Sensor Fusion via Alternating Correspondence Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sensor fusion techniques face challenges in accurately and reliably measuring scenes, particularly in environments where sensors provide incomplete or erroneous data, leading to inaccuracies in robot movement control and path planning.
Innovation Solution
A computer-implemented method for sensor fusion that receives data from multiple sensors with different modalities, determines corresponding features, performs mutual correspondence analysis considering performance specification data, and adjusts fusion based on these analyses to validate and combine data effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor fusion is performed using multiple sensors with different measurement modalities, then measurement accuracy and reliability of the scene is improved, but the complexity of data processing and fusion increases
Solution Approach 1:
The patent introduces an intermediary system that performs correspondence analysis between features from different sensors before fusion. This intermediary processing step validates whether features from different measurement modalities truly correspond to the same physical object, thereby improving measurement accuracy while managing processing complexity through structured validation protocols
Solution Approach 2:
The patent segments the sensor fusion process into distinct stages: feature extraction from individual sensors, correspondence analysis to validate feature matches, and final fusion. This segmentation allows each stage to be optimized independently, improving overall accuracy while making the complex processing more manageable and systematic
2Reliability
If correspondence analysis is performed between sensor features, then measurement reliability is improved, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary correspondence analysis before final sensor fusion, validating feature matches in advance. This preliminary action identifies and eliminates unreliable feature correspondences early in the process, ensuring higher measurement reliability while allowing the final fusion step to proceed more efficiently with pre-validated data
3Measurement precision
If performance specification data are taken into account in correspondence analysis, then measurement accuracy is improved, but device complexity and data requirements increase
Solution Approach 1:
The patent incorporates performance specification parameters (such as sensor accuracy ratings, measurement uncertainties, and detection thresholds) into the correspondence analysis process. By changing the parameters used in feature matching to include these performance metrics, the system achieves more accurate correspondence determination while managing complexity through parameter-based rather than structurally complex solutions
Data Source
AI summary
Sensor data are received from a multiplicity of sensors, wherein the multiplicity of sensors image a common scene using a multiplicity of measurement modalities. For each sensor of the multiplicity of sensors, at least one corresponding feature is determined in the respective sensor data and corresponding performance specification data are also obtained. In addition, mutual correspondence analysis between the features of the sensor data is carried out taking into account the corresponding performance specification data. Sensor fusion is also carried out on the basis of the correspondence analysis.