Bayesian Target Estimator Using Dynamic Mixed Quadrature
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current target identification systems face challenges in properly conditioning measurement variates and capturing visible and hidden stochastic information, leading to complexities in calculating posterior distributions and requiring efficient real-time implementation.
Innovation Solution
A target estimator and system model that condition measurement variates using a Bayes optimal estimation approach, incorporating dynamic mixed quadrature for real-time processing, which captures correlations between target types, states, and sensor measurements, facilitating accurate target identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If measurement variates are combined without proper conditioning, then calculation complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent applies preliminary action by performing conditioning operations on measurement variates before they are combined in the posterior distribution calculation. The system pre-processes measurements to account for correlations and conditional dependencies, ensuring that when variates are combined, the computational complexity remains manageable while measurement precision is preserved through proper statistical conditioning.
2Device complexity
If hidden states are replaced by estimated values, then device complexity is reduced, but loss of information increases
Solution Approach 1:
The patent implements feedback by using the estimated hidden states to update and refine the posterior distribution calculations. The system continuously incorporates measurement data to adjust estimates of hidden states, ensuring that stochastic information is preserved through iterative refinement rather than being lost in single-pass estimation. This feedback mechanism maintains accuracy while managing system complexity.
3Ease of operation
If expert knowledge is used to project to lower dimensional space, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the dimensional representation of measurement variates based on the specific problem context. Rather than fixed expert-defined projections, the system adapts the dimensionality and selection of variates to preserve the statistical properties necessary for accurate posterior distribution calculation, thereby maintaining measurement precision while achieving computational tractability.
Data Source
AI summary
A method for target identification includes receiving, at one or more processors, a number of measurements of a target, each measurement from the number of measurements being observed at a predetermined time (zk), a number of target types (T), each one of the number of measurements, each one the number of target type and each one of one or more hidden states, each hidden state (xk) being characterized at the predetermined time, being correlated to one another, providing, using the one or more processors, a first conditional probability distribution, a conditional probability of a target type given a number of measurements, defined inductively, and obtaining an estimate of the target type from the first conditional probability. Systems that implement the method are also disclosed.