Multi-Sensor Fusion Module for Vehicle Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current perception systems for autonomous vehicles face challenges in providing robust and accurate environmental awareness due to limitations in sensor data fusion, particularly in handling overlapping cell measurement uncertainties and the need for high-level information for navigation planning.
Innovation Solution
A device comprising multiple sensors (e.g., radar, video cameras) that produce synchronized data, a merging module for combining positions and speeds, and a risk evaluation module that generates risk indicators by combining scores from distance, angle, speed, and object type calculation blocks to enhance navigation decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple sensors are combined to improve perception reliability and accuracy, then the integrity of multi-sensor perception is improved, but the device complexity increases due to the need for synchronization and data fusion modules
Solution Approach 1:
The patent combines multiple sensors (radar, video cameras, ultrasonic sensors) and their data processing functions into a unified sensor fusion module. This merging approach integrates the detection capabilities of individual sensors while consolidating the synchronization and data fusion operations into a single coordinated system, thereby improving perception reliability without proportionally increasing overall system complexity
Solution Approach 2:
The sensor fusion module performs multiple functions simultaneously: it synchronizes data from different sensors, fuses the sensor data to improve detection accuracy, generates risk indicators, and provides information for navigation planning. This multi-functionality allows a single module to handle various perception tasks, reducing the need for separate dedicated components for each function
2Measurement precision
If sensor data is processed in real-time to provide accurate environmental perception, then the measurement precision of object positions and speeds is improved, but the processing time increases due to the complexity of data fusion operations
Solution Approach 1:
The system performs preliminary synchronization of sensor data in time coordinates before conducting the actual data fusion and analysis. By pre-aligning the temporal references of multiple sensors, the system prepares the data in an optimized state that facilitates faster and more accurate fusion operations, reducing the overall processing time while maintaining high measurement precision
Solution Approach 2:
The sensor fusion module autonomously manages the complex operations of data synchronization, fusion, and risk assessment without requiring external intervention. The system self-regulates the processing pipeline, dynamically adjusting the fusion operations based on the incoming data characteristics, which optimizes processing efficiency while maintaining detection accuracy
3Reliability
If a grid-based cell system is used to evaluate measurement quality, then the reliability of distance measurement is improved, but the measurement precision deteriorates due to uncertainties at cell boundaries and overlapping measurements
Solution Approach 1:
The patent extracts the measurement evaluation process from the rigid grid-based cell system and integrates it into the continuous coordinate space used by the sensor fusion module. By removing objects from the constrained cell framework and processing their positions in continuous coordinates, the system eliminates the boundary effects and overlapping uncertainties inherent in grid-based approaches while maintaining reliable measurement assessment
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The device for real-time signalling of at least one object to a vehicle navigation module (60) comprises a first sensor (1) arranged to produce first-sensor data comprising a first captured position and first captured speed of the object with respect to the vehicle. The device further comprises: at least one second sensor (2, 3, 4) arranged to produce second-sensor data comprising a second captured position and second captured speed of the object with respect to the vehicle; a synchronization module (20) arranged to produce synchronized data comprising a first synchronized position from the first captured position and first captured speed and at least one second synchronized position from the second captured position and second captured speed; a merging module (30) arranged to produce merged data comprising a merged position from the first synchronized position and the second synchronized position, in such a way as to signal said at least one object to the navigation module (60) by communicating all or part of the merged data thereto.