Multi-Sensor Target Information Fusion With Adaptive Kalman Weights

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional embedded processor chips struggle with irrational weight assignment in multi-sensor target information fusion due to low processing performance, leading to inaccurate fusion estimation results, especially in scenarios involving sensor jumps.

Innovation Solution

An optimization method and apparatus that calculates fusion weights in real-time using Kalman filtering to adjust weights based on sensor-specific covariances and measurement results, ensuring optimal fusion estimation by correcting covariances and using transformation matrices and Kalman gains.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional embedded processor chips are used for multi-sensor target information fusion, then the system structure is simple and easy to implement, but the processing performance is low leading to irrational weight assignment and inaccurate fusion estimation results

Engineering Contradiction:
Improvefusion estimation accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent implements dynamic weight adjustment through real-time covariance calculation and Kalman filtering. The fusion weights are not fixed but continuously adapted based on current sensor performance and environmental conditions, allowing the system to optimize fusion accuracy dynamically while maintaining real-time processing capability on embedded processors.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameters used for weight assignment from static sensor characteristics to dynamic covariance values calculated from actual measurement results. By using Kalman filtering to compute optimal weights based on current estimation errors and sensor covariances, the system achieves higher fusion accuracy without requiring excessive processing power.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If real-time weight calculation using Kalman filtering is implemented, then fusion estimation accuracy is improved, but processing power consumption increases

Engineering Contradiction:
Improvefusion estimation accuracyVSAvoidprocessing power consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent segments the fusion process into independent Kalman filter modules for each sensor type. Each sensor's measurement results are processed separately to calculate individual covariances and weights, which are then combined in the final fusion step. This modular approach reduces computational complexity and power consumption compared to processing all sensor data simultaneously.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by calculating weights only for the current time step using recursive Kalman filtering, rather than reprocessing all historical data. The method uses the previous estimation results and current measurements to update weights incrementally, significantly reducing processing power requirements while maintaining accuracy.

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If fixed fusion weights are assigned in advance, then processing is simple and fast, but the weight assignment is irrational and not universally applicable especially in sensor jump scenarios

Engineering Contradiction:
Improveprocessing speedVSAvoidweight assignment adaptability
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent implements feedback mechanisms where the Kalman filter continuously monitors the consistency between sensor measurements and predicted target states. When sensor jumps or anomalies occur, the covariance calculations automatically reflect the degraded sensor performance, leading to reduced weights for unreliable sensors. This feedback loop enables adaptive weight adjustment without manual intervention.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs self-service by automatically adjusting fusion weights based on its own internal state and sensor performance metrics. The Kalman filter uses the system's own prediction errors and measurement covariances to determine optimal weights, eliminating the need for external calibration or manual weight assignment while maintaining processing efficiency.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3726428B1Optimization method and apparatus for multi-sensor target information fusion, computer device, and recording medium
Publication Date: 2025.10.01 NIO ANHUI HLDG CO LTD
  • EP3726428B1 patent drawingFigure 1
  • EP3726428B1 patent drawingFigure 2~3
  • EP3726428B1 patent drawingFigure 4~5

AI summary

An optimization method and apparatus for multi-sensor target information fusion, a computer device, and a recording medium. The optimization method for multi-sensor target information fusion comprises the steps of: obtaining, for each time, a fusion prediction result for all sensors concerning a target state at the current time; obtaining actual measurement results from each sensor concerning the target state at the current time; obtaining, for each set of actual measurement results, an optimal estimation result for a corresponding sensor concerning the target state at the current time based on the fusion prediction result and the set of actual measurement results; and fusing optimal estimation results for all the sensors to determine a corresponding weight of each of the optimal estimation results and thus to obtain an optimal fusion estimation result concerning the target state at the current time.