Vehicle safety control method based on historical driving data and related products

Abstract

The invention provides a vehicle safety control method based on historical driving data and a related product. The method comprises the following steps: acquiring historical driving data of a current vehicle in a set time period, and clustering the historical driving data to obtain a first safe driving model; acquiring newly-added driving data of the current vehicle after the set time period, and clustering the newly-added driving data under the condition that the data volume of the newly-added driving data is greater than a first set data volume threshold value to obtain a second safe driving model; and detecting the current driving environment, and controlling the current vehicle to safely drive according to the first safe driving model and the second safe driving model under the condition that the driving risk index is greater than a set risk threshold value. According to the technical scheme, the safety of vehicle driving can be improved, and the purpose of improving user experience is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle safety control, and in particular to a vehicle safety control method based on historical driving data and related products. Background Art

[0002] With the continuous improvement of people's living standards and the rapid development of science and technology, cars have become a common industrial product in many families. At the same time, the traffic volume on the road is also increasing, and the problem of road driving safety is becoming more and more serious.

[0003] Intelligent assisted driving technology can formulate the best driving strategy according to the current driving environment of the vehicle to help the driver cope with the current driving environment and improve the safety of the car. The intelligent assisted driving technology in the prior art can obtain the driver's response measures to various driving environments based on the vehicle's historical driving data to build a data model, and generate a driving strategy for the current driving environment based on the data model. As the driver's driving skills improve and the driving season and driving environment change, the driver's response measures to different driving environments will also change, so the data model is regularly updated according to the current vehicle's driving data.

[0004] However, updating the data model takes a long time. If the update is performed while driving, the assisted driving function cannot be provided during the update process, which affects the driving safety of the vehicle. Summary of the invention

[0005] One purpose of the present invention is to provide a vehicle safety control method and related products based on historical driving data, so as to improve the safety of vehicle driving and achieve the purpose of improving user experience.

[0006] Specifically, in a first aspect, the present invention provides a vehicle safety control method based on historical driving data, comprising:

[0007] The steps of constructing a safe driving model for the current vehicle, and controlling the safe driving of the current vehicle according to the safe driving model;

[0008] The step of constructing a safe driving model for the current vehicle includes:

[0009] Acquire historical driving data of the current vehicle in a set time period, and acquire optimal driving modes for multiple historical driving environments from the historical driving data;

[0010] Obtaining adjacent driving environments of each of the historical driving environments according to the similarities between the historical driving environments, and selecting a plurality of reference driving environments from each of the historical driving environments according to the number of the adjacent driving environments;

[0011] Clustering the similarities between the reference driving environments to obtain a plurality of historical environment clusters, and calculating the cluster center of each of the historical environment clusters to obtain a first safe driving model;

[0012] Acquire new driving data of the current vehicle after the set time period, and when the amount of the new driving data is greater than a first set data amount threshold, acquire optimal driving modes for multiple new driving environments from the new driving data;

[0013] Clustering each of the newly added driving environments to obtain a plurality of newly added environment clusters, and calculating a cluster center of each of the newly added environment clusters to obtain a second safe driving model;

[0014] The step of controlling the current vehicle to drive safely according to the safe driving model comprises:

[0015] Detecting the current driving environment, and when the driving risk index is greater than a set risk threshold, selecting a target historical environment cluster from each of the historical environment clusters according to the similarity between the cluster center and the current driving environment, and taking the reference driving environment with the greatest similarity to the current driving environment in the target historical environment cluster as the target historical driving environment;

[0016] Selecting a new target cluster from each of the new environment clusters according to the similarity between the cluster center and the current driving environment, and taking multiple new driving environments in the new target clusters with the greatest similarity to the current driving environment as new target driving environments;

[0017] The neighboring driving environments of the target historical driving environment are integrated with the optimal driving mode of the newly added target driving environment to generate a safe driving strategy, and the current vehicle is controlled to drive safely according to the safe driving strategy.

[0018] Further, after the step of acquiring the newly added driving data of the current vehicle after the set time period, the method further includes:

[0019] Acquiring an update condition of the safe driving model according to the historical driving data;

[0020] Determining whether the amount of the newly added driving data is greater than a second set data amount threshold;

[0021] If so, detecting whether the current vehicle meets the update condition;

[0022] If it is less than, return to the step of obtaining the historical driving data of the set time period.

[0023] Furthermore, the step of calculating the cluster center of each of the historical environment clusters includes:

[0024] The fitness values ​​of the reference driving environments as the cluster centers of the corresponding historical environment clusters are calculated respectively, and the reference driving environments with the largest fitness values ​​in the historical environment clusters are taken as the cluster centers of the corresponding historical environment clusters.

[0025] Furthermore, the step of clustering each of the newly added driving environments to obtain a plurality of newly added environment clusters includes:

[0026] A first preset number of the newly added environment clusters is obtained, and a k-means clustering algorithm is used to cluster each of the newly added driving environments according to the first preset number to obtain a plurality of the newly added environment clusters.

[0027] Furthermore, the step of selecting a new target cluster from each of the new environment clusters according to the similarity between the cluster center and the current driving environment includes:

[0028] A second preset number is obtained, and a second preset number of newly added environment clusters whose cluster centers have the greatest similarity with the current driving environment are all used as the newly added target clusters.

[0029] In a second aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above vehicle safety control methods are implemented.

[0030] In a third aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any of the above vehicle safety control methods.

[0031] In the technical solution of the present invention, a first safe driving model can be constructed based on the historical driving data of the current vehicle in a set time period, and when the amount of the newly added driving data of the current vehicle is greater than the first set data amount threshold, a second safe driving model can be constructed based on the newly added driving data; during the driving process of the current vehicle, the target historical driving environment most similar to the current driving environment is obtained according to the first safe driving model, and the newly added target driving environment most similar to the current driving environment is obtained according to the second safe driving model, and then the neighboring driving environments of the target historical driving environment are merged with the optimal driving mode of the newly added target driving environment to generate a safe driving strategy, and the current vehicle is controlled to drive safely according to the safe driving strategy. Because in the technical solution of the present invention, after obtaining the newly added driving data of the current vehicle, the newly added driving data is not directly added to the first safe driving model, but the second safe driving model is constructed based on the newly added driving data, and the construction of the second safe driving model only requires clustering of the newly added driving data, which requires less time, so it can avoid the inability to provide safe driving services due to the update of the safe driving model, and the newly added driving data can also be used to improve the reliability of safe driving.

[0032] Based on the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art will become more aware of the above and other objects, advantages and features of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Hereinafter, some specific embodiments of the present invention will be described in detail in an exemplary and non-limiting manner with reference to the accompanying drawings. The same reference numerals in the accompanying drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the accompanying drawings:

[0034] Figure 1 is a schematic flow chart of a vehicle safety control method based on historical driving data according to an embodiment of the present invention;

[0035] Figure 2 is a schematic flow chart of constructing a safe driving model of a current vehicle according to an embodiment of the present invention;

[0036] Figure 3 is a schematic flow chart of controlling the safe driving of the current vehicle according to the constructed safe driving model according to one embodiment of the present invention;

[0037] Figure 4 is a schematic flow chart of constructing a safe driving model of a current vehicle according to another embodiment of the present invention;

[0038] Figure 5is a schematic flow chart of calculating the cluster center of each historical environment cluster according to an embodiment of the present invention;

[0039] Figure 6 is a schematic flow chart of clustering according to the similarity between each newly added driving environment according to an embodiment of the present invention;

[0040] Figure 7 is a schematic flow chart of updating each initial cluster according to an embodiment of the present invention;

[0041] Figure 8 is a schematic diagram of a computer program product according to an embodiment of the present invention; and

[0042] Fig. 9 is a schematic diagram of a computer-readable storage medium according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] Refer to the following Figures 1 to 9 To describe a vehicle safety control method and related products based on historical driving data of an embodiment of the present invention. In the description of this embodiment, it should be understood that the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features, that is, include one or more of the features. In the description of the present invention, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. When a feature "includes or contains" one or some of the features it covers, unless otherwise specifically described, this indicates that other features are not excluded and may further include other features.

[0044] See also Figure 1 , Figure 1 It is a schematic flow chart of a vehicle safety control method based on historical driving data according to an embodiment of the present invention, which method can not only improve the safety of current vehicle driving, but also improve the timeliness and reliability of the current vehicle safe driving response.

[0045] like Figure 1 As shown, the vehicle safety control method based on historical driving data of the present invention may generally include:

[0046] Step S101: constructing a safe driving model for the current vehicle;

[0047] Step S102: Control the current vehicle to drive safely according to the constructed safe driving model.

[0048] In this embodiment, the safe driving model constructed in the above step S101 includes a first safe driving model and a second safe driving model, and the method for constructing the safe driving model is as follows: Figure 2 As shown, the following steps are included:

[0049] Step S111: obtaining historical driving data of the current vehicle in a set time period, and obtaining multiple historical driving environments from the historical driving data;

[0050] Step S112: constructing a data vector of each historical driving data, and calculating the similarity between each historical driving environment according to the data vector of each historical driving data;

[0051] Step S113: obtaining neighboring driving environments of each historical driving data according to the similarity of each historical driving data, selecting reference driving environments from each historical driving data according to the number of neighboring driving environments, and obtaining the optimal driving mode of each reference driving environment;

[0052] Step S114: clustering is performed according to the similarity between the reference driving environments to obtain a plurality of historical environment clusters, and the cluster center of each historical environment cluster is calculated to obtain a first safe driving model;

[0053] Step S115: acquiring newly added driving data, and determining whether the amount of the newly added driving data is greater than a first set data amount threshold;

[0054] If yes, execute step S116;

[0055] Step S116: acquiring multiple newly added driving environments and the optimal driving mode for each newly added driving environment from the newly added driving data;

[0056] Step S117: constructing data vectors of each newly added driving environment, and calculating similarities between each newly added driving environment according to the data vectors of each newly added driving environment;

[0057] Step S118: clustering is performed according to the similarities between the newly added driving environments to obtain a plurality of newly added driving environment clusters, and the cluster center of each newly added driving environment cluster is calculated to obtain a second safe driving model.

[0058] In the above step S111, the historical driving environment obtained according to the historical driving data of the current vehicle includes a traffic jam environment, a road maintenance environment, a vehicle accident environment, a road failure environment, a waiting street light environment, and a road turning environment.

[0059] In the above step S112, taking one of the historical driving environments as an example, the method for constructing the data vector of the historical driving environment includes: obtaining data of the historical driving environment in multiple preset dimensions, and constructing the data vector of the historical driving environment according to the data of each preset dimension.

[0060] In this embodiment, the preset dimensions may include the distance between the current vehicle and the vehicle or obstacle ahead, the moving speed of the vehicle or obstacle ahead, the distance between the current vehicle and the side vehicle or obstacle, the moving speed of the side vehicle or obstacle, the ambient light intensity, the weather severity index, and the road smoothness. In the process of constructing the data vector of one of the historical driving environments, the data values ​​of each preset dimension of the historical driving environment may be normalized to obtain the data vector of the historical driving environment.

[0061] Taking one of the preset dimensions of the above driving history environment as an example, the method for normalizing the data value of the preset dimension includes:

[0062] Get the data value of the preset dimension as T, the maximum value of the preset dimension as T max , then the data value of the preset dimension after the normalization process is T'=T / T max .

[0063] Taking two of the driving environments as an example, the method for calculating the similarity of the two driving environments includes:

[0064] Assume that both driving environments have n preset dimensions, and in the data vector of one driving environment, the index of the i-th preset dimension is P 1,i , in another driving environment data vector, the index of the i-th preset dimension is P 2,i , then the similarity between the two driving environments for:

[0065]

[0066] In the above step S113, taking one of the historical driving environments as an example, the method for obtaining the neighboring driving environment of the historical driving environment includes: taking the historical driving environment whose similarity with the historical driving environment is greater than a preset similarity threshold as the neighboring driving environment of the historical driving environment.

[0067] In this embodiment, the method of selecting a reference driving environment from each historical driving data according to the number of neighboring driving environments includes:

[0068] The historical driving environments are arranged in order of the number of neighboring driving environments from large to small, and the historical driving environments with the first set ratio in the arrangement are used as reference driving environments. For example, assuming that there are N historical driving environments and the first set ratio is 80%, then the N×80% historical driving environments with the most neighboring driving environments are used as reference driving environments.

[0069] Taking one of the reference driving environments as an example, the method for obtaining the optimal driving mode for the reference driving environment includes:

[0070] Firstly, a plurality of driving modes corresponding to the reference driving environment are obtained from each historical driving data, then the risk index of each driving mode is calculated, and finally the optimal driving mode of the reference driving environment is obtained according to each risk index.

[0071] Taking one of the driving modes in a reference driving environment as an example, the method for obtaining the risk index of the driving mode includes:

[0072] First, determine whether a collision with surrounding vehicles or obstacles occurs during the driving mode. If a collision occurs, the risk index of the driving mode is set to 0;

[0073] If no collision occurs, the number of turns and the turning angle of each turn in the driving mode are obtained, and a first risk value of the driving mode is calculated according to the number of turns and each turning angle, including:

[0074] First, obtain the preset correspondence between the turn number interval and the risk value, then obtain the turn number interval in which the turn number in the driving mode is located, and obtain the preset risk value k1 corresponding to the number interval according to the above correspondence; let the turning angle of the i-th turn in the driving mode be θ i , then the first risk value d1 of this driving mode is:

[0075]

[0076] Wherein α1 is the first steering matching coefficient, α2 is the second steering matching coefficient, and a is the steering angle correction coefficient.

[0077] Periodically acquiring the driving speed, the driving speed of the front vehicle or obstacle, and the distance between the front vehicle or obstacle and the front vehicle or obstacle in the driving mode to obtain a second risk value of the driving mode, including:

[0078] Assume the i-th driving speed obtained is v i , the distance between the vehicle or obstacle in front is l i , the driving speed is v i , the speed of the vehicle or obstacle in front is v' i, then the second risk value of this driving mode is:

[0079]

[0080] Among them, β1 is the motion correction coefficient.

[0081] The number of times that the distance between the vehicle and the side vehicle or obstacle is less than the set vehicle distance threshold in the driving mode is periodically obtained as sid, and the third risk value d3 of the driving mode is obtained according to the number, including:

[0082]

[0083] Among them, β2 is the side vehicle distance correction coefficient, and β3 is the correction constant.

[0084] Finally, according to the first risk value, the second risk value and the third risk value, the risk index p of the driving mode is calculated, including:

[0085] p=ω1d1+ω2d2+ω3d3

[0086] Among them, ω1, ω2, ω3 are the first safety factor, the second safety factor and the third safety factor respectively, and ω1+ω2+ω3=1.

[0087] After the risk index of each driving mode in one of the driving environments is obtained, the driving mode with the smallest risk index is used as the optimal driving mode for the reference driving environment.

[0088] In the above step S114, a preset clustering algorithm can be used to cluster the reference driving environments according to the similarity between the reference driving environments, so as to obtain multiple historical environment clusters. Then, according to the similarity between the reference driving environments in each historical environment cluster, the cluster center of each historical environment cluster can be calculated respectively.

[0089] In this embodiment, after obtaining the cluster center of each historical environment cluster, the obtained first safe driving model includes multiple historical environment clusters, the cluster center of each historical environment cluster, the optimal driving mode of each reference driving environment and the neighboring driving environment.

[0090] In the above step S115, if the amount of newly added driving data of the current vehicle is less than the first set data amount threshold, it can be considered that the newly added driving data is too small to provide a reference for the safe driving of the current vehicle. Therefore, when the amount of newly added driving data of the current vehicle is greater than the first set data amount threshold, a second safe driving model is constructed based on the newly added driving data.

[0091] In the above step S116, the newly added driving environment obtained from the newly added driving data includes a traffic jam environment, a road maintenance environment, a vehicle accident environment, a road failure environment, a waiting streetlight environment, and a road turning environment. In this embodiment, the method for obtaining the optimal driving mode for each newly added driving environment is the same as the method for obtaining the optimal driving mode for each reference driving environment described above, and no further description is given here to avoid redundancy.

[0092] In the above step S117, the method of constructing the data vectors of each newly added driving environment and the method of calculating the similarity between each newly added driving environment based on the data vectors of each newly added driving environment are respectively the same as the method of constructing the data vectors of each historical driving environment and the method of calculating the similarity between each historical driving environment in the above text. In order to avoid redundancy, they are not described here.

[0093] In the above step S118, a preset clustering algorithm can be used to cluster the newly added driving environments according to the similarity between the newly added driving environments, so as to obtain a plurality of newly added environment clusters. Then, according to the similarity between the newly added driving environments in each newly added environment cluster, the cluster center of each newly added environment cluster can be calculated respectively.

[0094] In this embodiment, the obtained second safe driving model includes multiple newly added environment clusters and the cluster center of each newly added environment cluster.

[0095] In some embodiments of the present invention, the method of controlling the safe driving of the current vehicle according to the constructed safe driving model in step S102 is as follows: Figure 3 As shown, the following steps are included:

[0096] Step S121: Detect the current driving environment, calculate the driving risk index of the current driving environment, and determine whether the driving risk index is greater than a set risk threshold;

[0097] If yes, execute step S122;

[0098] Step S122: calculating the similarity between the cluster center of each historical environment cluster and the current driving environment, and taking the historical environment cluster whose cluster center has the greatest similarity with the current driving environment as the target historical environment cluster;

[0099] Step S123: calculating the similarity between each reference driving environment in the target historical environment cluster and the current driving environment, and taking the reference historical driving environment with the greatest similarity as the target historical driving environment;

[0100] Step S124: calculating the similarity between the cluster center of each newly added environment cluster and the current driving environment, and taking the newly added environment cluster whose cluster center has the greatest similarity with the current driving environment as the newly added target cluster;

[0101] Step S125: calculating the similarity between each newly added driving environment in the newly added target cluster and the current driving environment, and taking a first set number of newly added driving environments with the greatest similarity as newly added target driving environments;

[0102] Step S126: The optimal driving modes of the neighboring driving environments of the target historical driving environment and the optimal driving modes of each newly added target driving environment are integrated to obtain a safe driving method, and the current vehicle driving is controlled according to the safe driving method.

[0103] In the above step S121, the current driving environment can be obtained by using the driving environment detection device set on the current vehicle and the road traffic flow information at the location. Then, the driving risk evaluation model is used to obtain the driving risk index of the current driving environment according to the traffic flow of the current driving environment, the surrounding vehicle information and the obstacle information.

[0104] For example, assuming that the current lane width is L, the speed of the vehicle ahead is V, the distance to the vehicle ahead is D, and the traffic volume of the lanes on both sides is h1 and h2 respectively, then when the lane width L is greater than the set width threshold, the driving risk index R of the current driving environment is:

[0105]

[0106] When the lane width L is greater than the set width threshold, the driving risk index R of the current driving environment is

[0107]

[0108] Among them, ε1, ε2, and ε3 are the first risk matching coefficient, the second risk matching coefficient, and the third risk matching coefficient, respectively.

[0109] If the driving risk index of the current driving environment is greater than the set risk threshold, it can be considered that the current driving environment has a high risk of causing a traffic accident, and a safe driving method matching the current driving environment needs to be obtained to improve the safety of the current vehicle driving.

[0110] In the above step S122, the method of constructing the data vector of the current driving environment is the same as the method of constructing the data vectors of each historical driving environment in the above text; after constructing the data vector of the current driving environment, the similarity between the cluster center of each historical environment cluster cluster and the current driving environment can be calculated respectively based on the data vector and the data vector of the cluster center of each historical environment cluster cluster. The calculation method of this similarity is the same as the calculation method of the similarity between each historical driving environment in the above text, and it will not be introduced here to avoid repetition.

[0111] The method for calculating the similarity between each reference driving environment in the target historical environment clustering cluster and the current driving environment in the above step S123, the method for calculating the similarity between the cluster center of each newly added environment clustering cluster and the current driving environment in the above step S124, and the method for calculating the similarity between each newly added driving environment in the newly added target clustering cluster and the current driving environment in the above step S125 are all the same as the calculation method for calculating the similarity between each historical driving environment mentioned above, and will not be introduced here to avoid repetition.

[0112] In the above step S126, each neighboring driving environment of the target historical driving environment and each newly added target driving environment can be used as a standby driving environment, and then the optimal driving mode of each standby driving environment is integrated to obtain a safe driving method for the current driving environment.

[0113] In this embodiment, the method of fusing the optimal driving modes of each waiting driving environment includes:

[0114] Assume that there are M standby driving environments, take the ith standby driving environment as an example, and set the ambient light intensity of the ith standby driving environment to be light i , the ambient light intensity of the current driving environment is light0, then the similarity of the ambient light intensity between the ith waiting driving environment and the current driving environment is sli i for:

[0115]

[0116] The weather severity index of the i-th standby driving environment is weather i , the weather severity index of the current driving environment is weather0, then the weather severity index similarity swea between the ith standby driving environment and the current driving environment i for:

[0117]

[0118] The road smoothness of the i-th driving environment to be used is smooth iIf the road smoothness of the current driving environment is smooth0, then the road smoothness similarity sm between the i-th待用 driving environment and the current driving environment i is

[0119]

[0120] The environmental similarity senv between the i-th待用 driving environment and the current driving environment i is

[0121] senv i = sli i × swea i × sm i

[0122] Assume that in the current driving environment, the distance between the current vehicle and the vehicle or obstacle in front is l0. Then the similarity sfrobs of the front obstacle between the i-th待用 driving environment and the current driving environment i is as follows:

[0123]

[0124] Assume that in the current driving environment, the moving speed of the vehicle or obstacle in front is v0. Then the speed similarity sv between the i-th待用 driving environment and the current driving environment i is as follows:

[0125]

[0126] Assume that in the current driving environment, the distance between the current vehicle and the vehicle or obstacle on the left is lef0, and the distance between the current vehicle and the vehicle or obstacle on the right is rig0. In the i-th待用 driving environment, the distance between the current vehicle and the vehicle or obstacle on the left is lef i , and the distance between the current vehicle and the vehicle or obstacle on the right is rig i . Then the similarity sidobs of the side obstacle between the i-th待用 driving environment and the current driving environment i is

[0127]

[0128] Among the optimal driving methods of each待用 driving environment obtained, the number of optimal driving methods with steering at the t-th moment and the same steering direction is m. If m < M × μ, then in the safe driving method of the previous driving environment, at the t-th moment, control the current vehicle not to steer, and the driving speed V t is

[0129]

[0130] Among them, μ is the second setting ratio, v i,t is the driving speed of the current vehicle at the tth moment in the optimal driving mode of the i-th standby driving environment.

[0131] If m ≥ M × μ, then the optimal driving mode of the m waiting driving environments is obtained. In the safe driving mode of the previous driving environment, the current vehicle is controlled to turn at the tth moment, and the direction of the turning is the turning direction of the waiting driving mode. The driving speed V t and steering angle They are:

[0132]

[0133] Among them, senv′ i , sv′ i , sfrobsi′ i and sidobs′ i are respectively the environment similarity, speed similarity, front obstacle similarity and side obstacle similarity between the ith standby driving environment and the current driving environment among the m standby driving environments obtained, v′ i,t and θ′ i,t are respectively the moving speed and turning angle of the current vehicle at the tth moment in the optimal driving modes of the m standby driving environments obtained.

[0134] After obtaining the safe driving mode of the front driving environment, the driving speed and turning angle of the current vehicle at each moment are controlled according to the safe driving mode to improve the driving safety of the current vehicle.

[0135] According to the above content, the technical solution of this embodiment can construct a first safe driving model based on the historical driving data of the current vehicle in a set time period, and construct a second safe driving model based on the newly added driving data when the amount of data of the new driving data of the current vehicle is greater than the first set data amount threshold; in the process of driving of the current vehicle, the target historical driving environment most similar to the current driving environment is obtained according to the first safe driving model, and the newly added target driving environment most similar to the current driving environment is obtained according to the second safe driving model, and then the neighboring driving environments of the target historical driving environment are merged with the optimal driving mode of the newly added target driving environment to generate a safe driving strategy, and the current vehicle is controlled to drive safely according to the safe driving strategy. Since in the technical solution of this embodiment, after obtaining the newly added driving data of the current vehicle, the newly added driving data is not directly added to the first safe driving model, but the second safe driving model is constructed based on the newly added driving data, and the construction of the second safe driving model only requires clustering of the newly added driving data, which requires less time, so it can avoid the inability to provide safe driving services due to the update of the safe driving model, and the newly added driving data can also be used to improve the reliability of safe driving.

[0136] In some embodiments of the present invention, Figure 4 As shown, after the above step S115 acquires the newly added driving data, it also includes:

[0137] Step S131: obtaining an update condition of a safe driving model of the current vehicle according to historical driving data of the current vehicle;

[0138] Step S132: determining whether the amount of the newly added driving data of the current vehicle is greater than a second set data amount threshold;

[0139] If yes, execute step S133;

[0140] Step S133: Detect whether the current vehicle meets the above update conditions;

[0141] If satisfied, return to step S111.

[0142] In the above step S131, the usage pattern of the current vehicle can be obtained based on the historical driving data of the current vehicle to obtain the time period with the least data processing of the current vehicle, and this time period is used as the time period for model update. The interval time between reaching this time period and the last time a safe driving model was generated is greater than the set interval time, which is used as the update condition of the safe driving model.

[0143] For example, assuming that the usage time period of the current vehicle is obtained based on the usage record data of the current vehicle, which is concentrated between 7:00 and 9:00 in the morning and between 6:00 and 10:00 in the afternoon on weekdays, and the number of times the vehicle is used in other time periods accounts for a smaller proportion. Therefore, a specific time period can be selected outside the time periods of 7:00 to 9:00 in the morning and 6:00 to 10:00 in the afternoon on weekdays, and reaching the specific time period can be used as an update condition for the safety driving model of the current vehicle.

[0144] Therefore, after determining in the above step S132 that the amount of newly added driving data of the current vehicle is greater than the second set data amount threshold, if entering the selected specific time period, the update conditions of the safety driving model of the current vehicle are met.

[0145] Through the technical solution of this embodiment, the update conditions of the safe driving model can be determined based on the historical driving data of the current vehicle, so as to avoid the problem of being unable to perform safety control due to the update of the safe driving model during the driving of the current vehicle.

[0146] In some embodiments of the present invention, the method for calculating the cluster center of each historical environment cluster in step S114 is as follows: Figure 5 As shown, the following steps are included:

[0147] Step S321: Calculate the fitness of each reference driving environment as the cluster center of the corresponding historical environment cluster respectively;

[0148] Step S322: taking the reference driving environment with the greatest fitness in each historical environment cluster as the cluster center of the corresponding historical environment cluster.

[0149] Taking one of the historical environment clusters as an example, the method for obtaining the cluster center of the historical environment cluster includes:

[0150] A reference driving environment in the historical environment cluster is randomly obtained, and then the similarity between the reference driving environment and other reference driving environments in the historical environment cluster is calculated, and the average value of each similarity is used as the fitness of the historical reference driving environment.

[0151] Through the technical solution of this embodiment, the reference driving environment with the highest fitness in each historical environment clustering cluster can be used as the corresponding clustering center, so that each clustering center is an element in the corresponding historical environment clustering cluster, so that in the process of controlling the safe driving of the current vehicle according to the safe driving model, the target historical clustering cluster can be accurately and quickly obtained, thereby further improving the response speed of the current vehicle's safe driving.

[0152] In some embodiments of the present invention, the clustering algorithm used in the above step S114 for clustering according to the similarity between the reference driving environments, and the clustering algorithm used in the above step S118 for clustering according to the similarity between the newly added driving environments, are both k-means clustering algorithms.

[0153] For example, in the above step S118, clustering is performed according to the similarity between each newly added driving environment, specifically as follows: Figure 6 As shown, the following steps are included:

[0154] Step S301: obtaining a first preset number of newly added environment clusters, then randomly selecting a newly added driving environment from each newly added driving environment as a first initial cluster center, and moving the first initial cluster center to the cluster center set;

[0155] Step S302: determining whether the number of initial cluster centers in the cluster center set is a first preset number;

[0156] If not, execute step S303; if yes, execute step S304;

[0157] Step S303: Calculate the similarity between the remaining newly added driving environments and the initial cluster centers in the cluster center set, obtain a new initial cluster center according to the similarity, move the new initial cluster center to the cluster center set, and then return to step S302;

[0158] Step S304: clustering according to the similarities between the remaining newly added driving environments and the initial cluster centers to obtain a first preset number of initial cluster clusters;

[0159] Step S305: updating each initial cluster to obtain a first preset number of newly added environment clusters.

[0160] In the above step S303, taking one of the newly added driving environments as an example, after calculating the similarity between the newly added driving environment and each initial cluster center in the cluster center set, the average value of each similarity is used as the average similarity of the newly added driving environment. After obtaining the average similarity of each newly added driving environment, the newly added driving environment with the largest average phase velocity is used as the new initial cluster center.

[0161] In the above step S305, the method for updating each initial cluster is as follows: Figure 7 As shown, the following steps are included:

[0162] Step S311: calculating new cluster centers of each initial cluster, and re-clustering each newly added driving environment according to each new cluster center to obtain a new cluster;

[0163] Step S312: determining whether the update of the initial clustering cluster satisfies a preset termination condition;

[0164] If not, return to step S311; if satisfied, execute step S313;

[0165] Step S313: taking the cluster obtained by the last update as the newly added environment cluster.

[0166] In the above step S312, the preset termination condition includes that the number of updates to the initial clusters reaches a preset update number threshold, or the cluster centers of the clusters obtained twice in succession are the same.

[0167] In this embodiment, a k-means clustering algorithm can be used to cluster each reference driving environment into multiple historical environment clustering clusters, and cluster each new driving environment into multiple new environment clustering clusters, so as to improve the controllability and work efficiency of building a safe driving model.

[0168] In some embodiments of the present invention, the method of using the newly added environment cluster whose cluster center has the greatest similarity with the current driving environment as the newly added target cluster in step S124 comprises the following steps:

[0169] A second set number is obtained, and a second set number of newly added environment clusters whose cluster centers have the greatest similarity with the current driving environment are all used as newly added target clusters.

[0170] In this embodiment, the value of the second preset number can be 2 or other integers greater than 1. If there is only one newly added target cluster, there may be a problem that the target historical driving environment obtained is inaccurate due to the edge effect. The edge effect refers to the edge part of the first newly added environment cluster that exists in the reference driving environment with the greatest similarity to the current driving environment, but the similarity between the cluster center of the second newly added environment cluster and the current driving environment is greater than the similarity between the cluster center of the first newly added environment cluster and the current driving environment. Therefore, the target newly added driving environment will be searched in the second newly added environment cluster, but not in the first newly added environment cluster, resulting in inaccurate queried target historical driving environment, affecting the reliability of the obtained safe driving method.

[0171] Through the technical solution of this embodiment, the second preset number of new environmental clusters whose cluster centers have the greatest similarity with the current driving environment can be used as target new clusters to avoid the problem of affecting the safe driving mode of the current driving environment due to edge effects.

[0172] The flow chart provided by the present embodiment is not intended to indicate that the operation of the method will be performed in any particular order, or that all operations of the method are included in all every case. In addition, each of the above methods may include additional operations. Within the scope of the technical thinking provided by the present embodiment method, additional changes may be made to the above method.

[0173] It should be understood that in some embodiments, each part can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system.

[0174] This embodiment also provides a computer program product 10 and a computer readable storage medium 20 . Figure 8 is a schematic diagram of a computer program product 10 according to an embodiment of the present invention, Fig. 9 is a schematic diagram of a computer-readable storage medium 20 according to an embodiment of the present invention. The computer program product 10 includes a computer program 11, which implements the steps of any of the above-mentioned vehicle safety control methods based on historical driving data when the computer program 11 is executed by the processor 32. The computer-readable storage medium 20 stores the above-mentioned computer program 11, which implements the steps of any of the above-mentioned vehicle safety control methods based on historical driving data when the computer program 11 is executed by the processor 32. The computer device 30 may include a memory 31, a processor 32, and a computer program 11 stored in the memory 31 and running on the processor 32.

[0175] The computer program 11 for performing the operation of the present invention may be an assembly instruction, an instruction set architecture (ISA) instruction, a machine instruction, a machine-related instruction, a microcode, a firmware instruction, a state setting data, a configuration data of an integrated circuit, or a source code or an object code written in any combination of one or more programming languages ​​and process programming languages. The computer program 11 may be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer may be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider via the Internet). In some embodiments, in order to perform various aspects of the present invention, an electronic circuit including, for example, a programmable logic circuit, a field programmable gate array (FPGA) or a programmable logic array (PLA) may execute computer-readable program instructions by utilizing the state information of the computer-readable program instructions to personalize the electronic circuit.

[0176] In the description of this embodiment, the computer program product 10 is a related product including the computer program 11 .

[0177] For the purpose of the description of the present embodiment, the computer readable storage medium 20 is a tangible device capable of retaining and storing the computer program 11, which can be any device that can contain, store, communicate, propagate or use the computer program 11 for an instruction execution system, device or apparatus or in conjunction with these instruction execution systems, devices or apparatuses. More specific examples (a non-exhaustive list) of the computer readable storage medium 20 include the following: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, and any suitable combination of the above.

[0178] At this point, those skilled in the art should recognize that, although multiple exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications that conform to the principles of the present invention can still be directly determined or derived based on the content disclosed in the present invention without departing from the spirit and scope of the present invention. Therefore, the scope of the present invention should be understood and recognized as covering all these other variations or modifications.

Claims

1. A vehicle safety control method based on historical driving data, characterized in that: include: The steps of constructing a safe driving model for the current vehicle, and controlling the safe driving of the current vehicle according to the safe driving model; The step of constructing a safe driving model for the current vehicle includes: Acquire historical driving data of the current vehicle in a set time period, and acquire optimal driving modes for multiple historical driving environments from the historical driving data; Obtaining adjacent driving environments of each of the historical driving environments according to the similarities between the historical driving environments, and selecting a plurality of reference driving environments from each of the historical driving environments according to the number of the adjacent driving environments; Clustering the similarities between the reference driving environments to obtain a plurality of historical environment clusters, and calculating the cluster center of each of the historical environment clusters to obtain a first safe driving model; Acquire new driving data of the current vehicle after the set time period, and when the amount of the new driving data is greater than a first set data amount threshold, acquire optimal driving modes for multiple new driving environments from the new driving data; Clustering each of the newly added driving environments to obtain a plurality of newly added environment clusters, and calculating a cluster center of each of the newly added environment clusters to obtain a second safe driving model; The step of controlling the current vehicle to drive safely according to the safe driving model comprises: Detecting the current driving environment, and when the driving risk index is greater than a set risk threshold, selecting a target historical environment cluster from each of the historical environment clusters according to the similarity between the cluster center and the current driving environment, and taking the reference driving environment with the greatest similarity to the current driving environment in the target historical environment cluster as the target historical driving environment; Selecting a new target cluster from each of the new environment clusters according to the similarity between the cluster center and the current driving environment, and taking multiple new driving environments in the new target clusters with the greatest similarity to the current driving environment as new target driving environments; The neighboring driving environments of the target historical driving environment are integrated with the optimal driving mode of the newly added target driving environment to generate a safe driving strategy, and the current vehicle is controlled to drive safely according to the safe driving strategy.

2. The vehicle safety control method according to claim 1, characterized in that: After the step of acquiring the newly added driving data of the current vehicle after the set time period, the method further includes: Acquiring an update condition of the safe driving model according to the historical driving data; Determining whether the amount of the newly added driving data is greater than a second set data amount threshold; If so, detecting whether the current vehicle meets the update condition; If it is less than, return to the step of obtaining the historical driving data of the set time period.

3. The vehicle safety control method according to claim 1, characterized in that: The step of calculating the cluster center of each historical environment cluster includes: The fitness values ​​of the reference driving environments as the cluster centers of the corresponding historical environment clusters are calculated respectively, and the reference driving environments with the largest fitness values ​​in the historical environment clusters are taken as the cluster centers of the corresponding historical environment clusters.

4. The vehicle safety control method according to claim 1, characterized in that: The step of clustering each of the newly added driving environments to obtain a plurality of newly added environment clusters includes: A first preset number of the newly added environment clusters is obtained, and a k-means clustering algorithm is used to cluster each of the newly added driving environments according to the first preset number to obtain a plurality of the newly added environment clusters.

5. The vehicle safety control method according to claim 1, characterized in that: The step of selecting a new target cluster from each of the new environment clusters according to the similarity between the cluster center and the current driving environment includes: A second preset number is obtained, and a second preset number of newly added environment clusters whose cluster centers have the greatest similarity with the current driving environment are all used as the newly added target clusters.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the vehicle safety control method according to any one of claims 1 to 5 are implemented.

7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the vehicle safety control method according to any one of claims 1 to 5 are implemented.

Images