A safety access detection method and system for an unmanned mine dump truck

Abstract

The application discloses a kind of unmanned mine dump truck's safety access detection method and system, effectively solve the problem that a large number of unmanned mine dump trucks are difficult to put into actual operation due to lack of unified inspection and detection standard for unmanned mine dump trucks.The method comprises: configuring a plurality of detection dimensions for the to-be-tested unmanned mine dump truck in advance, and configuring a corresponding detection environment for each detection dimension, calling a target detection method to detect the to-be-tested unmanned mine dump truck to obtain detection data; the detection data includes vehicle body data and simulation data; the simulation data and the vehicle body data of the plurality of detection dimensions are input into the access evaluation model and the fault prediction model respectively to obtain evaluation data and fault data; the access detection result of the to-be-tested unmanned mine dump truck is obtained by fusing the fault data and the evaluation data of the plurality of detection dimensions.

Description

Technical Field

[0001] This invention relates to the field of intelligent mining and unmanned vehicle technology, and in particular to a safety access detection method and system for unmanned mining dump trucks. Background Technology

[0002] With the development of unmanned driving technology in open-pit mines, a large number of unmanned mining dump trucks have been put into actual operation. The main reason for this is the current lack of unified pre-entry inspection and testing standards in the industry, and the shortcomings of the inspection and testing standards used in some mining areas:

[0003] 1. Single detection dimension: Most mining areas only test vehicle hardware or basic performance, ignoring the safety verification of the integrated hardware and software of the autonomous driving module.

[0004] 2. Non-standard methods: Different mining areas have different self-built testing procedures, resulting in some vehicles entering operation with potential hazards.

[0005] 3. Insufficient coverage of extreme working conditions: In extreme environments such as high altitude, high cold, and high dust, insufficient detection and verification can easily lead to sensor failure or control abnormalities.

[0006] 4. Lack of intelligent monitoring: Existing detection methods mostly rely on manual and single-point testing, lacking data-driven multi-dimensional monitoring and prediction capabilities.

[0007] Therefore, there is an urgent need to establish a modular and intelligent safety access inspection and testing system for unmanned mining dump trucks to achieve comprehensive, standardized, and quantifiable testing and evaluation of vehicles before they are put into use. Summary of the Invention

[0008] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a safety access inspection method and system for unmanned mining dump trucks, which effectively solves the problem that the lack of unified inspection and testing standards for unmanned mining dump trucks makes it difficult to put a large number of unmanned mining dump trucks into actual operation.

[0009] The technical solution provided in this application is: a safety access detection method for an unmanned mining dump truck, the method comprising:

[0010] Multiple detection dimensions for the unmanned mining dump truck under test are pre-configured, and a corresponding detection environment is configured for each detection dimension; different detection dimensions correspond to different detection methods; the detection environment includes typical real-world environments and simulated environments;

[0011] The unmanned mining dump truck under test is controlled to be in a target detection environment, and the corresponding target detection method is called to detect the unmanned mining dump truck under test to obtain the corresponding detection data; the detection data includes vehicle body data and simulation data;

[0012] Simulated data from multiple detection dimensions and vehicle body data are input into pre-trained admission evaluation models and fault prediction models, respectively, to process the detection data from the multiple detection dimensions and vehicle body data to obtain evaluation data and fault data.

[0013] By integrating fault data and evaluation data from multiple detection dimensions, the access detection results of the unmanned mining dump truck under test are obtained, and the safety access detection of the unmanned mining dump truck is completed based on the access detection results.

[0014] Furthermore, the pre-configured multiple detection dimensions for the unmanned mining dump truck under test include:

[0015] Based on the working environment and functional characteristics of the unmanned mining dump truck under test, multiple detection dimensions are set for the unmanned mining dump truck.

[0016] Extract the detection elements for each detection dimension, set the corresponding detection method based on the detection elements, and bind the detection method to the detection dimension.

[0017] Furthermore, the detection dimensions include at least the functional safety detection dimensions;

[0018] The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes:

[0019] Multiple core execution modules of the unmanned mining dump truck under test were identified, and the detection parameters of each core execution module were determined.

[0020] The detection method is controlled to test each detection parameter, and detection data for the functional safety detection dimension is collected through multiple acquisition methods.

[0021] Furthermore, the detection dimensions include at least the environmental adaptability detection dimension;

[0022] The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes:

[0023] The detection dimensions include at least redundancy and emergency detection dimensions;

[0024] The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes:

[0025] Based on virtual technology, a failure scenario was constructed for the unmanned mining dump truck under test in terms of braking, controller and manual control interface.

[0026] The unmanned mining dump truck under test is controlled to operate at full load in failure scenarios involving braking, controller, and manual intervention interfaces to obtain corresponding detection data. Furthermore, the detection dimensions include at least the unmanned driving dimension.

[0027] The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes:

[0028] The driving module and communication module that affect the unmanned driving of the unmanned mining dump truck under test were identified.

[0029] Test and collect detection data for the driving module and communication module in typical real-world and simulated scenarios.

[0030] Furthermore, the access detection result of the unmanned mining dump truck under test, obtained by fusing fault data and evaluation data from multiple detection dimensions, includes:

[0031] Determine whether both the evaluation data and the fault data meet the preset admission criteria;

[0032] If so, the obtained access test result is qualified, and an access test report is generated based on the access test result.

[0033] Furthermore, the detection data from the multiple detection dimensions and the vehicle body data are processed separately to obtain evaluation data and fault data, including:

[0034] Trend prediction is performed based on the aforementioned detection data and historical detection data to predict intermediate data;

[0035] Determine whether the intermediate data meets the preset fault conditions, so as to obtain the fault data based on the intermediate data.

[0036] This application provides another solution: a safety access detection system for an unmanned mining dump truck, the module comprising:

[0037] The configuration module is used to pre-configure multiple detection dimensions for the unmanned mining dump truck under test, and to configure a corresponding detection environment for each detection dimension; different detection dimensions correspond to different detection methods; the detection environment includes typical real-world environments and simulated environments;

[0038] The calling module is used to control the unmanned mining dump truck under test to be in a target detection environment, and to call the corresponding target detection method to detect the unmanned mining dump truck under test and obtain the corresponding detection data; the detection data includes vehicle body data and simulation data;

[0039] The processing module is used to input simulated data from multiple detection dimensions and vehicle body data into the pre-trained admission evaluation model and fault prediction model, respectively, so as to process the detection data from the multiple detection dimensions and vehicle body data to obtain evaluation data and fault data.

[0040] The fusion module is used to fuse fault data and evaluation data from multiple detection dimensions to obtain the access detection results of the unmanned mining dump truck under test, so as to complete the safety access detection of the unmanned mining dump truck based on the access detection results.

[0041] This application also provides a solution: an electronic device, comprising: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory via the bus, and when the machine-readable instructions are executed by the processor, the steps of any one of the safety access detection methods for an unmanned mining dump truck are performed.

[0042] This application also provides another solution: a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of any one of the safety access detection methods for an unmanned mining dump truck.

[0043] This application provides a safety access detection method for an unmanned mining dump truck. The method first pre-configures multiple detection dimensions for the unmanned mining dump truck under test and configures a corresponding detection environment for each detection dimension; different detection dimensions correspond to different detection methods; the detection environment includes typical real-world environments and simulated environments; secondly, it controls the unmanned mining dump truck under test to be in the target detection environment and calls the corresponding target detection method to detect the unmanned mining dump truck under test to obtain corresponding detection data; the detection data includes vehicle body data and simulated data; then, it inputs the simulated data and vehicle body data of multiple detection dimensions into a pre-trained access evaluation model and fault prediction model respectively to process the detection data and vehicle body data of the multiple detection dimensions to obtain evaluation data and fault data; finally, it merges the fault data and evaluation data of multiple detection dimensions to obtain the access detection result of the unmanned mining dump truck under test, thereby completing the safety access detection of the unmanned mining dump truck based on the access detection result. Based on the above methods, a scientific, effective, and comprehensive testing and inspection standard covering various extreme working conditions for unmanned mining dump trucks has been established. This also enables a comprehensive evaluation of unmanned mining dump trucks before they enter the mine, ensuring that they meet the access requirements in terms of performance, safety, reliability, and environmental adaptability. The method provided in this application can be applied to various mining areas, effectively solving the problem that the lack of a unified testing and inspection standard for unmanned mining dump trucks makes it difficult to put a large number of them into actual operation. Attached Figure Description

[0044] Figure 1 This is a flowchart illustrating a safety access detection method for an unmanned mining dump truck provided in an embodiment of this application.

[0045] Figure 2 This is a schematic diagram of the process for obtaining the detection dimension provided in an embodiment of this application.

[0046] Figure 3 This is a structural block diagram of a safety access detection system for an unmanned mining dump truck provided in an embodiment of this application.

[0047] Figure 4 This is a structural block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0048] For the purposes of this invention, the foregoing and other technical contents, features and effects are described in conjunction with the appendix below. Figure 1-4 The detailed description of the embodiments will make this clear. All structural details mentioned in the following embodiments are based on the accompanying drawings.

[0049] Currently, a large number of driverless mining dump trucks have been put into actual operation. The main reason for this is that the industry lacks a unified pre-entry inspection and testing standard, and the inspection and testing standards used in some mining areas have the following shortcomings: single testing dimensions, non-standard methods, lack of coverage of extreme working conditions, and lack of intelligent monitoring.

[0050] Based on this, the present application provides a safety access detection method and system for unmanned mining dump trucks, which will be described below through embodiments and accompanying drawings.

[0051] Example 1

[0052] To facilitate understanding of this embodiment, a detailed description of the safety access detection method for an unmanned mining dump truck disclosed in this application embodiment will be provided first. For example... Figure 1 The present application provides a safety access detection method for an unmanned mining dump truck, the method comprising:

[0053] S101. Pre-configure multiple detection dimensions for the unmanned mining dump truck under test, and configure a corresponding detection environment for each detection dimension; different detection dimensions correspond to different detection methods; the detection environment includes typical real-world environment and simulated environment;

[0054] S102. Control the unmanned mining dump truck under test to be placed in the target detection environment, and call the corresponding target detection method to detect the unmanned mining dump truck under test to obtain the corresponding detection data; the detection data includes vehicle body data and simulation data;

[0055] S103. Input the simulation data of multiple detection dimensions and the vehicle body data into the pre-trained admission evaluation model and fault prediction model respectively, so as to process the detection data of the multiple detection dimensions and the vehicle body data respectively to obtain evaluation data and fault data.

[0056] S104. By integrating fault data and evaluation data from multiple detection dimensions, the access detection result of the unmanned mining dump truck under test is obtained, and the safety access detection of the unmanned mining dump truck is completed based on the access detection result.

[0057] The method described in this application is based on hardware equipment, which includes a detection terminal layer, a data acquisition layer, a cloud evaluation layer, and a result output layer. The detection terminal layer is installed on the unmanned mining dump truck and is used to connect to the interface of the unmanned mining dump truck. Specifically, it consists of sensors, test modules, and hardware interfaces, including a lidar detection system, a braking performance tester, a battery / power module diagnostic module, and a communication link detection module.

[0058] In step S101, based on the actual testing requirements of unmanned mining dump trucks, this application pre-configures multiple testing dimensions for the unmanned mining dump truck under test. These testing dimensions include functional safety testing, environmental adaptability testing, unmanned driving testing, and redundancy and emergency testing. A corresponding testing environment is configured for each testing dimension. Different testing dimensions correspond to different testing methods, thereby achieving targeted testing for each dimension and improving the accuracy and effectiveness of testing unmanned mining dump trucks. The testing environment includes typical real-world environments and simulated environments. The simulated environment includes general simulated environments and extreme simulated environments. The general simulated environment corresponds to the typical real-world environment, while the extreme simulated environment includes… This application covers various operating environments for unmanned mining dump trucks, including low-temperature, high-altitude, and high-dust scenarios, ensuring comprehensiveness. Different detection dimensions require the acquisition of corresponding detection data from the unmanned mining dump truck, which is achieved by a data acquisition layer. This layer acquires data on various parameters of the unmanned mining dump truck during real-time operation, such as braking distance, steering response, energy consumption, and sensor accuracy, through a detection terminal layer and a data bus. The typical real-world environment involves a cyclical driving process in a typical mining area, such as starting, accelerating, climbing, descending, braking, and obstacle avoidance, for real-vehicle testing. The simulated environment is based on various extreme environments simulated using multiple digital technologies to experiment on the components of the unmanned mining dump truck.

[0059] In the specific implementation of step S101, one embodiment is as follows: Figure 2 As shown, the pre-configured multiple detection dimensions for the unmanned mining dump truck under test include:

[0060] S1011. Based on the working environment and functional characteristics of the unmanned mining dump truck under test, multiple detection dimensions are set for the unmanned mining dump truck.

[0061] S1012. Extract the detection elements of each detection dimension, set the corresponding detection method based on the detection elements, and bind the detection method to the detection dimension.

[0062] In steps S1011-S1012, this application sets up multiple detection dimensions for the unmanned mining dump truck under test based on its working environment and functional characteristics. These dimensions include functional safety detection, environmental adaptability detection, unmanned driving, and redundancy and emergency detection. Each detection dimension is set based on the actual detection requirements of the unmanned mining dump truck under test. Detection elements are extracted for each detection dimension, with different elements corresponding to different detection dimensions. For example, in the functional safety dimension, the braking module, steering module, and drive module of the unmanned mining dump truck under test need to be tested to ensure basic driving safety. In the redundancy and emergency detection dimension, the braking redundancy module, emergency stop system, and manual takeover interface of the unmanned mining dump truck under test need to be tested. Corresponding detection methods are set based on the detection elements, and the detection methods are bound to the detection dimensions to improve the accuracy and effectiveness of the detection results under each detection dimension.

[0063] In step S102, this application controls the unmanned mining dump truck under test to be in a target detection environment. That is, during the detection process, the unmanned mining dump truck under test is controlled to be in a typical real environment and a simulated environment, and the unmanned mining dump truck under test is in a fully loaded state. The corresponding target detection method is called to detect the unmanned mining dump truck under test and obtain the corresponding detection data. That is, data of the corresponding detection dimension is collected based on the data acquisition layer. The detection data includes vehicle body data and simulation data. That is, data generated in both the typical real environment and the simulated environment are collected to ensure the accuracy of the obtained data. In practice, detection priorities can be set in the above detection dimensions. The detection priorities are sorted according to safety, function, status and environment to ensure that the core risks of the unmanned mining dump truck under test are covered first. Detection data of different detection dimensions can also be set to verify each other. For example, the root cause of the failure can be judged by combining tire wear and hydraulic oil status when braking performance deteriorates.

[0064] In the specific implementation of step S102, one embodiment is as follows: the detection dimension includes at least the functional safety detection dimension;

[0065] The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes:

[0066] A1. Identify multiple core execution modules of the unmanned mining dump truck under test, and determine the detection parameters for each core execution module;

[0067] A2. Control the detection method to test each detection parameter, and collect detection data of functional safety detection dimensions through multiple acquisition methods.

[0068] In steps A1-A2, this application pre-determines multiple core execution modules of the unmanned mining dump truck under test. These core execution modules are safety-related modules during the operation of the unmanned mining dump truck under test, such as braking, steering, and drive modules. The application also determines the detection parameters for each core execution module, such as braking sensitivity and braking distance parameters for the braking module, and steering accuracy and power assist status for the steering module. The application controls the detection method to test each detection parameter and collects functional safety testing data through various acquisition methods, such as using multiple sensors inherent to the vehicle and the vehicle's own sensors. The vehicle body data is transmitted from various data buses. In this dimension, the unmanned mining dump truck under test needs to be tested in both typical real-world scenarios and simulated scenarios. The extreme simulation scenarios in the simulation scenarios can be low temperature, high altitude, and high dust scenarios. When the mining dump truck under test is fully loaded, the braking module load is 3-5 times that of the unloaded load, and the sensitivity and braking distance will change significantly (e.g., the braking distance of the fully loaded truck may be more than twice that of the unloaded truck). The fully loaded state must be the core testing scenario. When the mining area is heavily loaded and going downhill, the braking module is prone to decreased sensitivity (thermal decay) due to continuous friction and heating. It is necessary to supplement "continuous braking test" (e.g., continuous braking 5 times, with an interval of 30 seconds between each time) to ensure that the safety requirements are still met after thermal decay.

[0069] In the specific implementation of step S102, another embodiment exists in which the detection dimensions include at least redundant and emergency detection dimensions.

[0070] The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes:

[0071] B1. Based on virtual technology, a failure scenario was constructed for the unmanned mining dump truck under test in terms of braking, controller and manual takeover interface.

[0072] B2. Control the unmanned mining dump truck under test to run at full load in failure scenarios of braking, controller and manual takeover interface, so as to obtain the corresponding test data.

[0073] In steps B1-B2, within this dimension, this application constructs failure scenarios for the unmanned mining dump truck under test based on virtual technology, focusing on its braking, controller, and manual intervention interfaces. This requires considering the specific working conditions of heavy loads, steep slopes, and dust in mines. The unmanned mining dump truck under test is operated at full load under these failure scenarios to obtain corresponding test data. The braking redundancy system, emergency stop system, and manual intervention interface of the unmanned mining dump truck under test are the last line of defense for safety. Verification must move beyond normal operating conditions and focus on the most dangerous main system failures, extreme environments, and automation anomalies in mining operations. Through practical simulation, data quantification, and repeated verification, it is ensured that each module can reliably function at critical moments, truly guaranteeing the safety of personnel and equipment in the mining area. The braking redundancy module must ensure that, after the main brake fails, the backup brake can bring the fully loaded vehicle to a stop within a specified distance, with a stable braking process and no safety risks. The emergency stop module, once triggered, can stop within ≤0.3... With a response time of less than 1 second, the system can forcibly cut off power and apply maximum braking force, stopping fully loaded vehicles within a safe distance without rolling away after stopping. The manual takeover interface must respond within ≤0.1 seconds after manual operation triggering, fully covering automated control. The takeover process is smooth, without delay or control conflicts.

[0074] In the specific implementation of step S102, there is another embodiment in which the detection dimension includes at least the autonomous driving dimension;

[0075] The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes:

[0076] C1. Identify the driving module and communication module that affect the unmanned driving of the unmanned mining dump truck under test;

[0077] C2. Test and collect test data for the driving module and communication module in typical real-world and simulated scenarios respectively.

[0078] In steps C1-C2, this application identifies the driving module and communication module that affect the unmanned driving of the unmanned mining dump truck under test. The driving module includes a perception module, positioning, and decision control for functional and accuracy testing. The perception module includes LiDAR, cameras, millimeter-wave radar, positioning module includes GNSS+IMU, and decision control module includes path planning and obstacle avoidance. The communication module includes V2X communication and vehicle network to prevent signal loss and hacker attacks. The testing data of the driving module and communication module in typical real-world scenarios and simulated scenarios are tested and collected to ensure the comprehensiveness of the test in the unmanned driving dimension.

[0079] In step S103, after obtaining the simulated data and vehicle body data corresponding to the multiple detection dimensions, this application inputs the simulated data and vehicle body data of the multiple detection dimensions into the pre-trained admission evaluation model and fault prediction model, respectively. The simulated data is data detected in a simulated scenario, while the vehicle body data is data detected in a typical real-world scenario. Before inputting the data into the admission evaluation model and fault prediction model, the data range to be processed needs to be determined, and then interference is eliminated through standardization cleaning to adapt to the special characteristics of mining scenarios, such as dust and vibration causing high data noise. All data is synchronized according to timestamps. Brake pedal trigger time, vehicle speed change, and hydraulic pressure are aligned to the millisecond level to ensure that multi-dimensional data can be correlated and analyzed, such as whether the braking response delay is synchronized with the hydraulic pressure drop. The admission assessment model and fault prediction model are both built and trained in the cloud to process the detection data and vehicle body data of the multiple detection dimensions to obtain assessment data and fault data, respectively. That is, the detection data is compared with the benchmark data, such as industry standards / vehicle calibration values / historical health values, to evaluate the assessment data. The assessment data is represented by a health score. The fault data identifies the magnitude and trend of the measured value deviating from the benchmark value to locate the fault type and risk level.

[0080] In a specific implementation of step S103, one embodiment involves processing the detection data from the multiple detection dimensions and the vehicle body data to obtain evaluation data and fault data, including:

[0081] S1031. Based on the detection data and historical detection data, perform trend prediction to predict intermediate data;

[0082] S1032. Determine whether the intermediate data meets the preset fault conditions, so as to obtain the fault data based on the intermediate data.

[0083] In steps S1031-S1032, the fault prediction model establishes a normal trend baseline based on historical detection data and fits a detection data curve based on the detection data for trend prediction. An early anomaly is identified by comparing the normal trend baseline and the detection data curve to obtain intermediate data. It is then determined whether the intermediate data meets a preset fault condition. The fault condition can be that the difference between the data on the normal trend baseline and the detection data curve is greater than a preset threshold, and that the difference remains greater than the preset threshold at the next moment. The fault condition can be set according to actual conditions. If the fault condition is not met, monitoring continues. If the fault condition is met, fault data is obtained. Based on this method, the accuracy of the identified fault data is ensured. All intermediate data and fault data must be bound to operating conditions, such as full load + 15° steep slope, to avoid misjudging normal data under abnormal operating conditions as faults, such as increased frame stress due to overload. The preset threshold is dynamically updated based on vehicle mileage and maintenance records, such as a new car brake pedal travel threshold of 15mm after 30,000km of driving. The thickness was later adjusted to 17mm to match the aging characteristics of the unmanned mining dump truck under test.

[0084] In step S104, after obtaining the fault data and evaluation data corresponding to a single detection dimension, the fault data and evaluation data of multiple detection dimensions are fused to obtain the access detection result of the unmanned mining dump truck under test. This is to avoid the data of a single monitoring dimension being easily affected by working conditions. For example, a long braking distance may be a road problem rather than a system fault. The fault data of multiple detection dimensions are verified to obtain the final fault result. The fusion of evaluation data involves assigning corresponding weights to multiple detection dimensions, and then weighting and fusing the weights corresponding to the detection dimensions with the evaluation data to obtain the evaluation result of the unmanned mining dump truck under test. Based on the fault result and the evaluation result, the access detection result of the unmanned mining dump truck under test is obtained. The access detection result includes qualified and unqualified. If any one of the evaluation result and the fault result is abnormal, the access detection result is unqualified. The safety access detection of the unmanned mining dump truck is completed based on the access detection result.

[0085] In a specific implementation of step S104, one embodiment is as follows: the access detection result of the unmanned mining dump truck under test is obtained by fusing fault data and evaluation data from multiple detection dimensions, including:

[0086] S1041. Determine whether the evaluation data and the fault data both meet the preset admission conditions;

[0087] S1042. If so, the obtained access test result is qualified, and an access test report is generated based on the access test result.

[0088] In steps S1041-S1042, this application determines whether the evaluation data and the fault data both meet the preset admission conditions. The admission conditions are that both the evaluation result and the fault result are normal, and the admission detection result is qualified. If so, the obtained admission detection result is qualified. If either the evaluation result or the fault result is abnormal, the admission detection result is unqualified. An admission detection report is generated based on the admission detection result to form a written and interpretable report.

[0089] This application proposes for the first time a detection system covering hardware, software, environment, network, and redundancy dimensions. Through AI prediction and digital twin simulation, it improves detection efficiency and accuracy, ensuring that all unmanned mining dump trucks entering the mining area meet strict safety access standards. The method provided in this application can serve as an industry-standardized detection platform with broad application value. It differs from traditional single performance testing and forms a new multi-dimensional and intelligent detection and access model.

[0090] In practice, this was verified on a 140-ton unmanned mining dump truck:

[0091] Functional safety dimension: Test braking distance, braking distance at full load and 30km / h ≤25m;

[0092] In terms of autonomous driving: LiDAR point cloud recognition rate ≥95%, camera target detection error ≤0.3 m; battery capacity retention rate ≥80% in a -30℃ environmental chamber test; V2X link packet loss rate ≤1% within a 2 km range in the mining area.

[0093] Redundancy and emergency detection dimensions: Emergency braking system response time ≤ 100 ms;

[0094] After the test results are evaluated by a model in the cloud, an "access qualification report" is generated, and the vehicle is approved to enter the mining area for operation.

[0095] Example 2

[0096] This application also provides a safety access detection system for unmanned mining dump trucks, such as... Figure 3 The diagram shows a block diagram of a safety access detection system for an unmanned mining dump truck. The functions implemented by this system correspond to the steps described above in executing a safety access detection method for an unmanned mining dump truck on a terminal device. This system can be understood as a server component including a processor. The module of the safety access detection system for an unmanned mining dump truck described in this application includes:

[0097] The configuration module 301 is used to pre-configure multiple detection dimensions for the unmanned mining dump truck under test, and configure a corresponding detection environment for each detection dimension; different detection dimensions correspond to different detection methods; the detection environment includes typical real environment and simulated environment;

[0098] The calling module 302 is used to control the unmanned mining dump truck under test to be in a target detection environment, and to call the corresponding target detection method to detect the unmanned mining dump truck under test to obtain the corresponding detection data; the detection data includes vehicle body data and simulation data;

[0099] The processing module 304 is used to input simulated data from multiple detection dimensions and vehicle body data into the pre-trained admission evaluation model and fault prediction model, respectively, so as to process the detection data from the multiple detection dimensions and vehicle body data to obtain evaluation data and fault data.

[0100] The fusion module 304 is used to fuse fault data and evaluation data from multiple detection dimensions to obtain the access detection result of the unmanned mining dump truck under test, so as to complete the safety access detection of the unmanned mining dump truck based on the access detection result.

[0101] In one feasible implementation, the configuration module includes:

[0102] The setting module is used to set multiple detection dimensions for the unmanned mining dump truck based on its working environment and functional characteristics.

[0103] The binding module is used to extract the detection elements of each detection dimension, set the corresponding detection method based on the detection elements, and bind the detection method to the detection dimension.

[0104] In one feasible implementation, the calling module includes:

[0105] The determination module is used to identify multiple core execution modules of the unmanned mining dump truck under test, and to determine the detection parameters of each core execution module.

[0106] The data acquisition module is used to control the detection method to test each detection parameter and to acquire detection data of the functional safety detection dimension through multiple acquisition methods.

[0107] In one feasible implementation, the calling module further includes:

[0108] The module is used to construct failure scenarios of the unmanned mining dump truck under test in terms of braking, controller and manual takeover interface based on virtual technology.

[0109] The operation module is used to control the unmanned mining dump truck under test to run at full load in failure scenarios of braking, controller and manual takeover interface, so as to obtain the corresponding test data.

[0110] In one feasible implementation, the calling module also includes:

[0111] A driving module is used to identify the driving module and communication module that affect the unmanned driving of the unmanned mining dump truck under test.

[0112] The testing module is used to test and collect detection data of the driving module and the communication module under typical real-world and simulated scenarios.

[0113] In one feasible implementation, the fusion module includes:

[0114] The judgment module is used to determine whether the evaluation data and the fault data both meet the preset admission conditions;

[0115] The generation module is used to determine whether the obtained access detection result is qualified and to generate an access detection report based on the access detection result.

[0116] In one feasible implementation, the fusion module further includes:

[0117] The prediction module is used to perform trend prediction based on the detection data and historical detection data in order to predict intermediate data.

[0118] The module is used to determine whether the intermediate data meets the preset fault conditions, so as to obtain the fault data based on the intermediate data.

[0119] Example 3

[0120] This application also provides an electronic device, such as Figure 4 As shown, it includes: a processor 401, a memory 402, and a bus 403. The memory 402 stores machine-readable instructions that can be executed by the processor 401. When the electronic device is running, the processor 401 and the memory 402 communicate through the bus 403. When the machine-readable instructions are executed by the processor 401, the steps of any one of the safety access detection methods for an unmanned mining dump truck are performed.

[0121] Example 4

[0122] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of any one of the safety access detection methods for an unmanned mining dump truck.

Claims

1. A safety access detection method for an unmanned mine dump truck, characterized in that, The method includes: Multiple detection dimensions for the unmanned mining dump truck under test are pre-configured, and a corresponding detection environment is configured for each detection dimension; different detection dimensions correspond to different detection methods; the detection environment includes typical real-world environments and simulated environments; The unmanned mining dump truck under test is controlled to be in a target detection environment, and the corresponding target detection method is called to detect the unmanned mining dump truck under test to obtain the corresponding detection data; the detection data includes vehicle body data and simulation data; Simulated data from multiple detection dimensions and vehicle body data are input into pre-trained admission evaluation models and fault prediction models, respectively, to process the detection data from the multiple detection dimensions and vehicle body data to obtain evaluation data and fault data. By integrating fault data and evaluation data from multiple detection dimensions, the access detection results of the unmanned mining dump truck under test are obtained, and the safety access detection of the unmanned mining dump truck is completed based on the access detection results.

2. The method of claim 1, wherein the method further comprises: The pre-configured multiple detection dimensions for the unmanned mining dump truck under test include: Based on the working environment and functional characteristics of the unmanned mining dump truck under test, multiple detection dimensions are set for the unmanned mining dump truck. Extract the detection elements for each detection dimension, set the corresponding detection method based on the detection elements, and bind the detection method to the detection dimension.

3. The safety access detection method for unmanned mining dump trucks as described in claim 2, characterized in that, The detection dimensions include at least the functional safety detection dimensions; The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes: Multiple core execution modules of the unmanned mining dump truck under test were identified, and the detection parameters of each core execution module were determined. The detection method is controlled to test each detection parameter, and detection data for the functional safety detection dimension is collected through multiple acquisition methods.

4. The safety access detection method for unmanned mining dump trucks as described in claim 2, characterized in that, The detection dimensions include at least redundancy and emergency detection dimensions; The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes: Based on virtual technology, a failure scenario was constructed for the unmanned mining dump truck under test in terms of braking, controller and manual control interface. The unmanned mining dump truck under test is controlled to run at full load in failure scenarios of braking, controller and manual takeover interface, so as to obtain the corresponding test data.

5. The safety access detection method for unmanned mining dump trucks as described in claim 2, characterized in that, The detection dimensions include at least the autonomous driving dimension; The step of calling the corresponding target detection method to detect the unmanned mining dump truck under test and obtaining the corresponding detection data includes: The driving module and communication module that affect the unmanned driving of the unmanned mining dump truck under test were identified. Test and collect detection data for the driving module and communication module in typical real-world and simulated scenarios.

6. The safety access detection method for unmanned mining dump trucks as described in claim 2, characterized in that, The access detection results of the unmanned mining dump truck under test are obtained by integrating fault data from multiple detection dimensions and evaluation data, including: Determine whether both the evaluation data and the fault data meet the preset admission criteria; If so, the obtained access test result is qualified, and an access test report is generated based on the access test result.

7. The safety access detection method for unmanned mining dump trucks as described in claim 1, characterized in that, The detection data from the multiple detection dimensions and the vehicle body data are processed separately to obtain evaluation data and fault data, including: Trend prediction is performed based on the aforementioned detection data and historical detection data to predict intermediate data; Determine whether the intermediate data meets the preset fault conditions, so as to obtain the fault data based on the intermediate data.

8. A safety access detection system for an unmanned mining dump truck, characterized in that, The system includes: The configuration module is used to pre-configure multiple detection dimensions for the unmanned mining dump truck under test, and to configure a corresponding detection environment for each detection dimension; different detection dimensions correspond to different detection methods; the detection environment includes typical real-world environments and simulated environments; The calling module is used to control the unmanned mining dump truck under test to be in a target detection environment, and to call the corresponding target detection method to detect the unmanned mining dump truck under test and obtain the corresponding detection data; the detection data includes vehicle body data and simulation data; The processing module is used to input simulated data from multiple detection dimensions and vehicle body data into the pre-trained admission evaluation model and fault prediction model, respectively, so as to process the detection data from the multiple detection dimensions and vehicle body data to obtain evaluation data and fault data. The fusion module is used to fuse fault data and evaluation data from multiple detection dimensions to obtain the access detection results of the unmanned mining dump truck under test, so as to complete the safety access detection of the unmanned mining dump truck based on the access detection results.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, they perform the steps of a safety access detection method for an unmanned mining dump truck as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of a safety access detection method for an unmanned mining dump truck as described in any one of claims 1 to 7.

Images