Intelligent compactness detection system and method integrated on road roller

Abstract

The invention belongs to the technical field of engineering machinery intelligence and construction quality detection, and provides an intelligent compaction degree detection system and method integrated on a road roller. The detection system comprises a multi-mode sensing unit which comprises a vibration sensing module, a machine vision module, an inertia measurement positioning module and a flatness detection module; the detection method comprises the following steps: S1, synchronously collecting data; s2, vibration signal processing and compaction index calculation; s3, image analysis and apparent feature extraction; s4, a multi-source data fusion and machine learning model; s5, positioning and drawing; in conclusion, the invention provides the intelligent compaction degree detection system and method integrated on the road roller. The method is reasonable in design and simple to operate, can realize real-time, comprehensive, lossless and high-precision detection of the compaction degree, and can evaluate the construction apparent quality at the same time, so that the method is suitable for industrial popularization.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent construction machinery and construction quality detection technology, and in particular to an intelligent compaction degree detection system integrated on a road roller and a method thereof. BACKGROUND

[0002] In the construction of infrastructure such as highways, railways, airports and dams, the compaction degree of fill material is a key core indicator for measuring the construction quality of roadbed and pavement base. The traditional compaction degree detection methods mainly rely on post-sampling detection, such as the ring knife method, sand replacement method, and nuclear density gauge method. These methods have the following inherent defects: first, point sampling and insufficient representation: only discrete points can be detected, and the compaction quality of the entire work surface cannot be fully reflected, and weak areas are prone to be missed; second, the detection results are slow to issue, and cannot guide the rolling operation in real time. If unqualified points are found, rework is required, which is inefficient and costly; there are also damage detection / safety problems: the ring knife method and the sand replacement method will damage the compacted pavement; the nuclear density gauge has radiation safety hazards, and the operation requires professional qualifications and strict management; and the method is extremely dependent on manual operation: the degree of automation is low, and it is greatly affected by human factors; To solve the above problems, continuous compaction control technology has appeared in recent years, which indirectly evaluates the compaction degree by measuring the vibration response of the steel wheel of the road roller. However, this technology still has limitations: the evaluation results are easily affected by factors such as road roller model, driving speed, material type, etc., and it is an indirect and relative measurement, which usually needs to be calibrated with traditional methods, and the absolute accuracy needs to be improved.

[0003] Therefore, we propose an integrated system that can detect the compaction degree in real time, comprehensively, non-destructively and with high precision, and can also evaluate the construction surface quality (such as flatness). SUMMARY

[0004] In view of the defects in the prior art, the present application provides an intelligent compaction degree detection system integrated on a road roller and a method thereof to solve the above problems.

[0005] In a first aspect, the present application provides an intelligent compaction detection system integrated on a road roller, comprising: a multi-modal sensing unit, the multi-modal sensing unit comprising a vibration sensing module, a machine vision module, an inertial measurement positioning module and a flatness detection module; the vibration sensing module is used for collecting vibration acceleration signals when the road roller is working; the machine vision module comprises at least one camera; wherein the camera is installed at a high position of a rear frame of the road roller, and faces the surface that has been rolled behind the steel wheel; it is used for collecting image sequences of the rolled surface; the inertial measurement positioning module comprises at least one positioning and attitude determination system integrated with a global navigation satellite system receiver and an inertial measurement unit; and are both installed near the center position of the road roller, for real-time acquisition of three-dimensional coordinates, driving speed, heading angle of the road roller, and pitch angle and roll angle of the vehicle body; the flatness detection module is electrically connected with the machine vision module, and the geometric deformation of the images between adjacent frames is analyzed by computer vision algorithm to calculate the longitudinal flatness estimate of the road surface; a data acquisition and processing industrial computer, the data acquisition and processing industrial computer is installed in the cab of the road roller, and is used for data acquisition and output results; a display interaction unit comprising an industrial touch screen display installed in the cab, and electrically connected with the data acquisition and processing industrial computer, and used for real-time display of the output results of the data acquisition and processing industrial computer; and a fixed protection device, the fixed protection device comprising a protective shell; the protective shell is used for protecting the machine vision module and the inertial measurement positioning module.

[0006] Further, the vibration sensing module adopts a three-axis acceleration sensor, which is installed on the bearing seat of the vibrating steel wheel or the frame of the road roller; the three-axis acceleration sensor is used for collecting vibration information. In actual application, this signal is the basis for calculating traditional compaction indicators (such as CMV, CCV, Evd, etc.).

[0007] Further, the flatness detection module further comprises a ranging array, which is arranged in front of the frame of the road roller, and is used for measuring the relative distance between the vehicle body and the road surface. In actual application, the transverse and longitudinal flatness is obtained by directly measuring the relative distance between the vehicle body and the road surface.

[0008] Further, the ranging array adopts a plurality of laser range finders or a plurality of ultrasonic sensors.

[0009] Further, the fixed protection device further comprises no less than one mounting bracket, and the mounting bracket is respectively used for the vibration sensing module, the machine vision module and the inertial measurement positioning module.

[0010] From the above technical solution, the present application provides an intelligent compaction detection system integrated on a road roller, which has the following beneficial effects: (1) Real-time continuous measurement: The compaction degree and flatness detection are completed synchronously during the rolling process, realizing "rolling while measuring".

[0011] (2) Surface-wide evaluation: The two-dimensional cloud map of compaction degree and flatness is generated by scanning the entire rolling area, avoiding missed detection.

[0012] (3) High precision and reliability: Combining apparent image information and inertial measurement data, the traditional vibration measurement is compensated and corrected, improving the absolute accuracy and reliability of the measurement results.

[0013] (4) Intelligence and visualization: Automatically processing data, generating intuitive quality reports and construction maps to guide the driver to optimize the rolling process.

[0014] (5) Especially, the camera used is an industrial high-definition camera, which functions to collect high-definition images and video sequences of the road surface after rolling. It is mainly used for subsequent analysis of road surface texture, macrostructure, surface defects, and as an auxiliary positioning reference; (6) And its inertial measurement positioning module: contains a positioning and orientation system integrated with GNSS (Global Navigation Satellite System) receiver and IMU (Inertial Measurement Unit). It is installed near the center of the vehicle body and is used to obtain the three-dimensional coordinates (longitude, latitude, and elevation), speed, heading angle, and pitch and roll angles of the road roller in real time. This module provides a high-precision time and space reference for the entire system; (7) At the same time, the flatness detection module innovatively uses the image sequence collected by the machine vision module to analyze the geometric deformation of adjacent frames through computer vision algorithms (such as optical flow method, feature point tracking), and calculates the longitudinal flatness of the road surface (such as the estimated value of international flatness index IRI).

[0015] In the second aspect, the present application provides an intelligent compaction degree detection method integrated on a road roller, comprising the following steps: Step 1: Synchronous data acquisition, through the interface of the above-mentioned components, synchronously receive the original data from all sensors, and mark with a uniform time stamp; Step 2: Vibration signal processing and compaction index calculation, filter and Fourier transform the acceleration signal, and calculate a series of compaction index values; Step 3: Image analysis and apparent feature extraction: process the images collected by the camera and extract texture features related to compaction degree; and calculate the flatness index; wherein, the texture features (including contrast and entropy of gray level co-occurrence matrix), macrostructure features, etc. Also includes the features of rough surface texture and distinct aggregate particle edges when the newly laid asphalt mixture is not compacted enough; and the features of smooth surface and fine texture after sufficient compaction.

[0016] Step 4: Multi-source data fusion and machine learning model; by inputting the original data, compaction index value and flatness index into the multi-source data fusion and machine learning model, and outputting the nonlinear mapping relationship between the multi-dimensional features and the true compaction degree, the absolute accuracy and working condition adaptability of the measurement are improved; wherein the machine learning model (such as random forest, gradient boosting tree or neural network, etc.).

[0017] Step 5: Positioning and mapping: binding the predicted compaction degree value, flatness value of each measurement point and the accurate coordinate provided by the inertial measurement positioning module, generating the compaction degree distribution cloud map and flatness distribution map of the entire working surface, and outputting to the industrial touch screen display of the display interaction unit.

[0018] From the above technical solution, the intelligent compaction degree detection method integrated on the road roller provided by the present application has the following beneficial effects: (1) High measurement accuracy: breaking through the limitations of single vibration measurement technology, by fusing machine vision appearance information and attitude data, using machine learning model for compensation and correction, the absolute accuracy and adaptability of compaction degree measurement for different materials are significantly improved.

[0019] (2) Comprehensive information: one-time acquisition of two key quality indicators of compaction degree and flatness, and provision of intuitive two-dimensional distribution cloud map, realizing visual global management of construction quality.

[0020] (3) Strong real-time, guiding construction: truly realizing process quality control, the operator can adjust the rolling process (speed, number of passes, vibration mode) in time according to real-time feedback, greatly reducing the risk and cost of subsequent rework.

[0021] (4) High degree of automation and intelligence: from data acquisition, processing, analysis to mapping and alarming, the whole process is automatically completed, reducing the dependence on professional testing personnel and human error.

[0022] (5) Non-destructive and safe: completely non-contact, non-destructive detection, no radiation and other safety hazards, environmentally friendly and safe. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the specific embodiments of the present application, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. In all the drawings, the elements or parts are not necessarily drawn according to the actual proportion.

[0024] Fig. 1 The schematic diagram of the present application installed on the road roller is provided for the embodiments of the present application; Fig. 2 The schematic diagram of the intelligent compaction degree detection system integrated on the road roller is provided for the present application; Fig. 3 An intelligent compaction detection method integrated on a road roller is provided. DETAILED DESCRIPTION

[0025] The embodiments of the technical solutions of the present application will be described in detail below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and therefore only serve as examples, and cannot limit the protection scope of the present application.

[0026] The embodiments are basically as shown in the accompanying drawings: Figs. 1 to 3 ​In a first aspect, the present application provides an intelligent compaction degree detection system integrated on a road roller, comprising: a multi-modal sensing unit, the multi-modal sensing unit comprising a vibration sensing module, a machine vision module, an inertial measurement positioning module and a flatness detection module; the vibration sensing module is used for collecting vibration acceleration signals during the operation of the road roller; the machine vision module comprises at least one camera; wherein the camera is installed at a high position of the rear frame of the road roller, and faces the surface that has been rolled behind the steel wheel; it is used for collecting image sequences of the rolled road surface; the inertial measurement positioning module comprises at least one positioning and pose system integrated with a global navigation satellite system receiver and an inertial measurement unit; and are both installed near the central position of the road roller, for real-time acquisition of the three-dimensional coordinates, the driving speed, the heading angle of the road roller, and the pitch angle and roll angle of the vehicle body; the flatness detection module is electrically connected with the machine vision module, and the geometric deformation of the images between adjacent frames is analyzed through computer vision algorithm to inversely calculate the longitudinal flatness evaluation of the road surface; a data acquisition and processing industrial computer, the data acquisition and processing industrial computer is installed in the cab of the road roller, and is used for data acquisition and output results; a display interaction unit comprising an industrial touch screen display installed in the cab, and electrically connected with the data acquisition and processing industrial computer, and used for real-time display of the output results of the data acquisition and processing industrial computer; and a fixed protection device, the fixed protection device comprising a protective shell; the protective shell is used for protecting the machine vision module and the inertial measurement positioning module. Real-time continuous measurement is realized: the detection of compaction degree and flatness is completed synchronously during the rolling process, realizing "rolling while measuring". Face type global evaluation: through scanning detection of the entire rolling area, a two-dimensional cloud map of compaction degree and flatness is generated, avoiding missed detection. High precision and reliability: combining apparent image information and inertial measurement data, the traditional vibration measurement is compensated and corrected, improving the absolute accuracy and reliability of the measurement results. Intelligence and visualization: automatically processing data, generating intuitive quality reports and construction maps to guide the driver to optimize the rolling process. Especially, the camera used is an industrial high-definition camera, which is used to collect high-definition images and / video sequences of the rolled road surface. It is mainly used for subsequent analysis of road surface texture, macrostructure, surface defects and as an auxiliary positioning reference; and the inertial measurement positioning module: contains a positioning and pose system integrated with a GNSS (Global Navigation Satellite System) receiver and an IMU (Inertial Measurement Unit). It is installed near the central position of the vehicle body, for real-time acquisition of the three-dimensional coordinates (longitude, latitude, elevation), driving speed, heading angle of the road roller, and pitch angle and roll angle of the vehicle body. The module provides a high-precision space-time reference for the entire system; at the same time, the flatness detection module innovatively proposes that the image sequences collected by the machine vision module can be directly used to analyze the geometric deformation of the images between adjacent frames through computer vision algorithms (such as optical flow method, feature point tracking), and inversely calculate the longitudinal flatness of the road surface (such as the evaluation of international flatness index IRI).

[0027] In the embodiment, the vibration sensing module adopts a three-axis acceleration sensor, which is installed on a roller bearing seat or a vehicle frame of the road roller; the three-axis acceleration sensor is used to collect vibration information. In actual application, the signal is the basis for calculating traditional compaction indexes (such as CMV, CCV, Evd, etc.).

[0028] In the embodiment, the flatness detection module further comprises a distance measuring array, which is arranged in front of the vehicle frame of the road roller and is used to measure the relative distance between the vehicle body and the road surface. In actual application, the relative distance between the vehicle body and the road surface is directly measured, and the transverse and longitudinal flatness is calculated.

[0029] In the embodiment, the distance measuring array is composed of a plurality of laser range finders or a plurality of ultrasonic sensors.

[0030] In the embodiment, the fixed protection device further comprises no less than one mounting bracket, which is respectively used for the vibration sensing module, the machine vision module and the inertial measurement positioning module.

[0031] In a second aspect, the present application provides an intelligent compaction degree detection method integrated on a road roller, which comprises the following steps: Step 1: data synchronous acquisition, through the interface of the above-mentioned components, the original data from all sensors is synchronously received, and a uniform time stamp is added; Step 2: vibration signal processing and compaction index calculation, the acceleration signal is filtered and Fourier transformed, and a series of compaction index values are calculated; Step 3: image analysis and apparent feature extraction, the image collected by the camera is processed, and the texture features related to the compaction degree are extracted; and the flatness index is calculated; wherein the texture features (including contrast and entropy value of gray level co-occurrence matrix), macrostructure features, etc. are included. At the same time, the features of rough surface texture and distinct edges and corners of aggregate particles when the newly paved asphalt mixture is not compacted enough, and the features of smooth surface and fine texture when the compaction is sufficient are also included.

[0032] Step 4: multi-source data fusion and machine learning model; the original data, the compaction index values and the flatness index are input into the multi-source data fusion and machine learning model, and the nonlinear mapping relationship between the multi-dimensional features and the true compaction degree is output, so as to improve the absolute accuracy and working condition adaptability of the measurement; wherein the machine learning model (such as random forest, gradient boosting tree or neural network, etc.) Step 5: Positioning and Mapping: Bind the predicted compaction value and the flatness value of each measurement point with the accurate coordinates provided by the inertial measurement positioning module to generate a compaction distribution cloud map and a flatness distribution map of the entire work surface, and output to the industrial touch screen display of the display interaction unit. High measurement accuracy: Breakthrough the limitations of single vibration measurement technology, through the fusion of machine vision appearance information and attitude data, using machine learning model for compensation and correction, significantly improving the absolute accuracy of compaction measurement and the adaptability to different materials. Comprehensive information: Simultaneously obtain two key quality indicators of compaction and flatness, and provide intuitive two-dimensional distribution cloud map, realize the visual global management of construction quality. Strong real-time, guide construction: Realize the process quality control, the operator can adjust the rolling process (speed, number of passes, vibration mode) in time according to the real-time feedback, greatly reduce the risk and cost of subsequent rework. High degree of automation and intelligence: From data acquisition, processing, analysis to mapping and alarm, the whole process is completed automatically, reducing the dependence on professional testing personnel and human error. Non-destructive and safe: Non-contact, non-destructive testing, no radiation and other safety hazards, environmentally friendly and safe.

[0033] In actual work: System installation: Fix the vibration acceleration sensor on the bearing seat of the vibrating steel wheel through the magnetic seat or bolt.

[0034] Install the high-definition industrial camera on the rear beam of the road roller through a special bracket, adjust its pitch angle to ensure that its field of view can completely cover the rolled road surface within 1.5-3 meters behind the steel wheel. Add protective cover and automatic windshield wiper.

[0035] Install the GNSS / IMU combined antenna on the roof in an open and unobstructed place.

[0036] Fix the industrial computer in the cab, connect the power supply and all sensor cables.

[0037] Install the touch screen display at a position convenient for the operator to observe and operate.

[0038] 2. System calibration: During the initial rolling stage of the construction section, select several representative points.

[0039] The system is running normally, and all sensor data of these points are recorded.

[0040] At the same time, measure the true dry density and compaction of these points by traditional sand pouring method.

[0041] Use the true compaction data as a label, and input the multi-sensor feature data of the corresponding points into the machine learning model for training to obtain a compaction prediction model specific to the current material and working condition.

[0042] 3. Normal work: The roller starts working, and the system automatically powers on and runs.

[0043] The industrial computer synchronously collects all data, and calculates vibration indicators, image features, and flatness in real time.

[0044] The extracted features are input into the trained machine learning model, and the predicted compaction degree value is output in real time.

[0045] All data are matched with spatial coordinates, and the compaction degree cloud chart and flatness chart are dynamically updated.

[0046] The operator monitors the quality in real time through the display screen. When the rolling frequency of a certain area is sufficient and the compaction degree reaches the set target value, the system prompts green; for low compaction areas, red alarm is prompted to guide re-rolling.

[0047] In summary, the intelligent compaction degree detection system and method integrated on the roller are not only reasonable in design, but also simple to operate, can realize real-time, comprehensive, non-destructive, and high-precision detection of the compaction degree, and can simultaneously evaluate the apparent quality of construction, so the application is suitable for industry promotion.

[0048] In the specification of the present application, a large number of specific details are described. However, it can be understood that the embodiments of the present application can be practiced without these specific details. In some examples, well-known methods, structures, and techniques are not shown in detail in order not to obscure the understanding of the present specification. The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the specification of the present application.

Claims

1. An intelligent compaction detection system integrated on a roller, characterized in that, The application relates to an intelligent compaction degree detection system integrated on a road roller. The system comprises a multi-modal sensing unit, a data acquisition and processing industrial computer, a display interaction unit and a fixed protection device. The multi-modal sensing unit comprises a vibration sensing module, a machine vision module, an inertial measurement positioning module and a flatness detection module. The vibration sensing module is used for collecting vibration acceleration signals of the road roller during operation. The machine vision module comprises at least one camera. The camera is installed on a high position of a rear frame of the road roller and faces a surface that has been rolled by a steel wheel. The camera is used for collecting image sequences of the rolled surface. The inertial measurement positioning module comprises at least one positioning and pose system integrated with a global navigation satellite system receiver and an inertial measurement unit. The positioning and pose system is installed near a central position of the road roller and is used for acquiring three-dimensional coordinates, a driving speed, a heading angle, a pitch angle and a roll angle of the road roller in real time. The flatness detection module is electrically connected with the machine vision module and is used for calculating a longitudinal flatness estimation value of the road surface by analyzing geometric deformation of images between adjacent frames through a computer vision algorithm.

2. The intelligent compaction system integrated with a roller according to claim 1, wherein, The data acquisition and processing industrial computer is installed in a cab of the road roller and is used for data acquisition and output of results.

3. The intelligent compaction system integrated with a roller according to claim 1, wherein, The display interaction unit comprises an industrial touch screen display installed in the cab and is electrically connected with the data acquisition and processing industrial computer and is used for displaying the output results of the data acquisition and processing industrial computer in real time.

4. The intelligent compaction system integrated with a roller according to claim 3, wherein, The fixed protection device comprises a protection shell and is used for protecting the machine vision module and the inertial measurement positioning module.

5. The intelligent compaction system integrated with a roller according to claim 1, wherein, The vibration sensing module adopts a three-axis acceleration sensor which is installed on a bearing seat of a vibrating steel wheel or a frame of the road roller and is used for collecting vibration information.

6. A method of integrated intelligent compaction on a roller, characterized in that, The flatness detection module further comprises a ranging array which is arranged in front of a frame of the road roller and is used for measuring a relative distance between the frame and the road surface. The ranging array comprises a plurality of laser range finders or a plurality of ultrasonic sensors. The fixed protection device further comprises at least one mounting bracket which is respectively used for the vibration sensing module, the machine vision module and the inertial measurement positioning module. The intelligent compaction degree detection system integrated on the road roller is used for the following steps. Step one: data synchronous acquisition, original data from all sensors are received through interfaces of the components and are marked with uniform time stamps; Step two: vibration signal processing and compaction index calculation, acceleration signals are filtered and Fourier transformed to calculate a series of compaction index values; Step three: image analysis and apparent feature extraction, images collected by the camera are processed to extract texture features related to the compaction degree; And a flatness index is calculated; Step four: multi-source data fusion and machine learning model, original data, compaction index values and the flatness index are input into the multi-source data fusion and machine learning model, and a nonlinear mapping relationship between multi-dimensional features and real compaction degrees is output to improve absolute accuracy and working condition adaptability of measurement. Step five: positioning and mapping: the predicted compaction value and the flatness value of each measurement point are bound with the accurate coordinates provided by the inertial measurement positioning module to generate the compaction distribution cloud map and the flatness distribution map of the entire work surface, and output to the industrial touch screen display of the display interaction unit.

Images