Acoustic Fill-Level Detection in Irregular Containers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the fill level of containers, particularly waste containers, are inefficient, costly, and require homogeneous surfaces, making them unsuitable for complex or irregularly shaped containers and prone to measurement errors due to surface irregularities.
Innovation Solution
A device using acoustic signals in the audible range, combined with a machine learning model, analyzes the room impulse response of a container to determine fill levels, employing inexpensive components like loudspeakers and microphones, and integrates an AI unit for classification based on decision trees to provide a robust, adaptable, and precise fill level measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If time-of-flight measurement methods are used to determine fill level, then measurement precision can be improved, but the method requires homogeneous surfaces which are not available in complex or irregularly shaped containers
Solution Approach 1:
The patent replaces mechanical contact-based level sensors with an acoustic field-based measurement system. An acoustic signal source emits sound waves that propagate through the container, and a signal receiver detects the reflected or transmitted acoustic signals. This substitution eliminates the need for physical contact with the substance and homogeneous surfaces, enabling accurate fill level measurement in complex container geometries.
Solution Approach 2:
The acoustic measurement system is designed to be universally applicable to various container types and substances. By using acoustic wave propagation and reflection characteristics, the system can adapt to different container shapes, sizes, and filling materials without requiring recalibration or modification, thus achieving both precision and versatility.
2Device complexity
If traditional fill level detection methods are used, then device complexity is reduced, but measurement errors increase due to surface irregularities
Solution Approach 1:
The patent replaces complex mechanical contact sensors with a simpler acoustic field-based system. The acoustic signal source and receiver measure fill level by analyzing sound wave propagation and reflection, eliminating the need for mechanical components that are sensitive to surface irregularities. This substitution reduces device complexity while improving measurement reliability.
3Ease of manufacture
If contact-based sensors are used for fill level detection, then manufacturing cost is reduced, but component wear increases and maintenance requirements increase
Solution Approach 1:
The patent replaces contact-based mechanical sensors with non-contact acoustic sensors. The acoustic signal source emits sound waves that travel through the container and reflect off the substance surface, which is detected by the acoustic receiver. This non-contact measurement method eliminates mechanical wear and tear, extending component service life while maintaining ease of manufacture through the use of standard acoustic components.
4Ease of manufacture
If acoustic signals in the audible range are used, then component cost is reduced using standard loudspeakers and microphones, but signal susceptibility to environmental interference increases
Solution Approach 1:
The patent uses acoustic signals in the audible frequency range (20 Hz to 20 kHz) instead of ultrasonic frequencies, enabling the use of inexpensive standard loudspeakers and microphones. To mitigate environmental interference, the system processes the acoustic signals through algorithms that filter out background noise and distinguish the reflection signals from the emission signals based on their temporal and spectral characteristics.
5Measurement precision
If machine learning models are integrated into the device, then fill level categorization precision is improved, but device complexity increases
Solution Approach 1:
The patent integrates a machine learning model directly into the processing unit of the fill level detection device. The model is trained offline using labeled data and then deployed as a self-contained component within the device. During operation, the model automatically processes the acoustic signals and provides fill level categorization without requiring external computing resources, enabling the device to serve itself with advanced intelligence while maintaining manageable complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system offers a cost-effective, robust, and adaptable method for contactless fill level detection in various container types, reducing component wear, minimizing measurement errors, and providing precise fill level categorization, even in complex geometries, with low susceptibility to contamination and environmental interference.
Implementation Method 1
an acoustic signal source for emitting an acoustic transmission signal in the audible range based on the feed signal generated by the signal generator
Implementation Method 2
The reception signal is based on an isolated impulse response, which is the basis of the room impulse response of the container and its fill level
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A device (10) for detecting the fill level of a container (40) filled with a substance comprises a processing unit (50) with a microcontroller (52); a signal generator (58) for generating a feed signal for feeding a signal source (54); an acoustic signal source (54) for emitting an acoustic transmission signal in the audible range based on the feed signal generated by the signal generator (58); a signal receiver (60) for receiving an acoustic reception signal; and a storage unit (64) for temporarily storing the transmission signal and/or the reception signal. The signal generator (58) is designed to generate a sine signal in a predetermined frequency range. The processing unit (50) is designed to calculate the room impulse response from the reception signal and the transmission signal and thus to determine the fill level of the container (40).The determination of the fill level is based on classification data from a machine learning model (22) of an AI unit (20).