An AI-based intelligent driving assistance system and method for electric vehicles

Abstract

The present invention discloses an intelligent driving assistance system and method for electric vehicles based on artificial intelligence, belonging to the technical field of artificial intelligence. Through two influencing factor levels of road condition information and turning angle, neural network and Bayesian algorithm are combined to screen data, and various situations of driving state risks are refined. Then, combined with the similarity algorithm, the optimal solution of the driving state situation closest to the current state is obtained, and further whether there is a potential safety hazard in the current state is judged. If there is a potential safety hazard, the driving speed is further regulated through the intelligent regulation value, so as to help the driver accurately regulate the driving speed when the road conditions cannot be directly perceived. Especially, through the tire turning angle, the driver is helped to perform intelligent regulation and warning in both straight and turning modes, ensuring the safe driving of the driver.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and particularly to an intelligent driving assistance system and method for electric vehicles based on artificial intelligence. Background Technique

[0002] With the continuous increase in the penetration rate of vehicle electrification, the power and acceleration of various brands of vehicles tend to be homogenized. Therefore, intelligence and networking are important directions for the differential competition layout of automobile manufacturers. Among them, autonomous driving technology is one of the biggest selling points of future intelligent vehicles. According to the role allocation in performing dynamic driving tasks and whether there are design operation condition restrictions, driving automation is divided into levels 0 to 5. For example, under the operation conditions specified by the intelligent driving system, the vehicle itself can complete tasks such as steering, accelerating and decelerating, road condition detection and reaction.

[0003] Wheel slippage is a common phenomenon during vehicle driving. The causes of wheel slippage are mostly poor road conditions while the driver is still driving too fast, or the driver is forced to suddenly change direction or speed, resulting in the vehicle's gravity suddenly concentrating on one end or side, and the corresponding tire thus losing adhesion. At present, the problem of wheel slippage is often prevented through the driver's driving experience. When the road conditions are unclear, intelligent driving technology is needed to help the driver perceive the running state of the vehicle and ensure the driver's safe driving. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent driving assistance system and method for electric vehicles based on artificial intelligence to solve the problems raised in the above background technique.

[0005] To solve the above technical problems, the present invention provides the following technical solutions:

[0006] An intelligent driving assistance system for electric vehicles based on artificial intelligence, the system includes: a driving assistance database module, an intelligent driving assistance pre-judgment module, an intelligent driving assistance research and judgment module, and an intelligent driving control module;

[0007] The driving assistance database module is used to pre - construct a driving assistance database, which includes a road surface condition database and a driving condition database; the road surface condition database stores road surface condition information; the driving condition database stores the rotation angle and driving speed of the electric vehicle tires; the road surface condition information is collected by a laser remote sensing sensor, and the road surface condition information includes road surface water accumulation condition characteristic information, road surface freezing condition characteristic information, and road surface dust accumulation condition characteristic information, where the dust accumulation includes sandy and muddy substances attached to the road surface; the road surface condition characteristic parameters corresponding to the road surface water accumulation condition characteristic information, the road surface freezing condition characteristic information, and the road surface dust accumulation condition characteristic information are respectively extracted and stored in the road surface condition database; the rotation angle of the electric vehicle tires is obtained through an image acquisition sensor, and the rotation angle and driving speed of the electric vehicle tires are stored in the driving condition database;

[0008] The intelligent driving assistance prediction module is used to construct an intelligent driving assistance prediction condition model based on the road surface condition database and the driving condition database, and output the safe driving speed of the electric vehicle and the standard data set corresponding to the safe driving speed;

[0009] The intelligent driving assistance judgment module is used to collect the road surface condition data and driving condition data during the driving process of the electric vehicle, and calculate the intelligent driving assistance judgment similarity based on the intelligent driving assistance prediction condition model;

[0010] The intelligent driving control module is used to intelligently regulate and give early warning prompts for the driving speed of the electric vehicle according to the intelligent driving assistance judgment similarity.

[0011] Furthermore, the intelligent driving assistance prediction module further includes a data screening unit and a prediction state generation unit;

[0012] The data screening unit is used to use the road surface condition characteristic parameters and the tire rotation angle as input information, and the driving speed when the electric vehicle slips as output information, train through a neural network model, integrate the training results, and generate a sample set of the input information corresponding to the same driving speed, denoted as V X ={G1, G2,..., G Y}, where V X represents any same driving speed, X represents the number of the same driving speed, and any element in G1, G2,..., G Y represents one of the road surface condition characteristic parameters and the tire rotation angle;

[0013] The prediction state generation unit is used to construct an intelligent driving assistance prediction condition model. Taking any same driving speed as a safe driving speed, the total number of types of safe driving speeds is M, and M≥X; calculate the safe driving speed V XThe conditional probability P(V X |W) = P(W|V X ) * P(V X ) / P(W), and P(W) = P(W|V X ) × P(V X ), where W represents the sample set corresponding to other safe driving speeds with any element in the sample set V X . P(W|V X ) represents the conditional probability of W, and P(V X ) and P(W) represent the probabilities of V X and W respectively; it is also used to train and generate the optimal data corresponding to the road surface water condition feature information, road surface freezing condition feature information, road surface dust condition feature information, and tire rotation angle respectively, set the model training threshold, and when P(V X |W) is greater than or equal to the model training threshold, it indicates that the training of the optimal data is completed; integrate the training results, and record a set of data corresponding to V X as the standard data set AV X = {R1, R2, R3, R4}, where R1, R2, R3, and R4 correspond to the road surface water condition feature parameter, road surface freezing condition feature parameter, road surface dust condition feature parameter, and tire rotation angle respectively.

[0014] Further, the intelligent driving assistance judgment module further includes a judgment similarity calculation unit and a data refinement unit;

[0015] The judgment similarity calculation unit is used to collect the road surface condition data and driving condition data during the driving process of the electric vehicle, generate the real-time driving behavior set S = {r1, r2, r3, r4}, convert the real-time driving behavior set and the standard data set into matrix forms S = (r1, r2, r3, r4) and AV X = (R1, R2, R3, R4) respectively, and calculate the intelligent driving assistance judgment similarity. The specific calculation formula is as follows:

[0016]

[0017] where E represents the intelligent driving assistance judgment similarity, represents the transpose of AV X = (R1, R2, R3, R4), and ||S|| and ||AV X || represent the norms of S = (r1, r2, r3, r4) and AV X = (R1, R2, R3, R4) respectively;

[0018] The data refining unit is used to preset a similarity threshold. If E is greater than or equal to the similarity threshold, the standard data set AV is extracted. X ={R1, R2, R3, R4}.

[0019] Furthermore, the intelligent driving control module further includes a warning judgment unit and an intelligent regulation unit;

[0020] The warning judgment unit is used to obtain the extracted standard data set AV according to the similarity of intelligent driving assistance research and judgment. X ={R1, R2, R3, R4} corresponding to the safe driving speed V X , calculate the average value of the safe driving speed, and use the average value as the warning prompt value. If the current driving speed is less than the warning prompt value, no warning prompt is issued. If the current driving speed is greater than or equal to the warning prompt value, a warning prompt is issued;

[0021] The intelligent regulation unit is used to calculate an intelligent regulation value according to the warning prompt value. The specific calculation formula is as follows:

[0022]

[0023] Where, V0 represents the intelligent regulation value, H represents the total number of the extracted standard data sets, and L represents the warning prompt value;

[0024] Then the maximum value of the current driving speed is V Max =L - V0.

[0025] An intelligent driving assistance method for electric vehicles based on artificial intelligence, this method includes the following steps:

[0026] Step S100: Pre-construct a driving assistance database, the driving assistance database includes a road surface condition database and a driving condition database; the road surface condition database stores road surface condition information; the driving condition database stores the rotation angle and driving speed of the electric vehicle tires;

[0027] Step S200: According to the road surface condition database and the driving condition database, construct an intelligent driving assistance pre-judgment condition model, and output the safe driving speed of the electric vehicle intelligent driving and the standard data set corresponding to the safe driving speed;

[0028] Step S300: Collect road surface condition data and driving condition data during the driving process of the electric vehicle, and calculate the similarity of intelligent driving assistance research and judgment based on the intelligent driving assistance pre-judgment condition model;

[0029] Step S400: According to the similarity of intelligent driving assistance research and judgment, perform intelligent regulation and warning prompts on the driving speed of the electric vehicle.

[0030] Furthermore, the specific implementation process of step S100 includes:

[0031] Road condition information is collected through a laser remote sensing sensor, wherein the road condition information includes road surface waterlogging condition characteristic information, road surface icing condition characteristic information and road surface dust accumulation condition characteristic information, wherein the dust accumulation includes sand and mud attached to the road surface; road condition characteristic parameters corresponding to the road surface waterlogging condition characteristic information, road surface icing condition characteristic information and road surface dust accumulation condition characteristic information are extracted respectively, and stored in a road condition database; the rotation angle of the electric vehicle tire is obtained through an image acquisition sensor, and the rotation angle and driving speed of the electric vehicle tire are stored in a driving condition database.

[0032] Furthermore, the specific implementation process of step S200 includes:

[0033] Step S201: Take the road condition characteristic parameters and the tire rotation angle as input information, take the driving speed of the electric vehicle when it slips as output information, train the neural network model, integrate the training results, and generate a sample set corresponding to the input information with the same driving speed, denoted as V X = {G1, G2, ..., G Y}, where V X represents any same driving speed, X represents the number of the same driving speed, G1, G2, ..., G Y Any element in represents one of the characteristic parameters of road condition and the tire rotation angle;

[0034] Step S202: construct an intelligent driving assistance prediction model, take any same driving speed as a safe driving speed, then the total number of safe driving speed types is M, and M≥X; calculate the safe driving speed V X The conditional probability P(V X |W)=P(W|V X )*P(V X ) / P(W), and P(W)=P(W|V X )×P(V X ), where W represents the sample set V X The sample set corresponding to other safe driving speeds of any element in , P(W|V X ) represents the conditional probability of W, P(V X ) and P(W) represent V X and the probability of W;

[0035] Step S203: Train and generate the optimal data of the road surface condition characteristic parameters corresponding to the road surface water condition characteristic information, the road surface icy condition characteristic information and the road surface dust condition characteristic information and the tire rotation angle, set the model training threshold, and when P(VX When |W) is greater than or equal to the model training threshold, it indicates that the data for optimal training is completed; the training results are integrated, and V X A set of data corresponding to the training is denoted as the standard data set AV X ={R1, R2, R3, R4}, where R1, R2, R3, and R4 respectively correspond to the road surface water accumulation condition characteristic parameter, the road surface freezing condition characteristic parameter, the road surface dust accumulation condition characteristic parameter, and the tire rotation angle;

[0036] According to the above method, during the driving process of the electric vehicle, the driver needs to observe the road conditions and then control the vehicle speed. This process is related to the driver's driving experience. In practice, for potholed road conditions, the driver can intuitively feel the road conditions through the bumpiness of the tires and then control the vehicle speed well. However, for road conditions with water accumulation, freezing, and dust accumulation, the driver often cannot intuitively feel them and is prone to misjudging the vehicle speed during the driving process, especially during straight driving and turning, it is easy to exceed the speed, which may cause the tires to slip and result in traffic accidents; therefore, this application combines the road conditions that are not easily intuitively felt by the driver and the two states of straight driving and turning during the driving process to make an intelligent auxiliary judgment on the driving vehicle speed; an intelligent driving assistance pre-judgment condition model is established, and through the Bayesian probability theory, the road surface water accumulation condition characteristic parameter, the road surface freezing condition characteristic parameter, the road surface dust accumulation condition characteristic parameter, and the tire rotation angle are quantitatively analyzed, and the optimal data set that causes the vehicle to slip at different driving speeds is output, that is, the road surface water accumulation condition characteristic parameter, the road surface freezing condition characteristic parameter, the road surface dust accumulation condition characteristic parameter, and the tire rotation angle.

[0037] Further, the specific implementation process of step S300 includes:

[0038] Step S301: Collect the road surface condition data and driving condition data during the driving process of the electric vehicle, generate the real-time driving behavior set S={r1, r2, r3, r4}, convert the real-time driving behavior set and the standard data set into matrix forms S=(r1, r2, r3, r4) and AV X =(R1, R2, R3, R4), calculate the intelligent driving assistance research and judgment similarity, and the specific calculation formula is as follows:

[0039]

[0040] where E represents the intelligent driving assistance research and judgment similarity, represents the transpose of AV X =(R1, R2, R3, R4), ||S|| and ||AV X || respectively represent the norms of S=(r1, r2, r3, r4) and AV X =(R1, R2, R3, R4);

[0041] Step S302: Preset a similarity threshold. If E is greater than or equal to the similarity threshold, extract the standard data set AV X ={R1, R2, R3, R4};

[0042] According to the above method, match and extract the real-time data during driving with the standard data set, so as to obtain the critical driving speed under different standard data sets, that is, the maximum driving speed causing skidding or traffic accidents, from the perspective of the intuitive factor of driving speed that causes traffic accidents or skidding situations.

[0043] Further, the specific implementation process in the step S400 includes:

[0044] Step S401: Obtain the safety driving speed V corresponding to the extracted standard data set AV X ={R1, R2, R3, R4}, calculate the average value of the safety driving speed, and use the average value as the warning prompt value. If the current driving speed is less than the warning prompt value, no warning prompt is issued. If the current driving speed is greater than or equal to the warning prompt value, a warning prompt is issued and step S402 is executed; X

[0045] Step S402: Calculate the intelligent regulation value according to the warning prompt value. The specific calculation formula is as follows:

[0046]

[0047] Among them, V0 represents the intelligent regulation value, H represents the total number of the extracted standard data sets, and L represents the warning prompt value;

[0048] Then the maximum value of the current driving speed is V Max =L - V0;

[0049] According to the above method, the intelligent regulation value actually reflects the driving speed fluctuation situation under different standard data sets, reflects the driving stability through the fluctuation situation, and then regulates the driving speed stability through the intelligent regulation value to obtain the maximum safe driving speed for the driver's reference.

[0050] ​Compared with the prior art, the beneficial effects achieved by the present invention are as follows: In an intelligent driving assistance system and method for electric vehicles based on artificial intelligence provided by the present invention, through two influencing factor levels of road condition information and turning angle, neural network and Bayesian algorithm are combined to screen data, and various situations of driving state risks are refined. Then, combined with the similarity algorithm, the optimal solution of the driving state situation closest to the current state is obtained, and further, it is judged whether there is a potential safety hazard in the current state. If there is a potential safety hazard, the driving speed is further regulated through the intelligent regulation value, so as to help the driver accurately regulate the driving speed when the road conditions cannot be intuitively perceived. Especially, through the tire turning angle, the driver is helped to perform intelligent regulation and warning in both straight-line and turning modes, ensuring the safe driving of the driver. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0052] Figure 1 is a schematic structural diagram of an intelligent driving assistance system for electric vehicles based on artificial intelligence according to the present invention;

[0053] Figure 2 is a schematic step diagram of an intelligent driving assistance method for electric vehicles based on artificial intelligence according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0055] Please refer to Figure 1 - Figure 2 , the present invention provides the following technical solutions:

[0056] Please refer to Figure 1 , in the first embodiment: An intelligent driving assistance system for electric vehicles based on artificial intelligence is provided. The system includes: a driving assistance database module, an intelligent driving assistance pre-judgment module, an intelligent driving assistance research and judgment module, and an intelligent driving control module;

[0057] A driving assistance database module is used to pre - construct a driving assistance database. The driving assistance database includes a road surface condition database and a driving condition database. The road surface condition database stores road surface condition information. The driving condition database stores the rotation angle and driving speed of the electric vehicle's tires. The road surface condition information is collected by a laser remote sensing sensor, and the road surface condition information includes road surface water accumulation condition characteristic information, road surface freezing condition characteristic information, and road surface dust accumulation condition characteristic information, where the dust accumulation includes sandy and muddy substances attached to the road surface. The road surface condition characteristic parameters corresponding to the road surface water accumulation condition characteristic information, the road surface freezing condition characteristic information, and the road surface dust accumulation condition characteristic information are respectively extracted and stored in the road surface condition database. The rotation angle of the electric vehicle's tires is obtained through an image acquisition sensor, and the rotation angle and driving speed of the electric vehicle's tires are stored in the driving condition database.

[0058] An intelligent driving assistance prediction module is used to construct an intelligent driving assistance prediction condition model based on the road surface condition database and the driving condition database, and output the safe driving speed of the electric vehicle and the standard data set corresponding to the safe driving speed.

[0059] Among them, the intelligent driving assistance prediction module further includes a data screening unit and a prediction state generation unit.

[0060] The data screening unit is used to take the road surface condition characteristic parameters and the tire rotation angle as input information, take the driving speed when the electric vehicle slips as output information, train through a neural network model, integrate the training results, and generate a sample set of the input information corresponding to the same driving speed, denoted as V X ={G1, G2,..., G Y}, where V X represents any same driving speed, X represents the number of the same driving speed, and any element in G1, G2,..., G Y represents one of the road surface condition characteristic parameters and the tire rotation angle.

[0061] The prediction state generation unit is used to construct an intelligent driving assistance prediction condition model. Taking any same driving speed as a safe driving speed, the total number of types of safe driving speeds is M, and M≥X. Calculate the conditional probability P(V X |W) = P(W|V X )*P(V X ) / P(W), and P(W) = P(W|V X )×P(V X ), where W represents the sample set corresponding to other safe driving speeds with any element in the sample set V X , P(W|V X ), and P(W|V X) represents the conditional probability of W, P(V X ) and P(W) represent the probabilities of V X and W respectively; it is also used to train the optimal data corresponding to the road surface water condition feature information, road surface freezing condition feature information, road surface dust condition feature information, and tire rotation angle respectively, set the model training threshold, and when P(V X |W) is greater than or equal to the model training threshold, it means that the training of the optimal data is completed; integrate the training results, and record a set of data corresponding to V X as the standard data set AV X ={R1, R2, R3, R4}, where R1, R2, R3, and R4 correspond to the road surface water condition feature parameter, road surface freezing condition feature parameter, road surface dust condition feature parameter, and tire rotation angle respectively;

[0062] The intelligent driving assistance judgment module is used to collect the road surface condition data and driving condition data during the driving of the electric vehicle, and calculate the intelligent driving assistance judgment similarity based on the intelligent driving assistance pre-judgment condition model;

[0063] Among them, the intelligent driving assistance judgment module also includes a judgment similarity calculation unit and a data extraction unit;

[0064] The judgment similarity calculation unit is used to collect the road surface condition data and driving condition data during the driving of the electric vehicle, generate the real-time driving behavior set S={r1, r2, r3, r4}, convert the real-time driving behavior set and the standard data set into matrix forms S=(r1, r2, r3, r4) and AV X =(R1, R2, R3, R4) respectively, and calculate the intelligent driving assistance judgment similarity. The specific calculation formula is as follows:

[0065]

[0066] Among them, E represents the intelligent driving assistance judgment similarity, represents the transpose of AV X =(R1, R2, R3, R4), ||S|| and ||AV X || represent the norms of S=(r1, r2, r3, r4) and AV X =(R1, R2, R3, R4) respectively;

[0067] The data extraction unit is used to preset the similarity threshold. If E is greater than or equal to the similarity threshold, extract the standard data set AV X ={R1, R2, R3, R4};

[0068] The intelligent driving control module is used to intelligently regulate the driving speed of the electric vehicle and give early warning prompts according to the similarity of intelligent driving assistance judgment;

[0069] Among them, the intelligent driving control module further includes an early warning judgment unit and an intelligent regulation unit;

[0070] The early warning judgment unit is used to obtain the extracted standard data set AV X ={R1, R2, R3, R4} corresponding safe driving speed V X , calculate the average value of the safe driving speed, and use the average value as the early warning prompt value. If the current driving speed is less than the early warning prompt value, no early warning prompt is issued. If the current driving speed is greater than or equal to the early warning prompt value, an early warning prompt is issued;

[0071] The intelligent regulation unit is used to calculate the intelligent regulation value according to the early warning prompt value. The specific calculation formula is as follows:

[0072]

[0073] Among them, V0 represents the intelligent regulation value, H represents the total number of the extracted standard data sets, and L represents the early warning prompt value;

[0074] Then the maximum value of the current driving speed is V Max =L - V0.

[0075] Please refer to Figure 2 , in the second embodiment: Provide an intelligent driving assistance method for an electric vehicle based on artificial intelligence. The method includes the following steps:

[0076] Pre - construct a driving assistance database. The driving assistance database includes a road surface condition database and a driving condition database; The road surface condition database stores road surface condition information; The driving condition database stores the rotation angle of the electric vehicle tire and the driving speed;

[0077] Collect road surface condition information through a laser remote sensing sensor. The road surface condition information includes road surface water accumulation condition characteristic information, road surface freezing condition characteristic information, and road surface dust accumulation condition characteristic information. Among them, dust accumulation includes sandy mud attached to the road surface; Respectively extract the road surface condition characteristic parameters corresponding to the road surface water accumulation condition characteristic information, road surface freezing condition characteristic information, and road surface dust accumulation condition characteristic information, and store them in the road surface condition database; Obtain the rotation angle of the electric vehicle tire through an image acquisition sensor, and store the rotation angle of the electric vehicle tire and the driving speed in the driving condition database;

[0078] According to the road surface condition database and the driving condition database, construct an intelligent driving assistance pre - judgment condition model, and output the safe driving speed of the electric vehicle intelligent driving and the standard data set corresponding to the safe driving speed;

[0079] Taking the road surface condition characteristic parameters and the tire rotation angle as input information, and taking the driving speed when the electric vehicle slips as output information, training through a neural network model, integrating the training results, and generating a sample set from the input information corresponding to the same driving speed, denoted as V X ={G1, G2,..., G Y}, where V X represents any same driving speed, X represents the number of the same driving speed, and any element in G1, G2,..., G Y represents one of the road surface condition characteristic parameters and the tire rotation angle;

[0080] Construct an intelligent driving assistance pre-judgment condition model. Taking any same driving speed as a safe driving speed, the total number of types of safe driving speeds is M, and M≥X; calculate the conditional probability P(V X |W) = P(W|V X )*P(V X ) / P(W), and P(W) = P(W|V X )×P(V X ), where W represents the sample set corresponding to other safe driving speeds with any element in the sample set V X , P(W|V X ) represents the conditional probability of W, and P(V X ) and P(W) represent the probabilities of V X and W respectively;

[0081] Train and generate the optimal data in the road surface condition characteristic parameters and the tire rotation angle corresponding to the road surface water accumulation condition characteristic information, the road surface freezing condition characteristic information, and the road surface dust accumulation condition characteristic information respectively. Set the model training threshold. When P(V X |W) is greater than or equal to the model training threshold, it means that the training of the optimal data is completed; integrate the training results, and record a set of data corresponding to the training of V X as the standard data set AV X ={R1, R2, R3, R4}, where R1, R2, R3, and R4 correspond to the road surface water accumulation condition characteristic parameters, the road surface freezing condition characteristic parameters, the road surface dust accumulation condition characteristic parameters, and the tire rotation angle respectively; X Collect the road surface condition data and the driving condition data during the driving process of the electric vehicle, and calculate the intelligent driving assistance judgment similarity based on the intelligent driving assistance pre-judgment condition model;

[0082]

[0083] ​Collect the road condition data and driving condition data during the driving process of the electric vehicle, generate a real-time driving behavior set S = {r1, r2, r3, r4}, and convert the real-time driving behavior set and the standard data set into matrix forms S = (r1, r2, r3, r4) and AV X =(R1, R2, R3, R4), calculate the similarity of intelligent driving assistance judgment, and the specific calculation formula is as follows:

[0084]

[0085] Among them, E represents the similarity of intelligent driving assistance judgment, represents the transpose of AV X =(R1, R2, R3, R4), ||S|| and ||AV X || respectively represent the norms of S = (r1, r2, r3, r4) and AV X =(R1, R2, R3, R4);

[0086] Preset a similarity threshold. If E is greater than or equal to the similarity threshold, extract the standard data set AV X ={R1, R2, R3, R4};

[0087] According to the similarity of intelligent driving assistance judgment, perform intelligent regulation and warning prompts on the driving speed of the electric vehicle;

[0088] According to the similarity of intelligent driving assistance judgment, obtain the extracted standard data set AV X ={R1, R2, R3, R4} corresponding safe driving speed V X , calculate the average value of the safe driving speed, and use the average value as the warning prompt value. If the current driving speed is less than the warning prompt value, no warning prompt is issued. If the current driving speed is greater than or equal to the warning prompt value, a warning prompt is issued;

[0089] According to the warning prompt value, calculate the intelligent regulation value, and the specific calculation formula is as follows:

[0090]

[0091] Among them, V0 represents the intelligent regulation value, H represents the total number of the extracted standard data set, and L represents the warning prompt value;

[0092] Then the maximum value of the current driving speed is V Max= L - V0; For example, through the matching of Bayesian algorithm and similarity, 5 standard data sets are obtained, and the corresponding safe driving speeds are {50, 60, 65, 70, 55} respectively, then the warning prompt value L = 60; If the current driving speed is 55, no warning prompt is issued; If the current driving speed is 75, a warning prompt is issued, calculate the intelligent regulation value V0 = 6, then the maximum value V of the current driving speed Max = 54.

[0093] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent driving assistance method for electric vehicles based on artificial intelligence, characterized in that, The method includes the following steps: Step S100: Pre-construct a driving assistance database, where the driving assistance database includes a road surface condition database and a driving condition database; road surface condition information is stored in the road surface condition database; the rotation angle and driving speed of the electric vehicle tires are stored in the driving condition database; Step S200: Construct an intelligent driving assistance pre-judgment condition model based on the road surface condition database and the driving condition database, and output the safe driving speed of the electric vehicle and the standard data set corresponding to the safe driving speed; Step S300: Collect the road surface condition data and driving condition data during the driving process of the electric vehicle, and calculate the intelligent driving assistance research and judgment similarity based on the intelligent driving assistance pre-judgment condition model; Step S400: Intelligently regulate and give early warning prompts for the driving speed of the electric vehicle according to the intelligent driving assistance research and judgment similarity; The specific implementation process in step S400 includes: Step S401: Obtain the extracted standard data set AV according to the similarity judgment of intelligent driving assistance X ={R1, R2, R3, R4} corresponding to the safe driving speed V X , calculate the average value of the safe driving speed, use the average value as the warning prompt value. If the current driving speed is less than the warning prompt value, no warning prompt is issued. If the current driving speed is greater than or equal to the warning prompt value, a warning prompt is issued and step S402 is executed; Among them, X represents the number of the same driving speed; V X A set of data corresponding to the training is recorded as the standard data set AV X = {R1, R2, R3, R4}, where R1, R2, R3, and R4 respectively correspond to the road surface water accumulation condition characteristic parameter, the road surface freezing condition characteristic parameter, the road surface dust accumulation condition characteristic parameter, and the tire rotation angle; Step S402: Calculate the intelligent regulation value according to the early warning prompt value, and the specific calculation formula is as follows: Where, V0 represents the intelligent regulation value, H represents the total number of the extracted standard data sets, and L represents the early warning prompt value; Then the maximum value of the current driving speed is V Max = L - V0.

2. The intelligent driving assistance method for an electric vehicle based on artificial intelligence according to claim 1, wherein The specific implementation process of step S100 includes: Collect road surface condition information through a laser remote sensing sensor. The road surface condition information includes road surface water accumulation condition characteristic information, road surface freezing condition characteristic information, and road surface dust accumulation condition characteristic information, where the dust accumulation includes sandy and muddy substances attached to the road surface; respectively extract the road surface condition characteristic parameters corresponding to the road surface water accumulation condition characteristic information, road surface freezing condition characteristic information, and road surface dust accumulation condition characteristic information, and store them in the road surface condition database; obtain the rotation angle of the electric vehicle tires through an image acquisition sensor, and store the rotation angle and driving speed of the electric vehicle tires in the driving condition database.

3. An intelligent driving assistance method for electric vehicles based on artificial intelligence according to claim 2, characterized in that, The specific implementation process of step S200 includes: Step S201: Use the road surface condition characteristic parameters and the tire rotation angle as input information, and use the driving speed when the electric vehicle slips as output information. Train through a neural network model, integrate the training results, and generate a sample set for the input information corresponding to the same driving speed, denoted as V X ={G1, G2,..., G Y}, where any element in G1, G2,..., G Y represents one of the road surface condition characteristic parameters and the tire rotation angle; Step S202: construct an intelligent driving assistance prediction model, take any same driving speed as a safe driving speed, then the total number of safe driving speed types is M, and M≥X; calculate the safe driving speed V X The conditional probability P(V X |W)=P(W|V X )*P(V X ) / P(W), and P(W)=P(W|V X )×P(V X ), where W represents the sample set V X The sample set corresponding to other safe driving speeds of any element in , P(W|V X ) represents the conditional probability of W, P(V X ) and P(W) represent V X and the probability of W; Step S203: Respectively train the optimal data corresponding to the road surface condition characteristic parameters and the tire rotation angle for the road surface water accumulation condition characteristic information, the road surface freezing condition characteristic information, and the road surface dust accumulation condition characteristic information, and set the model training threshold. When P(V X |W) is greater than or equal to the model training threshold, it indicates that the training of the optimal data is completed; integrate the training results.

4. The intelligent driving assistance method for an electric vehicle based on artificial intelligence according to claim 3, wherein, The specific implementation process of step S300 includes: Step S301: Collect the road condition data and driving condition data during the driving of the electric vehicle, generate a real-time driving behavior set S = {r1, r2, r3, r4}, and convert the real-time driving behavior set and the standard data set into matrix forms S = (r1, r2, r3, r4) and AV X = (R1, R2, R3, R4), calculate the similarity of intelligent driving assistance judgment, and the specific calculation formula is as follows: Among them, E represents the similarity of intelligent driving assistance judgment, represents AV X =(transpose of R1, R2, R3, R4), ||S|| and ||AV X || represent the norms of S=(r1, r2, r3, r4) and AV X =(R1, R2, R3, R4) respectively; Step S302: Preset a similarity threshold. If E is greater than or equal to the similarity threshold, then extract the standard data set AV X ={R1, R2, R3, R4}.

5. An intelligent driving assistance system for electric vehicles based on artificial intelligence, characterized in that, The system includes: a driving assistance database module, an intelligent driving assistance pre-judgment module, an intelligent driving assistance research and judgment module, and an intelligent driving control module; The driving assistance database module is used to pre-construct a driving assistance database, where the driving assistance database includes a road surface condition database and a driving condition database; road surface condition information is stored in the road surface condition database; the rotation angle and driving speed of the electric vehicle tires are stored in the driving condition database; collect road surface condition information through a laser remote sensing sensor. The road surface condition information includes road surface water accumulation condition characteristic information, road surface freezing condition characteristic information, and road surface dust accumulation condition characteristic information, where the dust accumulation includes sandy and muddy substances attached to the road surface; respectively extract the road surface condition characteristic parameters corresponding to the road surface water accumulation condition characteristic information, road surface freezing condition characteristic information, and road surface dust accumulation condition characteristic information, and store them in the road surface condition database; obtain the rotation angle of the electric vehicle tires through an image acquisition sensor, and store the rotation angle and driving speed of the electric vehicle tires in the driving condition database; The intelligent driving assistance pre-judgment module is used to construct an intelligent driving assistance pre-judgment condition model based on the road condition database and the driving condition database, and output the safe driving speed of the electric vehicle and the standard data set corresponding to the safe driving speed; The intelligent driving assistance research and judgment module is used to collect the road condition data and driving condition data during the driving process of the electric vehicle, and calculate the intelligent driving assistance research and judgment similarity based on the intelligent driving assistance pre-judgment condition model; The intelligent driving control module is used to intelligently regulate and give early warning prompts to the driving speed of the electric vehicle according to the intelligent driving assistance research and judgment similarity; The intelligent driving control module further includes an early warning judgment unit and an intelligent regulation unit; The warning judgment unit is used to obtain the extracted standard data set AV according to the similarity of intelligent driving assistance research and judgment X The corresponding safe driving speed V of = {R1, R2, R3, R4} X , calculate the average value of the safe driving speed, and use the average value as the warning prompt value. If the current driving speed is less than the warning prompt value, no warning prompt is issued. If the current driving speed is greater than or equal to the warning prompt value, a warning prompt is issued; The intelligent regulation unit is used to calculate the intelligent regulation value according to the early warning prompt value, and the specific calculation formula is as follows: Wherein, V0 represents the intelligent regulation value, H represents the total number of the extracted standard data sets, and L represents the early warning prompt value; Then the maximum value of the current driving speed is V Max = L - V0; Among them, X represents the number of the same driving speed; V X A set of corresponding training data is denoted as the standard data set AV X = {R1, R2, R3, R4}, where R1, R2, R3, and R4 respectively correspond to the characteristic parameters of road surface water accumulation condition, road surface freezing condition, road surface ash accumulation condition, and tire rotation angle.

6. The intelligent driving assistance system for electric vehicles based on artificial intelligence according to claim 5, wherein: The intelligent driving assistance pre-judgment module further includes a data screening unit and a pre-judgment state generation unit; The data screening unit is used to take the road surface condition characteristic parameters and the tire rotation angle as input information, take the driving speed when the electric vehicle slips as output information, train through a neural network model, integrate the training results, and generate a sample set for the input information corresponding to the same driving speed, denoted as V X ={G1, G2,..., G Y}, where V X represents any same driving speed, and any element in G1, G2,..., G Y represents one of the road surface condition characteristic parameters and the tire rotation angle; The pre-judgment state generation unit is used to construct an intelligent driving assistance pre-judgment condition model. Taking any same driving speed as a safe driving speed, the total number of types of safe driving speeds is M, and M≥X; Calculate the safe driving speed V X The conditional probability P(V X |W) = P(W|V X ) * P(V X ), and P(W) = P(W|V X ) × P(V X ), where W represents the sample set corresponding to any other safe driving speed of the elements in the sample set V X . P(W|V X ) represents the conditional probability of W, and P(V X ) and P(W) represent the probabilities of V X and W respectively; it is also used to train the optimal data corresponding to the road surface water condition characteristic information, road surface freezing condition characteristic information, and road surface dust condition characteristic information respectively to generate the road surface condition characteristic parameters and the tire rotation angle, set the model training threshold. When P(V X |W) is greater than or equal to the model training threshold, it means that the training of the optimal data is completed; integrate the training results.

7. An intelligent driving assistance system for electric vehicles based on artificial intelligence according to claim 6, characterized in that: The intelligent driving assistance research and judgment module further includes a research and judgment similarity calculation unit and a data refinement unit; The research and judgment similarity calculation unit is used to collect road condition data and driving condition data during the driving process of the electric vehicle, generate a real-time driving behavior set S = {r1, r2, r3, r4}, convert the real-time driving behavior set and the standard data set into matrix forms S = (r1, r2, r3, r4) and AV respectively X = (R1, R2, R3, R4), and calculate the intelligent driving assistance research and judgment similarity. The specific calculation formula is as follows: Among them, E represents the similarity of intelligent driving assistance judgment, represents AV X =(R1, R2, R3, R4) transpose, ||S|| and ||AV X || respectively represent the modulus lengths of S=(r1, r2, r3, r4) and AV X =(R1, R2, R3, R4); The data extraction unit is used to preset a similarity threshold. If E is greater than or equal to the similarity threshold, the standard data set AV is extracted. X ={R1, R2, R3, R4}.

Images