Method, device and equipment for determining passing parameters of unmanned vehicle

Abstract

The present application relates to the determination method, device, equipment and storage medium of the passing parameter of unmanned vehicle, belongs to unmanned technical field.The present application includes: being applied to unmanned mine truck, collecting multiple historical passing data of mine truck, preprocessing historical passing data, obtaining normal passing data, according to event scene, data segmentation is carried out to normal passing data, according to the value of the driving speed and each passing parameter in normal passing data, regression analysis is carried out to normal passing data under each scene, obtain the relationship curve between multiple passing parameters and the driving speed under the scene, according to the relationship curve between passing parameter and driving speed, determine the standard value range of the passing parameter of unmanned mine truck under each scene.The present application is helpful to solve the problem of few systematic construction and determination method of passing parameter of mine truck in prior art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned driving, in particular to a method and device for determining passing parameters of an unmanned vehicle. BACKGROUND

[0002] In recent years, the unmanned driving technology has developed rapidly. For the unmanned driving technology, the unmanned vehicle should not make driving behavior beyond the expectation of human beings, that is, the driving behavior parameters of the unmanned vehicle need to be consistent with human beings.

[0003] In a mining area, due to the relatively closed road scene of the mining area, the single driving route of the transport vehicle, and the fewer surrounding traffic participants, it is a typical scenario for the landing and application of unmanned driving. It is of great significance to formulate human-like passing rules for mining unmanned vehicles to improve the transport efficiency and safety of the mining area. However, at present, there is few systematic construction and determination method of passing parameters of mining trucks for the unmanned driving technology of the mining area. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a method and device for determining passing parameters of an unmanned vehicle, so as to solve the problem of few systematic construction and determination method of passing parameters of mining trucks in the prior art.

[0005] According to a first aspect of an embodiment of the present application, a method for determining passing parameters of a vehicle is provided, which is applied to an unmanned mining truck, and includes:

[0006] Collecting a plurality of historical passing data of the mining truck with a driver, wherein the historical passing data includes an event scene, a driving speed and values of a plurality of passing parameters;

[0007] Preprocessing the historical passing data to obtain normal passing data;

[0008] According to the event scene, the normal passing data is divided to obtain normal passing data under a plurality of scenes;

[0009] According to the driving speed and the values of the passing parameters in the normal passing data, a regression analysis is performed on the normal passing data under each scene to obtain a relationship curve between the plurality of passing parameters and the driving speed under the scene;

[0010] According to the relationship curve between the passing parameters and the driving speed, a standard value range of the passing parameters under each scene of the unmanned mining truck is determined.

[0011] Preferably, the preprocessing of the historical passing data to obtain normal passing data includes:

[0012] determine whether the driving speed in the historical traffic data conforms to a normal distribution, if yes, take the historical traffic data as normal distribution traffic data, and if not, take the historical traffic data as non-normal distribution traffic data;

[0013] perform abnormal data elimination on the normal distribution traffic data and the non-normal distribution traffic data respectively to obtain normal traffic data.

[0014] Preferably, the abnormal data elimination on the normal distribution traffic data comprises:

[0015] calculate the mean and the standard deviation of the driving speed in all normal distribution traffic data respectively;

[0016] calculate a normal distribution abnormal interval according to the mean and the standard deviation;

[0017] determine whether the driving speed in the normal distribution traffic data is located in the normal distribution abnormal interval, if yes, determine that the normal distribution traffic data is abnormal data and eliminate it.

[0018] Preferably, the abnormal data elimination on the non-normal distribution traffic data comprises:

[0019] sort the driving speed in the non-normal distribution traffic data in ascending order;

[0020] calculate four quantiles of the sorted driving speed;

[0021] calculate a non-normal distribution abnormal interval according to the four quantiles;

[0022] determine whether the driving speed in the non-normal distribution traffic data is located in the non-normal distribution abnormal interval, if yes, determine that the non-normal distribution traffic data is abnormal data and eliminate it.

[0023] Preferably, the traffic parameters comprise a steering wheel angle parameter, a vehicle headway parameter and a post-intersection intrusion time parameter, and the regression analysis on the normal traffic data in each scenario to obtain a relationship curve between the driving speed and the traffic parameters in the scenario comprises:

[0024] establish a loss function between the traffic parameters and the driving speed;

[0025] According to the preset quantile, the loss function is optimized by using the values of the driving speed and the steering wheel angle parameter in the normal traffic data under the scene, and after optimization, a relationship curve between the steering wheel angle parameter and the driving speed under the scene is obtained;

[0026] According to the preset quantile, the loss function is optimized by using the values of the driving speed and the headway parameter in the normal traffic data under the scene, and after optimization, a relationship curve between the headway parameter and the driving speed under the scene is obtained;

[0027] According to the preset quantile, the loss function is optimized by using the values of the driving speed and the post-intersection invasion time parameter in the normal traffic data under the scene, and after optimization, a relationship curve between the post-intersection invasion time parameter and the driving speed under the scene is obtained.

[0028] Preferably, the standard value range of the traffic parameter of the unmanned mine truck under each scene is determined according to the relationship curve between the traffic parameter and the driving speed, comprising:

[0029] The standard value range of the steering wheel angle parameter of the unmanned mine truck under each scene is determined according to the relationship curve between the steering wheel angle parameter and the driving speed under each scene;

[0030] The standard value range of the headway parameter of the unmanned mine truck under each scene is determined according to the relationship curve between the headway parameter and the driving speed under each scene;

[0031] The standard value range of the post-intersection invasion time parameter of the unmanned mine truck under each scene is determined according to the relationship curve between the post-intersection invasion time parameter and the driving speed under each scene.

[0032] Preferably, the event scene includes meeting, following and intersection conflict, and the normal traffic data is divided into normal traffic data under multiple scenes according to the event scene, comprising:

[0033] The normal traffic data with the event scene of meeting is selected from all normal traffic data as the normal traffic data under the meeting scene;

[0034] The normal traffic data with the event scene of following is selected from all normal traffic data as the normal traffic data under the following scene;

[0035] The normal traffic data with the event scene of intersection conflict is selected from all normal traffic data as the normal traffic data under the intersection scene.

[0036] According to a second aspect of the embodiments of the present application, a mine truck control method applied to an unmanned mine truck is provided, comprising:

[0037] The method for determining the passing parameters of the vehicle according to any one of the above embodiments is used to determine the standard value range of the passing parameters of the unmanned mine truck in each scene.

[0038] The passing of the unmanned mine truck is controlled according to the standard value range.

[0039] According to a third aspect of the embodiments of the present application, a device for determining the passing parameters of a vehicle applied to an unmanned mine truck is provided, comprising:

[0040] A data collection module is configured to collect a plurality of historical passing data of a mine truck with a driver, wherein the historical passing data comprises an event scene, a driving speed and values of a plurality of passing parameters.

[0041] A data preprocessing module is configured to preprocess the historical passing data to obtain normal passing data.

[0042] A data segmentation module is configured to segment the normal passing data according to the event scene to obtain normal passing data in a plurality of scenes.

[0043] A regression analysis module is configured to perform regression analysis on the normal passing data in each scene according to the driving speed and the values of the passing parameters in the normal passing data to obtain a relationship curve between the passing parameters and the driving speed in the scene.

[0044] A parameter standard determination module is configured to determine a standard value range of the passing parameters of the unmanned mine truck in each scene according to the relationship curve between the passing parameters and the driving speed.

[0045] According to a fourth aspect of the embodiments of the present application, a device for determining the passing parameters of a vehicle is provided, comprising:

[0046] A memory having an executable program stored thereon;

[0047] A processor configured to execute the executable program in the memory to implement the steps of the method according to any one of the above embodiments.

[0048] According to a fifth aspect of the embodiments of the present application, a computer readable storage medium is provided, which stores computer instructions for causing a computer to execute the steps of the method according to any one of the above embodiments.

[0049] The technical solutions provided by the embodiments of the present application can have the following beneficial effects:

[0050] By having a plurality of historical passing data of the mining truck driven by the driver, the historical passing data including the event scene, the driving speed and the values of the plurality of passing parameters, pre-processing the historical passing data, eliminating the abnormal data, obtaining the normal passing data, according to the event scene, data segmentation is performed on the normal passing data, obtaining the normal passing data under a plurality of scenes, and according to the driving speed and the values of the plurality of passing parameters in the normal passing data, regression analysis is performed on the normal passing data under each scene, obtaining the relationship curve between the plurality of passing parameters and the driving speed under the scene, and according to the relationship curve between the passing parameters and the driving speed, the standard value range of the passing parameters of the unmanned mining truck under each scene is determined, so that the standard value range of the passing parameters of the unmanned mining truck can be determined more accurately according to a large amount of historical passing data when the driver drives, and the passing of the unmanned mining truck is controlled according to the standard value range, thereby effectively solving the problem of few systematic construction and determination methods of the passing parameters of the mining truck in the prior art, formulating the human-like passing rules for the unmanned mining truck, improving the transportation efficiency of the mining area, and improving the transportation safety of the mining area.

[0051] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0052] The accompanying drawings, which are incorporated into and form part of the specification, illustrate an embodiment consistent with the present application and, together with the description, serve to explain the principles of the application.

[0053] Figure 1 is a flow diagram of a determination method of a passing parameter of a vehicle according to an exemplary embodiment;

[0054] Figure 2 is a schematic diagram of a relationship curve of a passing parameter of a vehicle according to an exemplary embodiment;

[0055] Figure 3 is a block diagram of a determination device of a passing parameter of a vehicle according to an exemplary embodiment. DETAILED DESCRIPTION

[0056] The exemplary embodiments will be described in detail herein with reference to the attached drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of apparatuses and methods consistent with some aspects of the present application as detailed in the appended claims.

[0057] The application provides a method for determining a passing parameter of a vehicle, which is applied to an unmanned mine truck, and refers to Figure 1 , Figure 1 is a flowchart of a method according to an exemplary embodiment, which comprises:

[0058] Step S11, collecting a plurality of historical passing data of the mine truck with a driver, wherein the historical passing data comprises an event scene, a driving speed and values of a plurality of passing parameters;

[0059] Step S12, preprocessing the historical passing data to obtain normal passing data;

[0060] Step S13, performing data segmentation on the normal passing data according to the event scene to obtain normal passing data under a plurality of scenes;

[0061] Step S14, performing regression analysis on the normal passing data under each scene according to the driving speed and the values of the passing parameters in the normal passing data to obtain a relationship curve between the passing parameters and the driving speed under the scene;

[0062] Step S15, determining a standard value range of the passing parameters of the unmanned mine truck under each scene according to the relationship curve between the passing parameters and the driving speed.

[0063] Specifically, a plurality of historical passing data of the mine truck with a driver is collected, and the historical passing data comprises an event scene, a driving speed and values of a plurality of passing parameters. The historical passing data is passing data of the mine truck collected when the mine truck is driven by a person. The event scene is a traffic event encountered when collecting the passing data when the mine truck is driven by a person, such as a following traffic event, an intersection traffic event and a meeting traffic event. The driving speed is a current speed of the mine truck collected when the mine truck is driven by a person. The passing parameters can include a steering wheel rotation angle, a vehicle head distance, a post-intersection encroachment time, a brake pedal intensity and an accelerator pedal intensity.

[0064] The steering wheel rotation angle is an angle of rotation of the steering wheel of the mine truck under a current traffic event at a current collection time. The vehicle head distance can be a distance between the vehicle head of the collected mine truck and other vehicles under the current traffic event. The post-intersection encroachment time can be a post-encroachment time (PET) of the collected mine truck and other vehicles at an intersection.

[0065] It should be noted that PET is an index widely used to evaluate the severity of traffic conflicts, which refers to the difference between the time of the current mine truck and the surrounding conflict vehicles reaching the intersection conflict point, has measurability and reliability, and can be used as a safety evaluation standard for the intersection conflict point. Moreover, the smaller the PET value, the greater the risk of traffic conflict between the current mine truck and the surrounding conflict vehicles at the intersection. The calculation method of PET is as follows: according to the following formula, the PET between the current mine truck and the surrounding conflict vehicles is calculated:

[0066] PET = T i -T i-1 ,

[0067] wherein T i is the time of the current mine truck reaching the intersection conflict point, and T i-1 is the time of the surrounding conflict vehicle reaching the intersection conflict point.

[0068] Specifically, the collection of historical traffic data of mine trucks can be actual traffic data of human-driven mine trucks in real scenarios, or traffic data simulated by a mine truck driving simulation platform. The mine truck driving simulation platform can provide typical mine driving simulation scenarios, and the typical mine driving simulation scenarios can set more scene modes through mode setting, such as typical road scenes, surrounding mine scenes, weather scenes, and traffic participant scenes, so as to collect more diverse traffic data. The typical road scenes can further include sharp bends, steep slopes, long straight sections, and intersections, etc. The surrounding mine scenes can be set as mine, mine plant, mine mechanical facilities, and various mine scenes according to actual mine scenes. The adverse weather scenes include rain, snow, fog, dust, and various weather scenes. The traffic participant scene is a mine unmanned driving truck scene. According to the more diverse traffic data collected, the accuracy of the standard range of traffic parameters can be ensured when determining the standard range of traffic parameters, and more scenes can make the unmanned mine truck operate more standardized and safer when facing different scenes.

[0069] It should be noted that in order to ensure the integrity of the collected historical traffic data of mine trucks and generate more traffic events, the traffic participants can also set unmanned mine truck vehicles, and the unmanned mine truck vehicles can also drive in formation to create more traffic participants and obtain more driving scenes, thereby improving the accuracy of parameter determination of the unmanned mine truck.

[0070] In the adaptive cruise control (ACC) model, more unmanned mine trucks can be set as traffic participants to expand the collection amount of historical traffic data. For the unmanned mine truck as a traffic participant, the following method can be used to set the following acceleration: when the following vehicle speed is greater than 3 km / h and the following distance is less than 1.8 m, the maximum deceleration of the vehicle is used; when the following vehicle speed is greater than 3 km / h and the following distance is greater than 1.8 m, the acceleration a of the vehicle is calculated according to the following formula:

[0071] a = 1.12 * (d - v f *t safe ) + 1.70 * Δv.

[0072] Where d represents the headway, v f represents the speed of the vehicle (following vehicle), t safe represents the safe headway, and Δv represents the speed difference between the two vehicles.

[0073] It should be noted that the traffic data can also include the speed, acceleration, lateral offset, position and travel time of the mine truck. The collection frequency of the traffic data can be between 10 Hz and 100 Hz, or can be set according to specific needs, and the present application does not make specific limitations.

[0074] Specifically, the historical traffic data collected above is preprocessed to identify historical traffic data that does not obviously conform to the actual situation as abnormal data, and after removing abnormal data, normal traffic data is obtained. Because there are many scenarios of mine trucks during road travel, different abnormal data identification can be performed on historical traffic data according to different scenarios to obtain normal traffic data.

[0075] The method of combining kernel density estimation and frequency distribution histogram can also be used to preliminarily analyze the distribution characteristics of the speed, acceleration, headway and steering wheel angle of the mine truck during driving, and to determine whether the historical traffic data meets the requirements.

[0076] Because there are many scenarios of mine trucks, for different traffic events, the normal traffic data can be divided according to the event scenarios in the historical traffic data to obtain the normal traffic data under each event scenario, which facilitates the setting of different standard value ranges of traffic parameters for different event scenarios.

[0077] For the normal traffic data obtained by the above processing, for each scene, the relationship curve between each traffic parameter and the driving speed in the scene can be obtained according to the driving speed and the value of each traffic parameter in the normal traffic data in the scene. When performing regression analysis, different regression analysis methods can be used, and for the scene, when analyzing the relationship between each traffic parameter and the driving speed, the constraint condition of the relationship curve between the traffic parameter and the driving speed can be set as needed to obtain multiple relationship curves between the traffic parameter and the driving speed. In this way, the relationship curve between each traffic parameter and the driving speed in any scene can be obtained. As shown in Figure 2 Figure 2 is a schematic diagram of a relationship curve of a traffic parameter of a vehicle according to an example embodiment, in which the upper limit value and the lower limit value are the relationship curve between the headway parameter and the driving speed.

[0078] For any scene, the upper limit value and the lower limit value of each traffic parameter at any driving speed can be determined according to the relationship curve between each traffic parameter and the driving speed obtained by the above regression analysis, and the value between the determined upper limit value and the lower limit value of the traffic parameter is taken as the standard value range of the traffic parameter of the unmanned mine truck in the scene.

[0079] It can be understood that the technical solution provided by the embodiment can collect multiple historical traffic data of the mine truck, the historical traffic data including the event scene, the driving speed and the values of multiple traffic parameters, pre-process the historical traffic data to eliminate abnormal data and obtain normal traffic data, perform data segmentation on the normal traffic data according to the event scene to obtain normal traffic data in multiple scenes, perform regression analysis on the normal traffic data in each scene according to the driving speed and the values of each traffic parameter in the normal traffic data, obtain the relationship curve between multiple traffic parameters and the driving speed in the scene, and determine the standard value range of the traffic parameter of the unmanned mine truck in each scene according to the relationship curve between the traffic parameter and the driving speed, so as to accurately determine the standard value range of the traffic parameter of the unmanned mine truck according to a large amount of historical traffic data of the driver, effectively solve the problem of few systematic construction and determination methods of the traffic parameter of the mine truck in the prior art, formulate human-like traffic rules for the unmanned mine truck, improve the transportation efficiency of the mine area, and improve the transportation safety of the mine area.

[0080] Preferably, in step S12, the historical traffic data is pre-processed to obtain normal traffic data, including:

[0081] ​Step S1212, judging whether the driving speed in the historical traffic data conforms to normal distribution, if conforming to normal distribution, taking the historical traffic data as normal distribution traffic data, if not conforming to normal distribution, taking the historical traffic data as non-normal distribution traffic data;

[0082] Step S122, respectively performing abnormal data elimination on the normal distribution traffic data and the non-normal distribution traffic data, to obtain normal traffic data.

[0083] Specifically, when pre-processing the historical traffic data to obtain normal traffic data, the following steps are included:

[0084] For the driving speed in each piece of historical traffic data under each scene, for historical traffic data under a scene, judging whether the driving speed conforms to normal distribution, if conforming to normal distribution, taking the historical traffic data as normal distribution traffic data, if not conforming to normal distribution, taking the historical traffic data as non-normal distribution traffic data.

[0085] For the normal distribution traffic data and the non-normal distribution traffic data under each scene, different methods are used to identify abnormal data, and after the abnormal data is eliminated, all normal traffic data under the scene is obtained.

[0086] It can be understood that by eliminating abnormal data, the accuracy of the standard value range of the determined traffic parameter can be improved, and unnecessary influence caused by abnormal data can be excluded.

[0087] Preferably, in the step S122, the abnormal data elimination on the normal distribution traffic data includes:

[0088] The mean and the standard deviation of the driving speed in all normal distribution traffic data are respectively calculated;

[0089] According to the mean and the standard deviation, a normal distribution abnormal interval is calculated;

[0090] Judging whether the driving speed in the normal distribution traffic data is located in the normal distribution abnormal interval, if located in the normal distribution abnormal interval, determining that the normal distribution traffic data is abnormal data, and eliminating it.

[0091] Specifically, when performing abnormal data elimination on the normal distribution traffic data, the following steps are included:

[0092] For all the normal distribution traffic data obtained above, the mean and standard deviation of the driving speed are calculated respectively, and the normal interval range is set according to the mean and standard deviation, and the opposite area is the normal distribution abnormal interval. The driving speed in all the normal distribution traffic data is judged to determine whether the driving speed is located in the normal distribution abnormal interval. When the driving speed is located in the normal distribution abnormal interval, it is determined that the normal distribution traffic data is abnormal data, which is excluded. When the driving speed is not located in the normal distribution abnormal interval, it is determined that the normal distribution traffic data is normal data, which is retained for subsequent calculation of the standard value range of the traffic parameter.

[0093] In a specific example, starting from the mean (μ), a range of 3 standard deviations (σ) is extended to both sides, i.e. the range between μ-3σ to μ+3σ, as the normal interval range, and the opposite area is the normal distribution abnormal interval. When the historical traffic data point falls within the range of μ-3σ to μ+3σ, it is considered to be normal data, and when the historical traffic data point falls outside the range of μ-3σ to μ+3σ, it is considered to be abnormal data, and the historical traffic data point is excluded.

[0094] Preferably, in the step S122, the abnormal data exclusion of the non-normal distribution traffic data comprises:

[0095] The driving speeds in the non-normal distribution traffic data are sorted in ascending order;

[0096] The four quartiles of the sorted driving speeds are calculated;

[0097] The non-normal distribution abnormal interval is calculated according to the four quartiles;

[0098] It is judged whether the driving speed in the non-normal distribution traffic data is located in the non-normal distribution abnormal interval. If it is located in the non-normal distribution abnormal interval, it is determined that the non-normal distribution traffic data is abnormal data, which is excluded.

[0099] Specifically, when the abnormal data exclusion of the non-normal distribution traffic data is performed, it comprises:

[0100] For all the non-normal distribution traffic data obtained above, the driving speeds in all the non-normal distribution traffic data are sequentially sorted in ascending order, and the quartiles and interquartile ranges are used as the criteria for judging whether the non-normal distribution traffic data is abnormal data. The four quartiles Q1, Q2, Q3 and Q4 of the sorted driving speeds are calculated, and Q1, Q2, Q3 and Q4 are the driving speeds located at 25%, 50%, 75% and 100% respectively.

[0101] According to the values of the four quantiles Q1, Q2, Q3 and Q4, the bit distance h1 = Q1-1.5(Q2-Q1) and h2 = Q2+1.5(Q2-Q1) are calculated, h1 and h2 are taken as the upper limit value and the lower limit value of the driving speed in the non-normal distribution traffic data, greater than or equal to h1 and less than or equal to h2 is the non-normal distribution normal interval, less than h1 or greater than h2 is the non-normal distribution abnormal interval.

[0102] When the non-normal distribution traffic data point falls within the range of h1 to h2, the historical traffic data point is considered to be normal data, and when the non-normal distribution traffic data point falls outside the range of h1 to h2, the historical traffic data point is considered to be abnormal data, and the historical traffic data point is excluded.

[0103] Preferably, the traffic parameters include steering wheel angle parameters, headway parameters and intersection post-invasion time parameters, and in the step S14, the normal traffic data in each scene is analyzed by regression according to the values of the driving speed and each traffic parameter in the normal traffic data, and a relationship curve between the multiple traffic parameters and the driving speed in the scene is obtained, including:

[0104] A loss function between each traffic parameter and the driving speed is established.

[0105] According to the preset quantile, the loss function is optimized by using the values of the driving speed and the steering wheel angle parameter in the normal traffic data in the scene, and after optimization, a relationship curve between the steering wheel angle parameter and the driving speed in the scene is obtained.

[0106] According to the preset quantile, the loss function is optimized by using the values of the driving speed and the headway parameter in the normal traffic data in the scene, and after optimization, a relationship curve between the headway parameter and the driving speed in the scene is obtained.

[0107] According to the preset quantile, the loss function is optimized by using the values of the driving speed and the intersection post-invasion time parameter in the normal traffic data in the scene, and after optimization, a relationship curve between the intersection post-invasion time parameter and the driving speed in the scene is obtained.

[0108] Specifically, the traffic parameters can include steering wheel angle parameters, headway parameters and intersection post-invasion time parameters, different traffic parameters can be selected in different scenes, and multiple traffic parameters can also be set. When the normal traffic data in each scene is analyzed by regression according to the values of the driving speed and each traffic parameter, and a relationship curve between the multiple traffic parameters and the driving speed in the scene is obtained, including:

[0109] Following the "human concept" and "human vehicle rule" idea, in order to ensure the public trust in the unmanned mine truck and the road traffic safety, the passing rule of the unmanned mine truck needs to be consistent with human, that is, based on the normal passing data obtained after the above processing, the human-like passing rule is extracted through the quantile regression method. For the normal passing data in each scene, the loss function between each passing parameter and the driving speed is established, and the formula of the loss function is as follows:

[0110]

[0111] Wherein, β is the coefficient vector, x is the driving speed in each normal passing data, y is each passing parameter, τ is a preset value, and i is the number of normal passing data.

[0112] For each scene, according to the preset quantile, that is, τ, which can be 15% and 85%, or 25%, 75%, etc., the driving speed and the value of the passing parameter in all normal passing data in this scene are used to optimize the above loss function, so that the loss function is minimized, the optimization is completed, and the coefficient vector β is obtained after the optimization. According to the coefficient vector β, the relationship curve between the passing parameter and the driving speed in this scene can be drawn. The selected quantile is different, and the calculated relationship curve is also different.

[0113] According to the preset multiple quantiles, the driving speed and the value of the steering wheel angle parameter in the normal passing data in this scene are used to optimize the above loss function, and after minimizing the loss function, the optimization is completed, and the relationship curve between the steering wheel angle parameter and the driving speed in this scene is obtained. For example, multiple relationship curves between the steering wheel angle parameter and the driving speed in the meeting scene can be calculated.

[0114] According to the preset multiple quantiles, the driving speed and the value of the steering wheel angle parameter in the normal passing data in this scene are used to optimize the above loss function, and after minimizing the loss function, the optimization is completed, and the relationship curve between the steering wheel angle parameter and the driving speed in this scene is obtained. For example, multiple relationship curves between the steering wheel angle parameter and the driving speed in the meeting scene can be calculated.

[0115] According to the preset multiple quantiles, the driving speed and the value of the steering wheel angle parameter in the normal passing data in this scene are used to optimize the above loss function, and after minimizing the loss function, the optimization is completed, and the relationship curve between the steering wheel angle parameter and the driving speed in this scene is obtained. For example, multiple relationship curves between the steering wheel angle parameter and the driving speed in the meeting scene can be calculated.

[0116] Preferably, in the step S15, the standard value range of the passing parameter of the unmanned mine truck in each scene is determined according to the relationship curve between the passing parameter and the driving speed, comprising:

[0117] The standard value range of the steering wheel angle parameter of the unmanned mine truck in each scene is determined according to the relationship curve between the steering wheel angle parameter and the driving speed in each scene.

[0118] The standard value range of the headway parameter of the unmanned mine truck in each scene is determined according to the relationship curve between the headway parameter and the driving speed in each scene.

[0119] The standard value range of the post-intersection intrusion time parameter of the unmanned mine truck in each scene is determined according to the relationship curve between the post-intersection intrusion time parameter and the driving speed in each scene.

[0120] Specifically, the standard value range of the passing parameter of the unmanned mine truck in each scene is determined according to the relationship curve between the passing parameter and the driving speed, comprising:

[0121] For any one scene, according to the relationship curve between the steering wheel angle parameter and the driving speed in the scene, since a plurality of quantiles can be set, a plurality of relationship curves are obtained, and according to actual needs, two relationship curves obtained when the 15% and 85% quantiles are selected are taken as the upper limit relationship curve and the lower limit relationship curve between the steering wheel angle parameter and the driving speed, and the values between the upper limit relationship curve and the lower limit relationship curve are all standard relationship values between the steering wheel angle parameter and the driving speed, that is, the standard value range of the steering wheel angle parameter of the unmanned mine truck in the scene is obtained.

[0122] For any one scene, according to the relationship curve between the headway parameter and the driving speed in the scene, since a plurality of quantiles can be set, a plurality of relationship curves are obtained, and according to the plurality of relationship curves between the headway parameter and the driving speed in each scene, the standard value range of the headway parameter of the unmanned mine truck in the scene is determined by the same method.

[0123] For any one scene, according to the relationship curve between the post-intersection intrusion time parameter and the driving speed in the scene, since a plurality of quantiles can be set, a plurality of relationship curves are obtained, and according to the relationship curve between the post-intersection intrusion time parameter and the driving speed in each scene, the standard value range of the post-intersection intrusion time parameter of the unmanned mine truck in the scene is determined by the same method.

[0124] Preferably, the event scenarios include meeting, following and intersection conflict, and the step S13 of data segmentation of the normal traffic data according to the event scenarios includes:

[0125] selecting normal traffic data with the event scenario of meeting from all the normal traffic data as normal traffic data under the scenario of meeting;

[0126] selecting normal traffic data with the event scenario of following from all the normal traffic data as normal traffic data under the scenario of following;

[0127] selecting normal traffic data with the event scenario of intersection conflict from all the normal traffic data as normal traffic data under the scenario of intersection.

[0128] Specifically, the event scenarios include meeting, following and intersection conflict, and the step S13 of data segmentation of the normal traffic data according to the event scenarios includes:

[0129] selecting normal traffic data with the event scenario of meeting from all the normal traffic data as normal traffic data under the scenario of meeting; selecting normal traffic data with the event scenario of following from all the normal traffic data as normal traffic data under the scenario of following; and selecting normal traffic data with the event scenario of intersection conflict from all the normal traffic data as normal traffic data under the scenario of intersection. The traffic event scenarios include meeting, following and intersection conflict.

[0130] Among them, the intersection conflict events include seven typical intersection conflict types:

[0131] 1) Intersection conflict 1: the host vehicle is straight, and the event vehicle is right-side entry left turn;

[0132] 2) Intersection conflict 2: the host vehicle is left turn, and the event vehicle is right-side entry left turn;

[0133] 3) Intersection conflict 3: the host vehicle is left turn, and the event vehicle is right-side entry straight;

[0134] 4) Intersection conflict 4: the host vehicle is straight, and the event vehicle is opposite entry left turn;

[0135] 5) Intersection conflict 5: the host vehicle is straight, and the event vehicle is right-side entry straight;

[0136] 6) Intersection conflict 6: the host vehicle is right turn, and the event vehicle is opposite entry left turn;

[0137] 7) Intersection conflict 7: the host vehicle is straight, and the event vehicle is right-side entry right turn.

[0138] It can be understood that the technical scheme provided by the embodiment, by collecting a plurality of historical passing data of the mine truck, the historical passing data including an event scene, a driving speed and values of a plurality of passing parameters, pre-processing the historical passing data, eliminating abnormal data, obtaining normal passing data, according to the event scene, performing data segmentation on the normal passing data to obtain normal passing data under a plurality of scenes, and according to the driving speed and the values of the passing parameters in the normal passing data, performing regression analysis on the normal passing data under each scene to obtain a relationship curve between the plurality of passing parameters and the driving speed under the scene, the standard value range of the passing parameters of the unmanned mine truck under each scene is determined according to the relationship curve between the passing parameters and the driving speed, so that the standard value range of the passing parameters of the unmanned mine truck can be accurately determined according to a large amount of historical passing data of the driver, thereby effectively solving the problem of few systematic construction and determination methods of the passing parameters of the mine truck in the prior art, formulating human-like passing rules for the unmanned mine truck, improving the transportation efficiency of the mine area, and improving the transportation safety of the mine area.

[0139] The application provides a mine truck control method applied to an unmanned mine truck, comprising the following steps:

[0140] The method for determining the passing parameters of the vehicle according to any one of the above embodiments is used to determine the standard value range of the passing parameters of the unmanned mine truck under each scene.

[0141] The passing of the unmanned mine truck is controlled according to the standard value range.

[0142] Specifically, after the standard value range of the passing parameters of the unmanned mine truck under each scene is determined by the above method, the passing parameters under each scene can be set for all unmanned mine trucks according to the standard value range, and the passing of the unmanned mine truck is controlled according to the standard value range of the passing parameters, thereby improving the utilization rate of the mine truck, reducing the workload and pressure of the driver, and ensuring the driving safety of the mine truck and reducing the occurrence of safety accidents.

[0143] The application provides a device for determining passing parameters of a vehicle applied to an unmanned mine truck, which is described as follows: Figure 3 , Figure 3 is a block diagram of a device for determining passing parameters of a vehicle according to an exemplary embodiment, which comprises:

[0144] The data acquisition module 31 is used to collect a plurality of historical passing data of the mine truck with a driver, and the historical passing data includes an event scene, a driving speed and values of a plurality of passing parameters.

[0145] The data preprocessing module 32 is configured to preprocess the historical traffic data to obtain normal traffic data.

[0146] The data segmentation module 33 is configured to segment the normal traffic data according to the event scene to obtain normal traffic data in multiple scenes.

[0147] The regression analysis module 34 is configured to perform regression analysis on the normal traffic data in each scene according to the driving speed and the values of the traffic parameters in the normal traffic data to obtain a relationship curve between the driving speed and the traffic parameters in the scene.

[0148] The parameter standard determination module 35 is configured to determine a standard value range of the traffic parameters of the unmanned mine truck in each scene according to the relationship curve between the traffic parameters and the driving speed.

[0149] Preferably, the preprocessing of the historical traffic data to obtain normal traffic data comprises:

[0150] determining whether the driving speed in the historical traffic data conforms to a normal distribution, and if so, regarding the historical traffic data as normal distribution traffic data, and if not, regarding the historical traffic data as non-normal distribution traffic data;

[0151] performing abnormal data elimination on the normal distribution traffic data and the non-normal distribution traffic data respectively to obtain normal traffic data.

[0152] Preferably, the abnormal data elimination on the normal distribution traffic data comprises:

[0153] calculating the mean and the standard deviation of the driving speed in all normal distribution traffic data respectively;

[0154] calculating a normal distribution abnormal interval according to the mean and the standard deviation;

[0155] determining whether the driving speed in the normal distribution traffic data is located in the normal distribution abnormal interval, and if so, regarding the normal distribution traffic data as abnormal data and eliminating it.

[0156] Preferably, the abnormal data elimination on the non-normal distribution traffic data comprises:

[0157] sorting the driving speed in the non-normal distribution traffic data in ascending order;

[0158] calculating four quantiles of the sorted driving speed;

[0159] According to the four quantiles, a non-normal distribution abnormal interval is calculated;

[0160] It is judged whether the driving speed in the non-normal distribution traffic data is located in the non-normal distribution abnormal interval. If the driving speed is located in the non-normal distribution abnormal interval, it is determined that the non-normal distribution traffic data is abnormal data, and the non-normal distribution traffic data is excluded.

[0161] Preferably, the traffic parameters include a steering wheel angle parameter, a headway parameter, and a post-intersection intrusion time parameter. According to the driving speed and the values of the traffic parameters in the normal traffic data, a relationship curve between the driving speed and the traffic parameters in each scene is obtained by regression analysis of the normal traffic data in each scene, including:

[0162] A loss function between the driving speed and the traffic parameters is established;

[0163] According to a preset quantile, the loss function is optimized by using the values of the driving speed and the steering wheel angle parameter in the normal traffic data in the scene. After optimization, a relationship curve between the driving speed and the steering wheel angle parameter in the scene is obtained.

[0164] According to a preset quantile, the loss function is optimized by using the values of the driving speed and the headway parameter in the normal traffic data in the scene. After optimization, a relationship curve between the driving speed and the headway parameter in the scene is obtained.

[0165] According to a preset quantile, the loss function is optimized by using the values of the driving speed and the post-intersection intrusion time parameter in the normal traffic data in the scene. After optimization, a relationship curve between the driving speed and the post-intersection intrusion time parameter in the scene is obtained.

[0166] Preferably, the standard value range of the traffic parameter of the unmanned mine truck in each scene is determined according to the relationship curve between the traffic parameter and the driving speed, including:

[0167] The standard value range of the steering wheel angle parameter of the unmanned mine truck in each scene is determined according to the relationship curve between the steering wheel angle parameter and the driving speed in each scene.

[0168] The standard value range of the headway parameter of the unmanned mine truck in each scene is determined according to the relationship curve between the headway parameter and the driving speed in each scene.

[0169] According to the relationship curve between the intersection post-invasion time parameter and the driving speed in each scene, a standard value range of the intersection post-invasion time parameter of the unmanned mine truck in the scene is determined.

[0170] Preferably, the event scene includes meeting, following and intersection conflict, and the data segmentation of the normal traffic data according to the event scene to obtain the normal traffic data in multiple scenes includes:

[0171] The normal traffic data with the event scene of meeting is selected from all the normal traffic data as the normal traffic data in the meeting scene;

[0172] The normal traffic data with the event scene of following is selected from all the normal traffic data as the normal traffic data in the following scene;

[0173] The normal traffic data with the event scene of intersection conflict is selected from all the normal traffic data as the normal traffic data in the intersection scene.

[0174] It can be understood that the technical scheme provided by the embodiment determines the standard value range of the traffic parameter of the unmanned mine truck according to a large amount of historical traffic data of the driver, and more accurately determines the standard value range of the traffic parameter of the unmanned mine truck, thereby effectively solving the problem of few systematic construction and determination methods of the traffic parameter of the mine truck in the prior art, formulating the human-like traffic rule of the unmanned mine truck, improving the transportation efficiency and safety in the mine area.

[0175] The application further provides a traffic parameter determination device of a vehicle, which comprises:

[0176] A memory having an executable program stored thereon;

[0177] A processor configured to execute the executable program in the memory to implement the steps of any one of the above methods.

[0178] In addition, the present application also provides a computer readable storage medium, the computer readable storage medium stores computer instructions, the computer instructions are used for making computer execute steps of any one of the above-mentioned methods. Wherein, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD) and the like; the storage medium can also include a combination of the above-mentioned kinds of memories.

[0179] It can be understood that the same or similar parts in the above-mentioned embodiments can be mutually referred to, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0180] It should be noted that in the description of the present application, the terms "first", "second" and the like are only used for descriptive purposes and should not be construed as indicating or implying relative importance. In addition, in the description of the present application, unless otherwise specified, the meaning of "a plurality of" is at least two.

[0181] Any process or method descriptions in flow charts or described elsewhere herein can be understood as representing code modules, segments, or portions of code which include one or more executable instructions for implementing specific logic functions (or steps) in the process, and the preferred embodiments of the present application include additional implementations in which the order of execution of the code modules, segments, or portions of code can be changed, including substantially simultaneously, or in reverse order, or in any other order, depending on the functionality involved, as will be understood by those skilled in the art.

[0182] It should be understood that parts of the present application can be realized in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be realized by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if realized in hardware, and as in another embodiment, it can be realized by any one or a combination of the following technologies known in the art: discrete logic circuit with logic gate circuit for implementing logic function on data signal, special integrated circuit with suitable combination logic gate circuit, programmable gate array (PGA), field programmable gate array (FPGA) and the like.

[0183] Those skilled in the art can understand that all or part of the steps of the method carried out by the above-mentioned embodiments can be instructed by a program to the relevant hardware, and the program can be stored in a computer readable storage medium. When the program is executed, it includes one of the steps of the method embodiment or a combination thereof.

[0184] In addition, each functional unit in each embodiment of the present application can be integrated into one processing module, or each unit can exist physically independently, or two or more units can be integrated into one module. The integrated module can be realized in the form of hardware or in the form of a software functional module. When the integrated module is realized in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

[0185] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0186] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily mean the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0187] Although the embodiments of the present application have been shown and described above, it can be understood that the above-mentioned embodiments are exemplary and cannot be understood as limiting the present application. Those skilled in the art can make changes, modifications, replacements and variations to the above-mentioned embodiments within the scope of the present application.

Claims

1. A method for determining the passage parameters of a vehicle, applied to unmanned mining trucks, characterized in that, include: Collect multiple historical passage data of mining trucks with drivers. The historical passage data includes event scenarios, driving speeds, and values ​​of multiple passage parameters, including steering wheel angle parameters, vehicle headway parameters, and intersection intrusion time parameters. The historical passage data is preprocessed to obtain normal passage data; Based on the event scenario, the normal passage data is segmented to obtain normal passage data under multiple scenarios; Based on the driving speed and various traffic parameters in the normal traffic data, regression analysis is performed on the normal traffic data for each scenario to obtain the relationship curves between multiple traffic parameters and the driving speed in that scenario. This includes: establishing a loss function between each traffic parameter and the driving speed; optimizing the loss function based on a preset quantile using the values ​​of driving speed and steering wheel angle parameters in the normal traffic data for that scenario, resulting in the relationship curve between the steering wheel angle parameter and the driving speed in that scenario; optimizing the loss function based on a preset quantile using the values ​​of driving speed and vehicle headway parameters in the normal traffic data for that scenario, resulting in the relationship curve between the vehicle headway parameter and the driving speed in that scenario; and optimizing the loss function based on a preset quantile using the values ​​of driving speed and intersection intrusion time parameters in the normal traffic data for that scenario, resulting in the relationship curve between the intersection intrusion time parameter and the driving speed in that scenario. Based on the relationship curve between the traffic parameters and the driving speed, the standard value range of the traffic parameters for each scenario of the unmanned mining truck is determined, including: determining the standard value range of the steering wheel angle parameter for the unmanned mining truck in each scenario based on the relationship curve between the steering wheel angle parameter and the driving speed; determining the standard value range of the front-end spacing parameter for the unmanned mining truck in each scenario based on the relationship curve between the front-end spacing parameter and the driving speed; and determining the standard value range of the intersection intrusion time parameter for the unmanned mining truck in each scenario based on the relationship curve between the intersection intrusion time parameter and the driving speed.

2. The method according to claim 1, characterized in that, The preprocessing of the historical passage data to obtain normal passage data includes: Determine whether the driving speed in the historical traffic data conforms to a normal distribution. If it conforms to a normal distribution, the historical traffic data is treated as normally distributed traffic data. If it does not conform to a normal distribution, the historical traffic data is treated as non-normally distributed traffic data. Abnormal data is removed from both the normally distributed and non-normally distributed traffic data to obtain normal traffic data.

3. The method according to claim 2, characterized in that, The process of removing outlier data from the normally distributed traffic data includes: Calculate the mean and standard deviation of driving speeds for all normally distributed traffic data. Based on the mean and standard deviation, the abnormal intervals of the normal distribution are calculated; Determine whether the driving speed in the normally distributed traffic data is within the abnormal range of the normally distributed data. If it is within the abnormal range of the normally distributed data, determine that the normally distributed traffic data is abnormal data and remove it.

4. The method according to claim 2, characterized in that, The process of removing outlier data from the non-normally distributed traffic data includes: Sort the driving speeds in the non-normally distributed traffic data in ascending order; Calculate the four quantiles of the sorted driving speeds; Based on the four quantiles, the non-normal distribution anomaly intervals are calculated; Determine whether the driving speed in the non-normally distributed traffic data is within the non-normally distributed abnormal interval. If it is within the non-normally distributed abnormal interval, determine that the non-normally distributed traffic data is abnormal data and remove it.

5. The method according to claim 1, characterized in that, The event scenarios include meeting oncoming traffic, following other vehicles, and intersection conflicts. Based on these event scenarios, the normal traffic data is segmented to obtain normal traffic data for multiple scenarios, including: Select the event scenario as the normal passage data of meeting oncoming traffic from all normal passage data, and use it as the normal passage data under the meeting oncoming traffic scenario; Select normal traffic data with the event scenario of car-following from all normal traffic data, and use it as normal traffic data under the car-following scenario; Select normal traffic data with intersection conflicts from all normal traffic data scenarios, and use them as normal traffic data for intersection scenarios.

6. A control method for mining trucks, applied to unmanned mining trucks, characterized in that, include: Using the method for determining the vehicle's passage parameters as described in any one of claims 1-5, the standard range of values ​​for the passage parameters of the unmanned mining truck in various scenarios is determined. The passage of the unmanned mining truck is controlled according to the standard value range.

7. A device for determining the passage parameters of a vehicle, applied to an unmanned mining truck, characterized in that, The device includes: The data acquisition module is used to collect multiple historical passage data of mining trucks with a driver. The historical passage data includes event scenarios, driving speeds, and values ​​of multiple passage parameters, including steering wheel angle parameters, vehicle headway parameters, and intersection intrusion time parameters. The data preprocessing module is used to preprocess the historical passage data to obtain normal passage data; The data segmentation module is used to segment the normal passage data according to the event scenario to obtain normal passage data under multiple scenarios; The regression analysis module is used to perform regression analysis on the normal traffic data under various scenarios based on the driving speed and the values ​​of various traffic parameters in the normal traffic data, to obtain the relationship curves between multiple traffic parameters and the driving speed under the scenario; specifically, it is used to establish a loss function between each traffic parameter and the driving speed; based on a preset quantile, it optimizes the loss function using the values ​​of driving speed and steering wheel angle parameters in the normal traffic data under the scenario, and after optimization, it obtains the relationship curve between the steering wheel angle parameter and the driving speed under the scenario; based on a preset quantile, it optimizes the loss function using the values ​​of driving speed and vehicle headway parameters in the normal traffic data under the scenario, and after optimization, it obtains the relationship curve between the vehicle headway parameter and the driving speed under the scenario; based on a preset quantile, it optimizes the loss function using the values ​​of driving speed and intersection intrusion time parameters in the normal traffic data under the scenario, and after optimization, it obtains the relationship curve between the intersection intrusion time parameter and the driving speed under the scenario; The parameter standard determination module is used to determine the standard value range of the traffic parameters for each scenario of the unmanned mining truck based on the relationship curve between the traffic parameters and the driving speed. Specifically, it is used to determine the standard value range of the steering wheel angle parameter for the unmanned mining truck in each scenario based on the relationship curve between the steering wheel angle parameter and the driving speed; to determine the standard value range of the front-end spacing parameter for the unmanned mining truck in each scenario based on the relationship curve between the front-end spacing parameter and the driving speed; and to determine the standard value range of the intersection intrusion time parameter for the unmanned mining truck in each scenario based on the relationship curve between the intersection intrusion time parameter and the driving speed.

8. A device for determining the passage parameters of a vehicle, applied to unmanned mining trucks, characterized in that, include: Memory, on which executable programs are stored; A processor for executing the executable program in the memory to implement the steps of the method according to any one of claims 1 to 5.

Images