Ultrasound Object Localization Using Compressed Sensing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Ultrasonic sensors in driver assistance systems face challenges in determining the direction of ultrasonic echoes due to non-directional sound propagation, requiring improved directivity and signal processing methods to accurately locate objects.

Innovation Solution

A method involving the transmission of ultrasonic pulses, detection with multiple receivers, conversion to digital signals, summation, multiplication by a random matrix, and direction determination using compressed sensing, combined with beamforming algorithms, to accurately determine the direction of incidence and store object locations in an environment map.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If sound propagates in a non-directional manner, then ultrasonic sensors can detect obstacles from any direction, but the direction of incidence cannot be determined easily

Engineering Contradiction:
Improveomnidirectional detection capabilityVSAvoiddirection of incidence determination
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent divides the detection task into two independent components: distance measurement (using time-of-flight) and direction determination (using beamforming with multiple receivers). This segmentation allows each component to be optimized separately, maintaining omnidirectional coverage while achieving precise direction measurement through array signal processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from single-receiver detection to multi-receiver array detection, adding spatial dimensionality to the system. By arranging multiple receivers in specific geometric patterns (linear array, circular array, or planar array), the system gains the ability to determine direction of incidence while maintaining omnidirectional detection coverage.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If multiple receivers are used to determine direction of incidence, then directional accuracy improves, but device complexity increases

Engineering Contradiction:
Improvedirection of incidence determinationVSAvoidnumber of receivers and signal processing
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies compressed sensing techniques to determine direction of incidence with a reduced number of receivers compared to traditional beamforming requirements. By using sparse signal reconstruction algorithms, the system achieves accurate direction measurement with fewer sensors, reducing device complexity while maintaining measurement precision.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent replaces complex hardware configurations with sophisticated signal processing algorithms. Instead of using numerous receivers arranged in complex patterns, the system uses advanced algorithms (compressed sensing, sparse reconstruction) to extract directional information from limited receiver data, substituting computational complexity for hardware complexity.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Device complexity

If compressed sensing is used to determine direction of incidence, then the number of receivers can be reduced, but signal processing complexity increases

Engineering Contradiction:
Improvenumber of receiversVSAvoidsignal processing complexity
Core Design Contradiction:
Device complexityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent changes the processing approach from traditional frequency-domain beamforming to time-domain compressed sensing with sparse reconstruction. By formulating direction determination as a sparse optimization problem and using iterative algorithms (such as L1-minimization or greedy algorithms like OMP), the system achieves accurate direction estimation with fewer receivers, trading algorithmic complexity for reduced hardware requirements.

Inventive Principle:
Principle #35Parameter changes

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

This approach enhances the resolution and accuracy of object localization, allowing for a reduced number of receivers and improved signal processing, enabling precise object detection for driver assistance systems.

Implementation Method 1

ultrasonic transmitter emits a signal, which is reflected by an obstacle and then registered by a receiver

Methodology Applied
Scientific EffectSound propagation: Sound

Implementation Method 2

Capturing ultrasound echoes of the ultrasound pulses with at least two receivers

Methodology Applied
Scientific EffectEcho: Echo

Implementation Method 3

Converting the analog signals generated by each receiver from the ultrasound echoes into digital signals

Methodology Applied
Scientific EffectAnalog-to-digital conversion:

Data Source

PatentEP2817656B1Method for locating objects using ultrasound
Publication Date: 2019.12.11 ROBERT BOSCH GMBH
  • EP2817656B1 patent drawingFigure 1
  • EP2817656B1 patent drawingFigure 2a~2c
  • EP2817656B1 patent drawingFigure 3

AI summary

The invention relates to a method and a device for locating objects using ultrasound. The method has the following steps: a) emitting at least one ultrasonic pulse (22), b) detecting ultrasonic echoes (24, 25) of the at least one ultrasonic pulse (22) using at least two receivers (14), c) converting the analog signals (28) generated from the ultrasonic echoes (24, 25) by each receiver (14) into digital signals (31), d) adding the digital signals (31) into an output signal (38), e) multiplying the output signal (38) by a random matrix, f) ascertaining the direction of incidence of the ultrasonic echoes (24, 25) from the multiplied output signal (39) using a compressed sensing process, and g) storing the ascertained directions of incidence in a map (42) of the surroundings.