Sensor Time-Series Trend Detection With Linear Filter Control

Solution Overview

Problem

Existing methods for automated trend recognition in data time series are resource-intensive, particularly with deep learning approaches that require significant computational power and energy consumption.

Innovation Solution

A computer-implemented method that applies a linear filter function to a data time series with varying resolutions, detecting trends by comparing characteristic values with threshold values, and triggering a system response when a trend is identified.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If deep learning approaches are used for trend detection, then measurement precision is improved, but use of energy and device complexity increase significantly

Engineering Contradiction:

Improvetrend detection accuracyVSAvoidcomputational energy consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent replaces expensive, resource-intensive deep learning models with inexpensive linear filter functions that require minimal computational resources. The linear filter acts as a simplified, resource-efficient alternative that achieves sufficient trend detection accuracy without the high energy consumption and computational complexity of deep learning approaches.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent changes the computational parameters from complex deep learning operations to simple linear filtering operations. By transforming the approach from high-complexity neural network computations to low-complexity linear filter applications, the system achieves trend detection with significantly reduced energy consumption and computational requirements.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If deep learning approaches are used for trend detection, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improvetrend detection accuracyVSAvoidcomputational system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces expensive, resource-intensive deep learning models with inexpensive linear filter functions that require minimal computational resources. The linear filter acts as a simplified, resource-efficient alternative that achieves sufficient trend detection accuracy without the high energy consumption and computational complexity of deep learning approaches.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent substitutes complex computational mechanics (deep learning algorithms) with simpler mathematical operations (linear filtering). This replacement reduces the computational burden and system complexity while maintaining the essential functionality of trend detection through characteristic value analysis.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Use of energy by moving object

If linear filter function is applied for trend detection, then use of energy is reduced, but measurement precision may be compromised

Engineering Contradiction:

Improvecomputational energy consumptionVSAvoidtrend detection accuracy

Core Design Contradiction:

Use of energy by moving objectVSMeasurement precision

Solution Approach 1:

The patent changes the computational parameters from complex deep learning operations to simple linear filtering operations. By transforming the approach from high-complexity neural network computations to low-complexity linear filter applications, the system achieves trend detection with significantly reduced energy consumption and computational requirements.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent implements a feedback mechanism where the characteristic value obtained from linear filtering is compared against a threshold to detect trends. This feedback loop ensures that even with simplified linear filtering, the system can accurately identify when trends occur by monitoring whether the characteristic value exceeds predetermined thresholds, thereby maintaining measurement precision.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20250028309A1Computer-implemented method for controlling a system based on a detected trend

Publication Date: 2025.01.23 ROBERT BOSCH GMBH

AI summary

A computer-implemented method for controlling a system based on a trend detected in a data time series acquired by a sensor. The method includes: acquiring and providing a data time series by a sensor; applying at least one linear filter function with a resolution L to each point in time of the data time series and obtaining a characteristic value for quantifying the temporal development of the data time series for each resolution L; detecting a trend in the data time series when one of the characteristic values of the data time series reaches a threshold value; and triggering a response of the system when a trend has been detected in the data time series.