A road damage comfort evaluation method

Abstract

The present invention relates to a road disease comfort evaluation method, comprising the following steps: obtaining road surface appearance information and vehicle vibration data of a road through a test vehicle; identifying road diseases based on road surface appearance information and obtaining an appearance attribute set of the road diseases; obtaining a road wheel track belt area based on road surface appearance information; judging whether the road disease is located in the road wheel track belt area, if so, proceeding to the next step, otherwise judging that the comfort of the road disease is not falling into the regional comfort; extracting the vehicle vibration data of the corresponding time according to the appearance attribute set to obtain the vehicle instantaneous vibration impact value as a comfort evaluation index. Compared with the prior art, the present invention can objectively evaluate the driving comfort of the road, with good accuracy and consistent standards.

Description

Technical Field

[0001] The invention relates to the field of road quality detection, and in particular to a method for evaluating the comfort level of road damage. Background Art

[0002] The evaluation of road damage is of great significance to both road maintenance departments and travel users. The former uses accurate diagnosis of road damage status to reasonably judge the trend of road performance changes and scientifically formulate management and maintenance plans. The latter will consider the impact of road damage on travel when choosing travel routes, so as to reasonably plan travel routes. In the past few decades, many evaluation methods for road damage have been proposed and adopted, which can be mainly divided into three methods: rating method, weighted scoring method and deduction method. The earliest proposed is the Pavement Serviceability Rating (PSR) index, which is a rating method. It evaluates road conditions based on passengers' perceived rating of ride quality when driving on a certain road section. Similarly, the Pavement Condition Ratio (PCR) uses expert rating method to make an overall rating evaluation of road conditions. Representative weighted scoring methods are the Pavement Serviceability Index (PSI) and the Fuzzy Distress Index (FDI). On the basis of PSR, people have added considerations for road cracks, rutting and repair of diseases, and established correlation equations with such diseases to calculate PSI. Since the need for expert groups is reduced, it can be widely used in road network quality evaluation. In addition, another widely used indicator is the Pavement Condition Index (PCI) developed by the U.S. Army Corps of Engineers. It belongs to the evaluation indicator of the deduction method type. The index describes the number, category and severity of diseases on a specific length of road section in the range of 0 to 100. It is evaluated by deduction, and the deduction weights of different categories, quantities and severity are determined by expert experience. In actual applications, for different road types in different regions, local management departments will modify the deduction weights based on PCI according to local characteristics. Although the above road disease evaluation methods vary depending on the application scenario, they have one thing in common: they all use a comprehensive weighted method to describe the average performance of the road section. These evaluation methods first need to segment the road, usually taking 100 meters or 1000 meters as an evaluation section, and use artificial weighting to obtain the final evaluation value. This segment weighting method will ignore the impact of extreme values, making it difficult to reflect the state of extremely serious diseases and difficult to guide refined and daily maintenance.

[0003] In addition, it is difficult to formulate weights scientifically and reasonably, which makes it impossible to compare different indicators. At present, many studies have been carried out to explore more reasonable weight determination methods, such as fuzzy reasoning models, data-driven models, and machine learning models. However, artificial weights lack practical significance and are difficult to effectively support long-term performance analysis. They cannot truly reflect the generation and deterioration process of each disease in the long-term service of the road. Therefore, we need an evaluation method that takes a single road disease as the evaluation object as a supplement to describe a more detailed road quality condition.

[0004] As people's demand for travel quality increases, they pay more and more attention to the comfort during travel. Passenger comfort is becoming more and more important in evaluating road diseases, especially in urban roads and autonomous driving scenarios. Evaluating road diseases from the perspective of comfort is not only conducive to road managers to grasp the actual service level of roads more intuitively and objectively, so as to maintain and maintain road quality more scientifically and effectively; it can also assist autonomous driving vehicles to reasonably plan driving routes and provide passengers with a more comfortable travel experience. However, in the research of driving comfort, it is mainly the automobile manufacturers who actively adjust and control, by detecting the vibration of the vehicle, adjusting the angle and direction of the seat to relieve the impact of bumps for passengers, but this evaluation object is concentrated on the interior or suspension of the vehicle, which belongs to passive reduction of driving discomfort. The existing evaluation methods based on road conditions mainly evaluate the impact on driving fatigue from the perspectives of road curves, road transverse and longitudinal slopes, and road landscapes, and rarely evaluate road diseases from the perspective of abnormal vibration of the vehicle caused by road diseases, thereby affecting comfort. Summary of the invention

[0005] The purpose of the present invention is to provide a road hazard comfort evaluation method in order to overcome the defects of the above-mentioned prior art.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A road damage comfort evaluation method comprises the following steps:

[0008] S1: Obtaining road surface information and vehicle vibration data through the test vehicle;

[0009] S2: Identify road damage based on road surface information and obtain a set of road damage surface attributes;

[0010] S3: obtaining the road wheel track area based on the road surface information;

[0011] S4: Determine whether the road defect is located in the road wheel track zone. If so, proceed to step S5. Otherwise, determine that the comfort level of the road defect is not within the zone comfort level.

[0012] S5: Extracting the vehicle vibration data at the corresponding time according to the apparent attribute set to obtain the vehicle instantaneous vibration impact value as a comfort evaluation index.

[0013] Preferably, the method further comprises:

[0014] S6: Rating the comfort level according to the comfort evaluation index.

[0015] Preferably, the step of comfort rating includes:

[0016] Obtain the comfort index range to which the comfort evaluation index belongs, and determine the comfort level of the road damage.

[0017] Preferably, the road surface information is road surface image data captured by a camera mounted on a test vehicle.

[0018] Preferably, the step S2 specifically includes:

[0019] Based on the recognition algorithm, the type, size and location of road defects in the road surface image are identified, and the acquisition timestamp of the road surface image data containing the road defects is obtained to form an apparent attribute set.

[0020] Preferably, the step S3 specifically includes:

[0021] S31: Identify the lane line position in the road surface information;

[0022] S32: With the lane lines on both sides as the starting point and the end point respectively, an area with a preset proportion is selected as the road wheel track area.

[0023] Preferably, the step S31 specifically includes:

[0024] S311: using a Gaussian filter to smooth the road surface image and remove image noise;

[0025] S312: Using the Canny operator to detect the edge position of the lane line;

[0026] S312: In the binary image of the Canny operator edge detection, the Hough line transform is used to extract the straight line in the image, which is used as the lane line. The pixel distance d between the lane lines on both sides is calculated, and the position with a distance from the left lane line in the interval [s1d, s2d] and [s3d, s4d] is selected as the road wheel track band area, where s1, s2, s3, and s4 are respectively the preset track band parameters.

[0027] Preferably, the step S5 specifically includes:

[0028] S51: Selecting a preset range before and after the image acquisition time of the road disease as a rough selection time window for vehicle vibration data;

[0029] S52: extracting the vehicle vibration data within the rough selection time window, extracting the high-frequency part of the vibration signal and obtaining abnormal data of the high-frequency part, and defining a precise time window according to the interval where the abnormal data is located;

[0030] S53: extracting the vehicle vibration data within the precise time window to calculate the instantaneous vibration impact value of the vehicle.

[0031] Preferably, the step S53 specifically includes:

[0032] S531: extracting stimulus response by sliding window without interval at the starting point of the precise time window;

[0033] S532: Calculate the vibration energy value according to the excitation response:

[0034]

[0035] Among them, a w (t0) is the equivalent acceleration at the end point t0, τ is the sliding window interval, t0 is the end point of the sliding window, a wz (t) is the vibration data at time t, and the sliding window calculation results constitute the sequence [a w (t1),a w (t2)......,a w (t n )];

[0036] S533: Calculate the maximum value in the sequence as the vehicle instantaneous vibration impact value of the road disease.

[0037] Preferably, the vehicle vibration data is three-axis acceleration of multiple measuring points on the test vehicle or three-axis velocity and three-axis displacement data of the acceleration that can be obtained by derivation.

[0038] Compared with the prior art, the present invention provides a method for evaluating the comfort of road defects. This method uses a single defect as the evaluation object, greatly improving the precision of road condition evaluation and maintenance, and can effectively reduce the deviation of road quality condition evaluation caused by individual extreme values ​​in traditional road section evaluation methods. In addition, this method makes an evaluation from the perspective of the impact of road defects on ride comfort. In addition to facilitating the road maintenance department to implement the functional attributes of road humanization and comfort, it also increases the consideration of driving comfort when passengers choose travel routes, which is of great significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a flow chart of the present invention;

[0040] Figure 2 A schematic diagram of a method for determining whether a road defect is in a wheel track area in an embodiment of the present invention;

[0041] Figure 3 A schematic diagram of determining a precise time window in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. Note that the following embodiments are merely illustrative in nature, and the present invention is not intended to limit its applicable objects or uses, and the present invention is not limited to the following embodiments.

[0043] Example

[0044] A method for evaluating the comfort level of road defects mainly realizes the identification, location and comfort evaluation of road defects through the fusion of image and vibration data. Defects in the road are detected using image data. When a defect is detected, the location of the defect is first determined through image data. If it occurs in the vehicle wheel track area, it will cause abnormal excitation vibration to the passing vehicles. At the same time, wavelet transform is used to extract the vibration data corresponding to the time, and instantaneous vibration impact is used to calculate the comfort level under the abnormal vibration condition as the comfort evaluation level of the defect. The identified road defect can also be a structure that affects the comfort level of the road surface, such as a manhole cover. Figure 1 As shown, the method comprises the following steps:

[0045] (1) Obtain road surface information and vehicle vibration data through test vehicles.

[0046] In this embodiment, an on-board industrial camera or a driving recorder is installed on the test vehicle to capture road surface images and obtain road surface surface information. The collected images are black and white grayscale images or color images. In addition, sensors are installed at the measuring points of the test vehicle to obtain the vibration data of the vehicle. The vibration data refers to the three-axis acceleration of each measuring point of the vehicle or the three-axis velocity and three-axis displacement data of the acceleration that can be obtained by derivation. The measuring points include the three placement points of the backrest, seat cushion, and footrest in the three positions of the driver's seat, the co-driver's seat, and the rear seat, totaling 9 measuring points. When conditions are limited, the measuring points can only select two measuring points on the left and right sides of the rear axle of the vehicle.

[0047] (2) Identify road hazards based on road surface appearance information and obtain a set of appearance attributes of road hazards. In this embodiment, the type, size, and location of road hazards in a road surface image are identified based on a recognition algorithm, and the acquisition timestamp of road surface image data containing road hazards is obtained to form an appearance attribute set D. In the present invention, a visual recognition algorithm such as a neural network can be used to identify road hazards in a road surface image and obtain the size and location, or it can be obtained by introducing artificial annotation information.

[0048] (3) Obtain the road wheel track area based on the road surface information.

[0049] The specific steps to obtain the road wheel track area are as follows: Figure 2 Shown are:

[0050] 31. Identify the lane line position in the road surface information;

[0051] 32. Take the lane lines on both sides as the starting point and the end point respectively, and select an area with a preset proportion as the road wheel track area.

[0052] In this embodiment, step 31 includes:

[0053] 311: Use Gaussian filter to smooth the road image and remove image noise;

[0054] 312: Use the Canny operator to detect the edge position of the lane line;

[0055] 312: In the binary image of the Canny operator edge detection, the Hough line transform is used to extract the straight line in the image, which is used as the lane line. The pixel distance d between the lane lines on both sides is calculated, and the position with a distance from the left lane line in the interval [s1d, s2d] and [s3d, s4d] is selected as the road wheel track zone, where s1, s2, s3, and s4 are the preset track zone parameters respectively.

[0056] (4) Determine whether the road defect is located in the road wheel track area. If so, proceed to step (5). Otherwise, determine that the comfort level of the road defect is not within the regional comfort level.

[0057] Specifically, this embodiment determines whether the location of the road defect falls within the wheel track area to determine whether the road defect is located in the road wheel track area. If the location of the road defect falls within the wheel track area, it is considered that the road defect is located in the wheel track area and will cause abnormal vibration to passing vehicles.

[0058] (5) The vehicle vibration data at the corresponding time is extracted according to the apparent attribute set to obtain the vehicle instantaneous vibration impact value as a comfort evaluation index.

[0059] 51: Select 3 seconds before and after the image acquisition time of the road disease as the rough selection time window T1 of the vehicle vibration data of abnormal vibration that may be caused by the disease V ;

[0060] 52: Extraction in the rough selection time window T1 VThe vehicle vibration data in the road is selected, and the appropriate wavelet basis function and decomposition layer number are selected to decompose the corresponding vehicle vibration data and obtain the high-frequency part of the vibration signal. The two abnormal data points of the high-frequency part are extracted from the high-frequency part of the vibration signal using the Laida criterion, and the time of the two abnormal data points is used as the precise starting and ending points of the abnormal vibration of the vehicle caused by the road disease, thereby determining the precise time window T2 of the vehicle vibration. V .

[0061] Extract the vehicle vibration data within the rough selection time window, extract the high-frequency part of the vibration signal and obtain the abnormal data of the high-frequency part, and define the precise time window T2 according to the interval where the abnormal data is located V ;

[0062] 53: Extract the vehicle vibration data within the precise time window to calculate the instantaneous vibration impact value of the vehicle and obtain the vehicle bumping condition. The vehicle bumping condition refers to the maximum value, minimum value, average value and jerk value of the acceleration in the XYZ direction calculated based on the vehicle vibration data corresponding to the time, and also includes the weighted acceleration root mean square value calculated separately at each measuring point.

[0063] Step 53 specifically includes:

[0064] 531: Select the time window, in the precise time window T2 V The starting point has no interval sliding window to extract the stimulus response;

[0065] 532: Calculate the vibration energy value based on the excitation response:

[0066]

[0067] Among them, a w (t0) is the equivalent acceleration at the end point t0, τ is the sliding window interval, t0 is the end point of the sliding window, a wz (t) is the vibration data at time t, and the sliding window calculation results constitute the sequence [a w (t1),a w (t2)......,a w (t n )];

[0068] 533: Calculate the maximum value in the sequence as the vehicle instantaneous vibration impact value of the road disease.

[0069] (6) Obtain the comfort index range to which the comfort evaluation index belongs and determine the comfort level of the road damage.

[0070] Specifically, the rating is performed according to the vehicle instantaneous vibration impact value range and the corresponding comfort level in Table 1 below.

[0071] Table 1 Comparison table of transient vibration shock (MTVV) and comfort level

[0072] <![CDATA[Instantaneous vibration shock / m / s 2 > People's subjective feelings Comfort level Less than 0.315 No discomfort A 0.315～0.63 Some discomfort B 0.5～1 Rather uncomfortable C 0.8～1.6 uncomfortable D 1.25～2.5 Very uncomfortable E Greater than 2 Very uncomfortable F

[0073] In this embodiment, when the road defect is not located in the road wheel track area, the corresponding comfort level of the area not falling into the area is comfort level B.

[0074] More specifically, in this embodiment, an on-board industrial camera is fixed to the rear window of the test vehicle to shoot the road image from above, and a vibration sensor is fixed above the rear axle of the vehicle to record the vibration of the test vehicle. The image capture frame rate is 5 frames per second, and the vibration data acquisition frame rate is 200Hz. A certain urban road is selected, and the road diseases on this section are all manhole covers. There are 18 manhole covers distributed continuously on this section, all located in the wheel track. The vehicle passes at a constant speed of 50km / h, and the image and vibration data are recorded at the same time.

[0075] The above steps are used to identify the road surface information, that is, the road image obtained by the on-board industrial camera, to obtain the apparent attribute set of road diseases. In this embodiment, s1, s2, s3, and s4 are respectively A road track belt area is formed.

[0076] In this embodiment, in step (5), the vehicle vibration data of 3 seconds before and after is extracted, the three-layer decomposition of the vibration data by the db4 wavelet basis function is selected, and its high-frequency part is extracted. The abnormal interval is determined according to the Laida criterion, and the corresponding vibration is extracted. Figure 3 shown.

[0077] In step 53 of this embodiment, 0.1 seconds is selected as the time window, and the sliding window is used to calculate the vibration energy value a of the vibration data within the precise time window. w (t0), and in the precise time window T2 V a calculated based on the sliding window w The maximum value is selected from (t0) to obtain the MTVV value as the comfort value caused by the disease. The MTVV value and comfort level corresponding to each manhole cover are determined as shown in the following table:

[0078] Table 2 Evaluation results of comfort level of manhole cover in Example 1

[0079]

[0080] The above embodiments are merely examples and do not limit the scope of the present invention. These embodiments can also be implemented in various other ways, and various omissions, substitutions, and changes can be made without departing from the technical concept of the present invention.

Claims

1. A road damage comfort evaluation method, characterized in that: The following steps are involved: S1: Obtaining road surface information and vehicle vibration data through the test vehicle; S2: Identify road damage based on road surface information and obtain a set of road damage surface attributes; S3: obtaining the road wheel track area based on the road surface information; S4: Determine whether the road defect is located in the road wheel track zone. If so, proceed to step S5. Otherwise, determine that the comfort level of the road defect is not within the zone comfort level. S5: Extracting the vehicle vibration data at the corresponding time according to the apparent attribute set to obtain the vehicle instantaneous vibration impact value as a comfort evaluation index.

2. A road damage comfort evaluation method according to claim 1, characterized in that: The method further comprises: S6: Rating the comfort level according to the comfort evaluation index.

3. A road damage comfort evaluation method according to claim 2, characterized in that: The steps of comfort rating include: Obtain the comfort index range to which the comfort evaluation index belongs, and determine the comfort level of the road damage.

4. A road damage comfort evaluation method according to claim 1, characterized in that: The road surface information is road surface image data captured by a camera mounted on a test vehicle.

5. A road damage comfort evaluation method according to claim 4, characterized in that: The step S2 specifically includes: Based on the recognition algorithm, the type, size and location of road defects in the road surface image are identified, and the acquisition timestamp of the road surface image data containing the road defects is obtained to form an apparent attribute set.

6. A road damage comfort evaluation method according to claim 1, characterized in that: The step S3 specifically includes: S31: Identify the lane line position in the road surface information; S32: With the lane lines on both sides as the starting point and the end point respectively, an area with a preset proportion is selected as the road wheel track area.

7. A road damage comfort evaluation method according to claim 6, characterized in that: The step S31 specifically includes: S311: using a Gaussian filter to smooth the road surface image and remove image noise; S312: Using the Canny operator to detect the edge position of the lane line; S312: In the binary image of the Canny operator edge detection, the Hough line transform is used to extract the straight line in the image, which is used as the lane line. The pixel distance d between the lane lines on both sides is calculated, and the position with a distance from the left lane line in the interval [s1d, s2d] and [s3d, s4d] is selected as the road wheel track band area, where s1, s2, s3, and s4 are respectively the preset track band parameters.

8. A road damage comfort evaluation method according to claim 1, characterized in that: The step S5 specifically includes: S51: Selecting a preset range before and after the image acquisition time of the road disease as a rough selection time window for vehicle vibration data; S52: extracting the vehicle vibration data within the rough selection time window, extracting the high-frequency part of the vibration signal and obtaining abnormal data of the high-frequency part, and defining a precise time window according to the interval where the abnormal data is located; S53: extracting the vehicle vibration data within the precise time window to calculate the instantaneous vibration impact value of the vehicle.

9. A road damage comfort evaluation method according to claim 8, characterized in that: The step S53 specifically includes: S531: extracting stimulus response by sliding window without interval at the starting point of the precise time window; S532: Calculate the vibration energy value according to the excitation response: Among them, a w (t0) is the equivalent acceleration at the end point t0, τ is the sliding window interval, t0 is the end point of the sliding window, a wz (t) is the vibration data at time t, and the sliding window calculation results constitute the sequence [a w (t1),a w (t2)......,a w (t n )]; S533: Calculate the maximum value in the sequence as the vehicle instantaneous vibration impact value of the road disease.

10. A road damage comfort evaluation method according to claim 8, characterized in that: The vehicle vibration data is the three-axis acceleration of multiple measuring points on the test vehicle or the three-axis velocity and three-axis displacement data of the acceleration that can be obtained by derivation.

Images