Automatic driving mine card motion state abnormity monitoring and early warning method

Abstract

The invention relates to an automatic driving mine card motion state abnormity monitoring and early warning method, belongs to the technical field of equipment health management, solves the problem that the prior art lacks the early warning capability for the abnormal state of an automatic driving mine card, and comprises the steps: S1, collecting the motion state data of a mine card under the normal running condition; s2, extracting key information in the motion state data; s3, performing moving average processing on factors influencing the motion state in the extracted key information; step S4, processing the smoothed data to obtain 27 types of different working conditions; s5, setting two levels of abnormal monitoring thresholds for the data of the transverse deviation and the course deviation of each type of working conditions; s6, acquiring real-time data of the mine card during running, comparing the real-time data with the two-level anomaly detection threshold values, and judging the motion state level of the real-time data as a judgment result; and S7, taking corresponding countermeasures according to a judgment result.

Description

Technical Field

[0001] The present invention relates to the field of equipment health management, and more specifically to a method for monitoring and warning abnormal motion status of an autonomous driving mining truck. Background Art

[0002] With the advancement of science and technology, smart mines have become an important development direction of the modern mining industry. In open-pit mines with complex and changeable environments, the introduction of autonomous driving technology and the realization of unmanned transportation can avoid major casualties and minimize safety risks. At present, autonomous driving mining trucks are in a stage of rapid development. The working environment in mining areas is harsh and complex, which has a significant impact on the performance and status of transportation vehicles and is prone to safety accidents. In current engineering applications, potential problems in the operation of mining trucks are difficult to be discovered and handled in a timely manner. Usually, corresponding measures can only be taken when significant abnormal phenomena are manifested. There is a lack of early warning capabilities for abnormal vehicle conditions, which increases maintenance costs and downtime, reduces operating efficiency, and seriously restricts the overall production capacity of mining areas and the development of the industry.

[0003] In actual applications in mining areas, there is still a lack of effective methods to monitor the motion status of autonomous mining trucks. Among the existing methods, the widely used threshold monitoring system based on manual experience provides a basic evaluation framework for the motion status to a certain extent, but sometimes it is difficult to accurately capture the complex and changeable motion characteristics of mining trucks under different working conditions, resulting in frequent false alarms or missed alarms in actual applications. False alarms not only increase the number of unnecessary downtimes and affect production efficiency, but may also cause fatigue and distrust of operators; missed alarms mean that potential faults are not discovered in time, which may further evolve into serious safety accidents.

[0004] Due to the lack of effective monitoring and early warning methods, the maintenance of mining trucks relies more on experience and judgment rather than scientific analysis results. This situation not only reduces maintenance efficiency, but may also lead to unnecessary maintenance activities or ignore real potential failures. Although this experience-oriented maintenance model can deal with daily problems to a certain extent, it is difficult to deal with complex and changeable failure modes and potential risks, and it is easy to lead to one-sidedness and lag in maintenance decisions. On the one hand, excessive maintenance may cause waste of resources; on the other hand, it may also bury safety hazards due to ignoring real potential failures. Summary of the invention

[0005] In view of this, the present invention provides a method for monitoring and warning the abnormal motion state of an autonomous driving mining truck. Aiming at the problem that potential abnormal problems in the operation of mining trucks are difficult to be discovered and handled in a timely manner, a multi-level abnormal state monitoring and warning method that takes into account the operating conditions is proposed.

[0006] According to an embodiment of the present invention, a method for abnormal motion state monitoring and early warning of an autonomous driving mining truck is provided, comprising the following steps:

[0007] Step S1, collecting the motion state data of the mining truck in a normal running state through a laser radar, a camera, a GPS system or a sensor;

[0008] Step S2, extracting key information from the motion state data of a normal sports car, the key information including factors affecting the motion state and the performance of the motion state, the factors affecting the motion state including slope, road curvature and vehicle speed, the performance of the motion state including lateral deviation and heading deviation;

[0009] Step S3, performing sliding average processing on the factors affecting the motion state in the extracted key information to obtain smoothed data of the slope, road curvature and vehicle speed as the smoothed key information;

[0010] Step S4, dividing the obtained smoothed slope, road curvature and vehicle speed data into three intervals according to their size, recording the boundary value of each interval, and arranging and combining the three intervals of the three factors to obtain 27 different working conditions;

[0011] Step S5, classifying the motion state data of the normal sports car collected in step S1 according to the intervals divided in step S4, and respectively counting the data of lateral deviation and heading deviation for each of the 27 different working conditions, and using the box plot method to set two levels of abnormal monitoring thresholds for the data of lateral deviation and heading deviation of each working condition;

[0012] Step S6, obtaining real-time data of the mining truck running by laser radar, camera, GPS system or sensor, extracting the data of lateral deviation and heading deviation in the real-time data, and comparing them with the two-level abnormality detection threshold, and determining the motion state level of the real-time data as the judgment result;

[0013] Step S7, taking corresponding countermeasures according to the judgment result obtained in step S6. If the judgment result is a normal state, no action is required; if it is a potential abnormal state, an abnormal reminder is issued; if it is a significant abnormal state, an abnormal alarm is issued, the fault is reported and the vehicle is stopped.

[0014] Optionally, the step S4 specifically includes the following steps:

[0015] Step S4.1, setting the obtained smoothed slope, road curvature and vehicle speed data as slope s, road curvature c and vehicle speed v respectively;

[0016] Step S4.2, setting the data ranges of the slope, road curvature and vehicle speed to be evenly divided into three continuous sub-intervals, setting the high-end interval of the slope to s1, the middle interval of the slope to s2, the initial interval of the slope to s3, the high-end interval of the road curvature to c1, the middle interval of the road curvature to c2, the initial interval of the road curvature to c3, the high-end interval of the vehicle speed to v1, the middle interval of the vehicle speed to v2, and the initial interval of the vehicle speed to v3;

[0017] Step S4.3, after dividing the intervals, the sub-intervals of slope, road curvature and vehicle speed are arranged and combined to obtain 27 different working conditions: (s1, v1, c1), (s2, v1, c1), (s3, v1, c1), (s1, v2, c1), (s2, v2, c1), (s3, v2, c1), (s1, v3, c1), (s2, v3, c1), (s3, v3, c1), (s1, v1, c2), (s2, v1, c2), (s3 ,v1,c2), (s1,v2,c2), (s2,v2,c2), (s3,v2,c2), (s1,v3,c2), (s2,v3,c2), (s3,v3,c2), (s1,v1,c3), ( s2,v1,c3), (s3,v1,c3), (s1,v2,c3), (s2,v2,c3), (s3,v2,c3), (s1,v3,c3), (s2,v3,c3), (s3,v3,c3).

[0018] Optionally, in step S5: the two-level abnormal monitoring threshold includes a first-level threshold and a second-level threshold higher than the first-level threshold; wherein, being lower than the first-level threshold represents that the movement state of the mining truck is normal, exceeding the first-level threshold but not reaching the second-level threshold represents a potential abnormal state, and exceeding the second-level threshold represents a significant abnormal state.

[0019] Optionally, the step S6 specifically includes the following steps:

[0020] Step S6.1, obtaining real-time data of the mining truck running by laser radar, camera, GPS system or sensor;

[0021] Step S6.2, comparing the real-time data with the 27 different types of working conditions obtained in step S4 to determine the type of working condition;

[0022] Step S6.3, according to the category of the working condition, select the two-level abnormal monitoring threshold of the corresponding category from the result obtained in step S5;

[0023] Step S6.4, extract the data of lateral deviation and heading deviation in the real-time data, and compare them with the two-level abnormal monitoring thresholds of the corresponding category to obtain the motion state level as the judgment result, wherein, if the data of lateral deviation and heading deviation are lower than the first-level threshold, it indicates that the mining truck is in a normal state; if the data of lateral deviation and heading deviation exceed the first-level threshold but do not reach the second-level threshold, it indicates that the mining truck is in a potential abnormal state; if the data of lateral deviation and heading deviation exceed the second-level threshold, it indicates that the mining truck is in a significant abnormal state.

[0024] Compared with the prior art, a method for abnormal motion status monitoring and early warning of an autonomous driving mining truck provided by the implementation of the present invention can effectively monitor the motion status of the mining truck for abnormalities, greatly reducing the false alarm or missed alarm of the threshold monitoring system set based on artificial experience. The set multi-level thresholds can realize the timely discovery and handling of potential problems, have the ability to warn of abnormal conditions, and improve the efficiency of operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. The features and advantages of the present invention can be more clearly understood by referring to the drawings. The drawings are schematic and should not be understood as limiting the present invention in any way. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 The present invention provides a flowchart of a method for monitoring and warning abnormal motion status of an autonomous driving mining truck according to an embodiment of the present invention.

[0027] Figure 2 An effect diagram of dividing the slope, curvature and vehicle speed in an embodiment of the method for abnormal motion state monitoring and early warning of an autonomous driving mining truck provided by an embodiment of the present invention.

[0028] Figure 3 A schematic diagram of calculating two-level thresholds of lateral deviation and heading deviation based on the box plot method in an embodiment of an abnormal motion state monitoring and early warning method for an autonomous driving mining truck provided by an embodiment of the present invention.

[0029] Figure 4 A flowchart of an embodiment of a method for abnormal motion status monitoring and early warning of an autonomous driving mining truck provided by an embodiment of the present invention.

[0030] Figure 5 A schematic diagram of a normal situation in which the abnormal level is determined according to the excess size in an embodiment of the method for abnormal motion state monitoring and early warning of an autonomous driving mining truck provided by an embodiment of the present invention.

[0031] Figure 6 A schematic diagram of an abnormal reminder situation in which an abnormal level is determined based on an excess size in an embodiment of an abnormal monitoring and early warning method for the motion state of an autonomous driving mining truck provided by an embodiment of the present invention.

[0032] Figure 7 A schematic diagram of an abnormal alarm situation in which an abnormal level is determined based on an excess size in an embodiment of an abnormal monitoring and early warning method for the motion state of an autonomous driving mining truck provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

[0034] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0035] A method for monitoring and warning abnormal motion status of an autonomous driving mining truck according to an embodiment of the present invention is described in detail below with reference to the accompanying drawings.

[0036] like Figure 1 As shown, according to an embodiment of the present invention, a method for abnormal motion state monitoring and early warning of an autonomous driving mining truck is provided, comprising the following steps.

[0037] Step S1, collecting the motion state data of the mining truck under normal running conditions through data collection methods such as laser radar, camera, GPS system, and sensor.

[0038] Step S2, extracting key information from the motion state data of a normal sports car. The key information includes factors affecting the motion state and the performance of the motion state, wherein the factors affecting the motion state may include slope, road curvature, vehicle speed, etc.; the performance of the motion state may include lateral deviation, heading deviation, etc.

[0039] Among them, the slope is the ratio of the vertical height h of the mining truck on the road, i.e. the slope surface, to the horizontal width l, i.e. the tangent value of the slope angle (tan∠a value, where ∠a is the angle between the slope and the horizontal plane). The slope has a significant impact on the driving force, braking force and driving stability of the mining truck. The road curvature describes the degree of the curvature of the road, and shows the rate of change of the tangent direction angle at a point on the curve with the arc length through a differential definition. The size of the curvature is directly related to the control difficulty and stability requirements of the mining truck when driving on a curve. The speed refers to the speed of the mining truck during the road driving process, which is a physical quantity that measures the speed of the mining truck.

[0040] Lateral deviation refers to the difference between the actual route and the planned route when the mining truck is traveling along the prescribed route. It reflects the degree of deviation between the actual route and the planned route of the mining truck and is an important basis for evaluating the mining truck's trajectory tracking capability. Heading deviation refers to the deviation between the heading value predicted by the model in the prediction time domain and the expected trajectory heading, reflecting the direction control accuracy of the mining truck.

[0041] By comprehensively considering factors such as slope, road curvature, vehicle speed, lateral deviation and heading deviation, the motion control performance of the mining truck can be fully evaluated, and a strong guarantee can be provided for its safe, stable and accurate driving under different working conditions.

[0042] Step S3, sliding average processing is performed on the factors affecting the motion state in the extracted key information, such as slope, road curvature, and vehicle speed, to obtain smoothed key information. In order to ensure the accuracy of the obtained data and prevent interference from erroneous data, the data of the key information is subjected to sliding average processing, and the sliding window size can be set to 20, that is, the data at a certain moment and the following 20 data are grouped as a group to obtain the average value as the data value at that moment. The smoothed key information includes the smoothed slope, road curvature, and vehicle speed data.

[0043] This implementation uses a moving average method: the data points within a certain time window are averaged to smooth the data. The choice of window size has a significant impact on the smoothing effect. The larger the window, the more obvious the smoothing effect, but it may mask the true changes in the data. Data smoothing can eliminate noise and outliers in the data, making the overall trend and pattern of the data more obvious.

[0044] Through data smoothing, noise and outliers can be effectively eliminated, the accuracy, reliability and visualization of data can be improved, and the precision of data analysis can be enhanced. Data patterns and trends can be revealed more clearly, providing a reliable basis for subsequent data analysis and mining. This is conducive to accurately judging the abnormal conditions of mining trucks and making corresponding decisions and operations, reducing the occurrence of failures, improving the level of safety protection, and improving the operating efficiency of mining trucks.

[0045] Step S4, the obtained smoothed slope, road curvature and vehicle speed data are evenly divided into three intervals according to size, the boundary value of each interval is recorded, and the three intervals of the three factors are arranged and combined to obtain 27 different working conditions. The step S4 specifically includes the following steps.

[0046] Step S4.1, setting the obtained smoothed slope, road curvature and vehicle speed data as slope s, road curvature c and vehicle speed v respectively.

[0047] Step S4.2, set the boundary values ​​of slope, road curvature and vehicle speed, and divide them into three intervals respectively, set the high-end interval of slope to s1, the middle interval of slope to s2, the initial interval of slope to s3, the high-end interval of road curvature to c1, the middle interval of road curvature to c2, the initial interval of road curvature to c3, the high-end interval of vehicle speed to v1, the middle interval of vehicle speed to v2, and the initial interval of vehicle speed to v3.

[0048] Step S4.3, next, perform permutations and combinations to obtain 27 different working conditions as follows: (s1, v1, c1), (s2, v1, c1), (s3, v1, c1), (s1, v2, c1), (s2, v2, c1), (s3, v2, c1), (s1, v3, c1), (s2, v3, c1), (s3, v3, c1), (s1, v1, c2), (s2, v1, c2), (s3, v1, c2), (s1,v2,c2), (s2,v2,c2), (s3,v2,c2), (s1,v3,c2), (s2,v3,c2), (s3,v3,c2), (s1,v1,c3), (s2,v 1,c3), (s3,v1,c3), (s1,v2,c3), (s2,v2,c3), (s3,v2,c3), (s1,v3,c3), (s2,v3,c3), (s3,v3,c3). See Figure 2 , dividing the data into intervals for analysis significantly refines the granularity of the analysis compared to overall analysis. By dividing a large data set into smaller intervals, the unique characteristics and changing trends of the data in each interval can be calculated more accurately. It avoids minor errors that may be ignored due to data aggregation in the overall analysis, providing a more accurate basis for decision-making. Since each interval is independent and targeted, it can reduce the errors and interference factors that may exist in the overall analysis, thereby drawing more accurate and reliable conclusions.

[0049] Step S5, classify the motion state data collected in step S1 under the condition of normal running according to the intervals divided in step S4, and for each of the 27 different working conditions, count the data of lateral deviation and heading deviation, and use the box plot method to set two-level abnormal monitoring thresholds for the data of lateral deviation and heading deviation of each type of working condition. The two-level abnormal monitoring threshold includes a first-level threshold and a second-level threshold higher than the first-level threshold. Among them, lower than the first-level threshold represents that the mining truck movement state is normal, exceeding the first-level threshold but not reaching the second-level threshold represents a potential abnormal state, and exceeding the second-level threshold represents a significant abnormal state.

[0050] Specifically, Figure 3As shown, the two-level abnormal monitoring threshold setting based on the box plot includes the following steps: the box plot consists of a rectangular box and two extended lines (box whiskers), and the two ends of the box represent the first quartile (Q1) and the third quartile (Q3), respectively, which contain the middle 50% of the data set. The length between the first quartile (Q1) and the third quartile (Q3) is the interquartile range (IQR). The box whiskers extend to the minimum and maximum values ​​of the data (excluding outliers), and their length is usually 1.5 times the interquartile range (IQR). The outliers are marked outside the box whiskers as independent points, revealing the outliers in the data. Setting 2 times the interquartile range (IQR) as the abnormal reminder point and 3 times the interquartile range (IQR) as the abnormal warning point can effectively divide the obtained data.

[0051] Step S6, obtain the real-time data of the mining truck running by laser radar, camera, GPS system or sensor, extract the data of lateral deviation and heading deviation in the real-time data, and compare it with the two-level abnormal detection threshold, and determine the motion state level of the real-time data as the judgment result. Each real-time data of the mining truck represents the state of the mining truck at a certain moment, and each real-time data includes data of three influencing factors: vehicle speed, curvature and slope. Under the state corresponding to these three influencing factors, there are lateral deviation and heading deviation data. After the three influencing factors of a real-time data are classified according to the boundary value, the lateral deviation and heading deviation data corresponding to the influencing factors of the real-time data will also be passively classified.

[0052] The following references Figure 2 The way of classifying the real-time data of the mining truck by interval is exemplified. In one example: assuming that the vehicle speed v ranges from 0 to 34, the curvature c ranges from 0 to 0.26, and the slope s ranges from -1 to 8 (this example is only used to facilitate the understanding of this embodiment, and the data does not serve as any limitation to the present invention):

[0053] c = [0, 0.09, 0.18, 0.26];

[0054] v = [0, 11, 23, 34];

[0055] s=[-1,-0.4,0.2,0.8].

[0056] Set the boundary values ​​and the corresponding sub-intervals, c1 = [0, 0.09), c2 = [0.09, 0.18), c3 = [0.18, 0.26]; v1 = [0, 11), v2 = [11, 23), v3 = [23, 34]; s1 = [-1, -0.4), s2 = [-0.4, 0.2), s3 = [0.2, 0.8]. Then perform permutations and combinations, and divide the obtained sub-intervals into 27 types of working conditions, such as Figure 2As shown schematically, each grid is a working condition.

[0057] Example of real-time data list of mining card 1

[0058]

[0059]

[0060] Taking the first data in the list of the above example as an example, the vehicle speed is 19, in the interval of v2, the slope (slope) is 0.3, in the interval of s3, the curvature is 0.007, and in the interval of c1, it is determined that the data is in the 20th category. It is necessary to compare the lateral deviation -0.056569 and the heading deviation -0.19644 of the data with the previously calculated thresholds of the lateral deviation and heading deviation of the 20th category.

[0061] Specifically, step S6 includes the following steps.

[0062] Step S6.1, obtaining real-time data of the mining truck running through laser radar, camera, GPS system or sensor.

[0063] Step S6.2, compare the real-time data with the 27 different working conditions obtained in step S4 to determine the working condition category. The real-time data is compared with the 27 different working conditions obtained in step S4, specifically including comparing the values ​​of the slope, road curvature and vehicle speed in the key information of the real-time data with the boundary values ​​of the above 27 different working conditions to determine the working condition category.

[0064] Step S6.3, according to the category of the working condition, select the two-level abnormal monitoring threshold of the corresponding category from the result obtained in step S5.

[0065] Step S6.4, extract the data of lateral deviation and heading deviation from the key information of real-time data, and compare them with the two-level abnormal monitoring thresholds of the corresponding categories to obtain the motion state level as the judgment result, wherein, if the data of lateral deviation and heading deviation are lower than the first-level threshold, it indicates that the mining truck is in a normal state; if the data of lateral deviation and heading deviation exceed the first-level threshold but do not reach the second-level threshold, it indicates that the mining truck is in a potential abnormal state; if the data of lateral deviation and heading deviation exceed the second-level threshold, it indicates that the mining truck is in a significant abnormal state.

[0066] Step S7, take corresponding countermeasures according to the judgment result obtained in step S6. Among them, if the judgment result is a normal state, no processing is required; if it is a potential abnormal state, an abnormal reminder is issued; if it is a significant abnormal state, an abnormal alarm is issued, the fault is reported and the vehicle is stopped. Among them, in a potential abnormal state, it is not necessary to take measures immediately, and monitoring is maintained. If the abnormal reminder continues to appear, it will be converted to a significant abnormal state and corresponding processing will be taken. Among them, in a significant abnormal state, it can also be set to take measures such as issuing an abnormal alarm and dispatching the mining truck to the maintenance area for inspection.

[0067] The following references Figures 4 to 7 , an exemplary embodiment of an abnormal motion state monitoring and early warning method for an autonomous driving mining truck applying an embodiment of the present invention is described in detail.

[0068] For ease of understanding, this embodiment uses the data of a mining truck in a mining area in a normal running state as the basic data, and the subsequent real-time running data of the mining truck is used as the experimental data. Figure 1 As shown, in this exemplary embodiment, abnormal status monitoring and early warning are performed on the autonomous driving mining truck, including performing the following steps.

[0069] Step S1, collect the motion state data of the mining truck under normal running conditions through data collection methods such as laser radar, camera, GPS system, sensor, etc. The collected motion state data of the mining truck under normal running conditions are all saved in the log file, and the motion state data are extracted from the log file of the normal running condition as the basic data of the mining truck. In order to ensure the accuracy and objectivity of data analysis, a large amount of normal running data under different weather conditions, different tasks, etc. can be collected.

[0070] Step S2, extracting key information from the motion state data of a normal sports car obtained in step S1, including: factors affecting the motion state: slope, road curvature, vehicle speed, etc.; manifestations of the motion state: lateral deviation, heading deviation, etc.

[0071] Step S3, sliding average processing is performed on the factors affecting the motion state in the extracted key information to obtain smoothed key information. In order to ensure the accuracy of the obtained data and prevent the interference of erroneous data, the data is processed by sliding average, and the sliding window size is set to 20, that is, the data at a certain moment and the following 20 data are grouped as a group to obtain the average value as the data value at that moment. The smoothed key information includes the smoothed slope, road curvature, and vehicle speed data.

[0072] Step S4, the obtained smoothed slope, road curvature, and vehicle speed data are evenly divided into three intervals according to their sizes, and the boundary value of each interval is recorded, so as to obtain 27 different working conditions in total.

[0073] Step S5, classify the motion state data of the normal running car collected in step S1 according to the intervals divided in step S4, and for each category, count the lateral deviation and heading deviation data, and use the box plot method to set two-level abnormal monitoring thresholds for the lateral deviation and heading deviation data respectively. The two-level abnormal monitoring threshold includes a first-level threshold and a second-level threshold higher than the first-level threshold. Among them, lower than the first-level threshold represents that the mining truck has a normal motion state, exceeding the first-level threshold but not reaching the second-level threshold represents a potential abnormal state, and exceeding the second-level threshold represents a significant abnormal state.

[0074] Step S6, acquiring real-time data, and determining the motion state level of the real-time data as a judgment result, that is, whether it is abnormal and the abnormal level, based on the relationship between the lateral deviation and heading deviation in the real-time data and the threshold.

[0075] Specifically, Figure 4 As shown, real-time abnormal monitoring and early warning of the motion state of the autonomous driving mining truck are performed, including the following steps: obtaining key information of the real-time data, including: slope, road curvature, vehicle speed, lateral deviation, heading deviation, etc., and determining which type of working condition the autonomous driving mining truck is currently in according to the 27 different working conditions obtained in step S4; comparing the lateral deviation and heading deviation in the real-time data with the two-level abnormal monitoring thresholds of the type obtained in step S5, and determining the motion state level of the real-time data, that is, determining whether it is abnormal and the abnormal level.

[0076] See the accompanying drawings, wherein Figure 5 FIG. 1 is a schematic diagram showing a normal state of real-time data in this embodiment. Figure 6 Schematic diagram of an abnormal reminder of real-time data in this embodiment. Figure 7 This is a schematic diagram of an abnormal alarm in real-time data in this embodiment. Figures 5 to 7 Each schematic diagram includes four parts from top to bottom, namely, lateral deviation, abnormal lateral deviation, heading deviation, and abnormal heading deviation. Figures 5 to 7 The top part of each schematic diagram is the lateral deviation part, in which the five lines from top to bottom are the abnormal alarm upper limit, abnormal reminder upper limit, real-time lateral deviation data, abnormal reminder lower limit, and abnormal alarm lower limit. Figure 5 As shown in Figure 1, the real-time lateral deviation data are all within the abnormal warning threshold, and this data group is normal data. Figure 6 As shown in the figure, the lateral deviation data line is between the abnormal reminder data line and the abnormal alarm data line, indicating that the data is abnormal data and an abnormal reminder is required. Figure 7 As shown, the lateral deviation data line is outside the abnormal alarm data line, indicating that the data is particularly abnormal data and an abnormal alarm is required.

[0077] exist Figures 5 to 7The second part of each schematic diagram in FIG. 1 is the abnormal situation part of the lateral deviation. Figure 5 As shown in the figure, the data status judgment of the lateral deviation abnormality diagram is 0, which is the data in normal state. Figure 6 As shown in , if the state judgment of the data displayed in the lateral deviation abnormal situation diagram is 1, it shows that the data is outside the abnormal reminder threshold and within the abnormal warning threshold. Figure 7 As shown, if the lateral deviation abnormal situation diagram shows that the state judgment of the data is 2, it means that the data is outside the abnormal alarm threshold.

[0078] Step S7, according to the judgment result of step S6, if the data is normal, output normal as the judgment result, otherwise, judge the data as abnormal, and determine the abnormal level according to the degree of abnormality, and then dispatch the mining truck to the maintenance area for inspection.

[0079] Specifically, Figure 4 As shown, the response measures taken in this embodiment are as follows: if the judgment result is a normal state, no processing is required; in a potential abnormal state, an abnormal reminder is given, and no immediate measures need to be taken. If the abnormal reminder continues to appear and appears for more than eight minutes within ten minutes, the fault is reported and the vehicle is stopped; in a significant abnormal state, an abnormal alarm is issued, the fault is immediately reported and the vehicle is stopped, or the mining truck is immediately dispatched to the maintenance area for further testing to find out the cause of the abnormality.

[0080] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.

[0081] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0082] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for monitoring and warning abnormal motion status of an autonomous driving mining truck, characterized in that: The following steps are involved: Step S1, collecting the motion state data of the mining truck in a normal running state through a laser radar, a camera, a GPS system or a sensor; Step S2, extracting key information from the motion state data of a normal sports car, the key information including factors affecting the motion state and the performance of the motion state, the factors affecting the motion state including slope, road curvature and vehicle speed, the performance of the motion state including lateral deviation and heading deviation; Step S3, performing sliding average processing on the factors affecting the motion state in the extracted key information to obtain smoothed data of the slope, road curvature and vehicle speed as the smoothed key information; Step S4, dividing the obtained smoothed slope, road curvature and vehicle speed data into three intervals according to their size, recording the boundary value of each interval, and arranging and combining the three intervals of the three factors to obtain 27 different working conditions; Step S5, classifying the motion state data of the normal sports car collected in step S1 according to the intervals divided in step S4, and respectively counting the data of lateral deviation and heading deviation for each of the 27 different working conditions, and using the box plot method to set two levels of abnormal monitoring thresholds for the data of lateral deviation and heading deviation of each working condition; Step S6, obtaining real-time data of the mining truck running by laser radar, camera, GPS system or sensor, extracting the data of lateral deviation and heading deviation in the real-time data, and comparing them with the two-level abnormality detection threshold, and determining the motion state level of the real-time data as the judgment result; Step S7, take corresponding countermeasures according to the judgment result obtained in step S6. If the judgment result is a normal state, no processing is required; if it is a potential abnormal state, an abnormal reminder is issued; if it is a significant abnormal state, an abnormal alarm is issued, the fault is reported and the vehicle is stopped.

2. The method for abnormal motion state monitoring and early warning of an autonomous driving mining truck according to claim 1 is characterized in that: The step S4 specifically comprises the following steps: Step S4.1, setting the obtained smoothed slope, road curvature and vehicle speed data as slope s, road curvature c and vehicle speed v respectively; Step S4.2, setting the data ranges of the slope, road curvature and vehicle speed to be evenly divided into three continuous sub-intervals, setting the high-end interval of the slope to s1, the middle interval of the slope to s2, the initial interval of the slope to s3, the high-end interval of the road curvature to c1, the middle interval of the road curvature to c2, the initial interval of the road curvature to c3, the high-end interval of the vehicle speed to v1, the middle interval of the vehicle speed to v2, and the initial interval of the vehicle speed to v3; Step S4.3, after dividing the intervals, the sub-intervals of slope, road curvature and vehicle speed are arranged and combined to obtain 27 different working conditions: (s1, v1, c1), (s2, v1, c1), (s3, v1, c1), (s1, v2, c1), (s2, v2, c1), (s3, v2, c1), (s1, v3, c1), (s2, v3, c1), (s3, v3, c1), (s1, v1, c2), (s2, v1, c2), (s3 ,v1,c2), (s1,v2,c2), (s2,v2,c2), (s3,v2,c2), (s1,v3,c2), (s2,v3,c2), (s3,v3,c2), (s1,v1,c3), ( s2,v1,c3), (s3,v1,c3), (s1,v2,c3), (s2,v2,c3), (s3,v2,c3), (s1,v3,c3), (s2,v3,c3), (s3,v3,c3).

3. The method for abnormal motion state monitoring and early warning of an autonomous driving mining truck according to claim 2 is characterized in that: In step S5: The two-level abnormality monitoring threshold comprises a first-level threshold and a second-level threshold higher than the first-level threshold; Among them, being lower than the first-level threshold value indicates that the mining truck movement state is normal, exceeding the first-level threshold value but not reaching the second-level threshold value indicates a potential abnormal state, and exceeding the second-level threshold value indicates a significant abnormal state.

4. The method for abnormal motion state monitoring and early warning of an autonomous driving mining truck according to claim 3 is characterized in that: The step S6 specifically comprises the following steps: Step S6.1, obtaining real-time data of the mining truck running by laser radar, camera, GPS system or sensor; Step S6.2, comparing the real-time data with the 27 different types of working conditions obtained in step S4 to determine the type of working condition; Step S6.3, according to the category of the working condition, select the two-level abnormal monitoring threshold of the corresponding category from the result obtained in step S5; Step S6.4, extract the data of lateral deviation and heading deviation in the real-time data, and compare them with the two-level abnormal monitoring thresholds of the corresponding category to obtain the motion state level as the judgment result, wherein, if the data of lateral deviation and heading deviation are lower than the first-level threshold, it indicates that the mining truck is in a normal state; if the data of lateral deviation and heading deviation exceed the first-level threshold but do not reach the second-level threshold, it indicates that the mining truck is in a potential abnormal state; if the data of lateral deviation and heading deviation exceed the second-level threshold, it indicates that the mining truck is in a significant abnormal state.

Images