Multi-Sensor Vehicle Object Grouping With Time-Series Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional obstacle detection systems face challenges in accurately grouping detection results from multiple sensors in real-time due to potential errors in instantaneous value grouping, while time-series grouping for higher accuracy requires longer time-series data processing.
Innovation Solution
A vehicle control system that integrates detection information from multiple sensors using an information integration device, which stores and processes time-series data to calculate correction parameters, allowing for accurate instantaneous value grouping and reducing processing time by performing calibration only when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If instantaneous value grouping is used to group detection results from multiple sensors, then processing time is short, but grouping accuracy deteriorates due to potential errors in determination
Solution Approach 1:
The system performs preliminary calibration processing to generate correction parameters in advance. These correction parameters are calculated based on time-series detection results and are stored for subsequent use in instantaneous value grouping, eliminating the need for time-consuming calibration during real-time operation
Solution Approach 2:
The system dynamically switches between two operating modes: calibration mode (for calculating correction parameters using time-series data) and detection mode (for rapid instantaneous value grouping using pre-calculated correction parameters). This dynamic adaptation allows the system to optimize between accuracy and speed based on operational requirements
2Measurement precision
If time-series grouping is used to improve grouping accuracy, then grouping accuracy is improved, but processing time increases due to longer time-series data requirements
Solution Approach 1:
The system extracts only the essential calibration information from time-series data to generate correction parameters. Instead of performing full time-series grouping analysis during real-time operation, the system extracts calibration parameters offline and applies them instantly during detection, achieving high accuracy without the time penalty of continuous time-series processing
Solution Approach 2:
The system performs the computationally intensive time-series analysis and correction parameter calculation in advance, before real-time detection is required. This preliminary calibration phase uses historical time-series data to establish correction parameters that are then reused for rapid instantaneous grouping
Data Source
AI summary
A vehicle control system is configured to group objects detected by a plurality of sensors. The vehicle control system includes an integration device that groups detection information from the plurality of sensors and outputs integrated detection information, and a vehicle control device that controls a vehicle on the basis of the integrated detection information. An arithmetic device of the information integration device stores first time-series information of the first detection information and second time-series information of the second detection information in a storage device, calculates a correction parameter of the first detection information by grouping the first time-series information and the second time-series information when the first sensor and the second sensor detect the same object, calculates correction information obtained by correcting the first detection information using the correction parameter, and outputs the integrated detection information by instantaneous value grouping using the correction information and the second detection information.