorates
Solution Approach 1:
The system performs preliminary buffering of measurement data with minimal delay, preparing data for ordered processing without significant time loss. The buffer operates efficiently to maintain data sequence while enabling rapid retrieval and processing, thus balancing order maintenance with real-time performance requirements
Solution Approach 2:
The patent implements dynamic processing where the system adapts its processing speed and buffer management based on real-time requirements. The fusion filter dynamically adjusts to process buffered data at optimal rates, maintaining both information order and real-time responsiveness by flexibly managing the trade-off between sequencing and speed
3Measurement precision
If SINS is integrated with GNSS and tachometer, then positioning accuracy is improved, but delay inconsistency causes disorder in measurement information
Solution Approach 1:
The buffering mechanism serves as an intermediary that receives measurement information from SINS, GNSS, and tachometer, then sequences them in the correct temporal order before processing. This intermediary layer prevents information loss due to disorder by ensuring proper sequencing while maintaining all measurement data for fusion processing
Data Source
AI summary
The present disclosure provides an integrated positioning method and system. The method comprises: receiving IMU data, and solving for the IMU data in a first Kalman filter to obtain a system state variable of an autonomous vehicle; receiving and buffering measurement data, and regarding the measurement data as current measurement data; performing measurement updating: obtaining the system state variable at a measurement time of current measurement data, and solving in the second Kalman filter to obtain an updated system state variable; if measurement data after the measurement time of the measurement data is already buffered, regarding the already-buffered measurement data after the measurement time as current measurement data, and performing the measurement update. The present disclosure can solve the problem about measurement information disorder caused by inconsistency of time delay in solving between different sensors, and improve real time and accuracy of the positioning.