Method and system for predicting injury severity of non-standard sitting passenger in intelligent cockpit

Abstract

The application belongs to the technical field of collision safety, and discloses a damage severity prediction method and system for non-standard sitting posture passengers in an intelligent cockpit, which comprises collecting real traffic accident cases, reconstructing simulation working conditions, obtaining a high-risk working condition data set, identifying the accident form features and passenger damage severity of the high-risk working condition data set, obtaining a standard posture data set under the high-risk working condition, considering the collision parameters and passenger posture parameters, obtaining a simulation data set coupled with the collision parameters and non-standard sitting posture parameters, fusing the standard posture data set and the simulation data set, obtaining a high-risk working condition global sitting posture database, training a neural network by using the high-risk working condition global sitting posture database, obtaining a trained prediction model, and predicting the passenger damage severity by using the trained prediction model. The application realizes accurate safety evaluation of the non-standard sitting posture in the intelligent cockpit, and completes the passenger damage severity prediction of the non-standard sitting posture in the intelligent cockpit through a machine learning algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of collision safety technology, in particular to a damage severity prediction method and system for non-standard sitting posture passengers in an intelligent cabin. BACKGROUND

[0002] The statements in this section merely provide background information related to the present application and do not necessarily constitute the prior art.

[0003] With the rapid development and popularization of automatic driving technology, the traditional driver-centered vehicle design concept is facing a fundamental change. In the highly automated driving (SAE L3 and above) scenario, passengers will be freed from the task of driving and turn to more free and diverse sitting postures, such as semi-laying, seat rotation or fully stretched legs, etc. This change brings comfort experience to passengers, but also poses new challenges to the passive safety system of the vehicle. The current collision safety protection system and evaluation standard based on standard driving posture have been difficult to meet future needs. SUMMARY

[0004] To solve the above problems, the present application provides a damage severity prediction method and system for non-standard sitting posture passengers in an intelligent cabin, which realizes accurate safety evaluation of non-standard sitting posture in an intelligent cabin, completes intelligent identification of high-risk working conditions and extraction of biomechanical characteristics, establishes a coupling simulation data set of collision parameters and non-standard sitting posture parameters, and completes prediction of passenger damage severity in an intelligent cabin with non-standard sitting posture through a machine learning algorithm.

[0005] To achieve the above purpose, the present application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a damage severity prediction method for non-standard sitting posture passengers in an intelligent cabin, comprising the following steps:

[0007] Collect real traffic accident case data and use multi-body dynamics to reconstruct simulation working conditions to enrich data and obtain a high-risk working condition data set;

[0008] Identify the accident morphology characteristics and passenger damage severity of the high-risk working condition data set, and obtain a standard posture data set under high-risk working conditions based on the accident morphology characteristics and passenger damage severity;

[0009] Based on the standard posture data set, consider the collision parameters and passenger posture parameters, and obtain a simulation data set of the coupling of collision parameters and non-standard sitting posture parameters under high-risk working conditions. Fuse the standard posture data set and the simulation data set to obtain a high-risk working condition global sitting posture database;

[0010] The neural network is trained by using the high-risk working condition global sitting posture database to obtain a trained prediction model, and the trained prediction model is used for predicting the passenger injury severity.

[0011] As an alternative embodiment, the final predicted passenger injury severity is classified into risk levels according to the AIS-ISS scoring standard.

[0012] As an alternative embodiment, the accident form features include accident frequency, severity, collision direction, collision position, collision angle, collision speed, passenger injury, and collision object.

[0013] As an alternative embodiment, the passenger injury severity includes casualty and abbreviated injury scale.

[0014] As an alternative embodiment, the collision parameters include collision speed, collision position, collision angle, and vehicle stiffness parameters.

[0015] As an alternative embodiment, the passenger posture parameters include seat orientation, seat angle, passenger posture, seat belt wearing state, and airbag triggering timing.

[0016] In a second aspect, the present application provides a damage severity prediction system for non-standard sitting posture passengers in an intelligent cockpit, comprising:

[0017] The data acquisition module is configured to collect real traffic accident case data, and to enrich the data by using multi-body dynamics to simulate the working condition reconstruction, and to obtain a high-risk working condition data set;

[0018] The standard posture data set construction module is configured to identify the accident form features and the passenger injury severity of the high-risk working condition data set, and to obtain a standard posture data set under high-risk working conditions based on the accident form features and the passenger injury severity;

[0019] The global sitting posture database construction module is configured to obtain a simulation data set coupled with non-standard sitting posture parameters under high-risk working conditions based on the standard posture data set, considering the collision parameters and the passenger posture parameters, to fuse the standard posture data set and the simulation data set, and to obtain a high-risk working condition global sitting posture database;

[0020] The model training and prediction module is configured to train the neural network by using the high-risk working condition global sitting posture database to obtain a trained prediction model, and to predict the passenger injury severity by using the trained prediction model.

[0021] Thirdly, the present invention provides an electronic device including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, perform the method described in the first aspect.

[0022] Fourthly, the present invention provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in the first aspect.

[0023] Fifthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0024] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0025] This disclosure proposes a method and system for predicting the severity of injuries to occupants in non-standard seating positions in smart cockpits. It employs multibody dynamics to reconstruct simulated operating conditions, enriches the dataset, and constructs a high-risk operating condition dataset. Based on this enriched high-risk operating condition dataset, intelligent identification of high-risk operating conditions and extraction of their biomechanical features are achieved. Furthermore, considering collision parameters and occupant posture parameters, a coupled simulation dataset of collision parameters and non-standard seating posture parameters is established. By fusing the standard posture dataset and the simulation dataset, a full-domain seating posture database for high-risk operating conditions is obtained. Using this full-domain seating posture database, machine learning algorithms are employed to predict the severity of occupant injuries in non-standard seating positions in smart cockpits. This achieves accurate safety assessment of non-standard seating positions in smart cockpits.

[0026] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0027] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0028] Figure 1 The flowchart of the injury severity prediction method for non-standard seating posture occupants in smart cockpits provided in Embodiment 1 of the present invention Figure 1 ;

[0029] Figure 2 This is a schematic diagram of the collision conditions of the present invention;

[0030] Figure 3 The flowchart of the injury severity prediction method for non-standard seating posture occupants in smart cockpits provided in Embodiment 1 of the present invention Figure 2 . Detailed Implementation

[0031] The application will be further described below in connection with the drawings and embodiments.

[0032] It should be noted that the following detailed description is exemplary in nature and is intended to provide further description of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0033] It is also to be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments in accordance with the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and "comprising", when used in this specification, specify the presence of stated features, integers, steps, or components, but do not preclude the presence or addition of one or more other features, integers, steps, components, or groups thereof.

[0034] The embodiments in the application and the features in the embodiments can be combined with each other without conflict.

[0035] Embodiment 1

[0036] As shown in the figure, the embodiment provides a damage severity prediction method for intelligent cockpit non-standard sitting passengers, including the following steps: Figure 1

[0037] S1, collect real traffic accident case data, and use multi-body dynamics to simulate the working condition reconstruction, enrich the data, and obtain a high-risk working condition data set;

[0038] S2, identify the accident morphology characteristics and passenger damage severity of the high-risk working condition data set, and obtain a standard posture data set under the high-risk working condition based on the accident morphology characteristics and the passenger damage severity;

[0039] S3, based on the standard posture data set, considering the collision parameters and the passenger posture parameters, obtaining a simulation data set of the coupling of the collision parameters and the non-standard sitting posture parameters under the high-risk working condition, and fusing the standard posture data set and the simulation data set to obtain a high-risk working condition global sitting posture database;

[0040] S4, training a neural network using the high-risk working condition global sitting posture database to obtain a trained prediction model, and using the trained prediction model to predict the passenger damage severity.

[0041] It also includes dividing the finally predicted passenger damage severity into risk levels according to the AIS-ISS scoring standard.​

[0042] Real traffic accident case data is collected, different boundary conditions are set, the accident form characteristics are taken as decision factors, and the occupant injury severity is taken as a decision target, and a multi-attribute decision method is used to obtain a plurality of high-risk working conditions. Due to the limitation of real traffic accident case data, a multi-body dynamics is used to simulate the working condition reconstruction, enrich the data, and construct a high-risk working condition data set.

[0043] The accident form feature recognition includes accident frequency, severity, collision direction, collision position, collision angle, collision speed, occupant injury condition and collision object.

[0044] The occupant injury severity includes personnel casualty condition and combined scoring standard (CTS).

[0045] The occupant casualty evaluation method of the abbreviated injury scale (AIS) can accurately quantify the injury degree of the occupant in the traffic accident through a systematic and standardized medical evaluation system, and provide a scientific basis for medical rescue, accident liability identification and vehicle safety performance improvement. Specifically, the method comprises the following steps:

[0046] 1) In the damage site division and grade evaluation stage, the application adopts the internationally recognized AIS standard, divides the human body into 9 main anatomical regions, including skin, head, jaw and face, neck, chest, abdomen and pelvic organs, spine, upper limbs and lower limbs. The injury degree of each part is divided into 6 grades according to the clinical severity, and the score range is from 1 to 6. Among them, 1 represents the lightest trauma; 6 represents the most serious trauma that endangers life. This grading system can comprehensively cover various trauma conditions from mild to fatal.

[0047] 2) In the expert evaluation and scoring stage, the application requires that the evaluation team is composed of clinical experts with rich trauma treatment experience. During the evaluation process, the experts need to consider many factors: first, the direct impact of the injury on the patient's vital signs; second, the degree of functional disability caused by the injury, including long-term effects such as loss of motor function and sensory impairment; third, the patient's pain degree and expected recovery situation. The evaluation experts need to combine imaging examination, laboratory test results and clinical observation, and strictly score each injury site according to the AIS scoring standard, to ensure the objectivity and consistency of the evaluation results.

[0048] 3) In the comprehensive evaluation stage, the application adopts different evaluation strategies for different injuries. For single-site injury, the AIS score of the site can directly reflect the severity of the injury. For multiple injury patients, a more scientific injury severity score (Injury Severity Score, ISS) calculation method is used. The specific ISS score calculation formula is as follows:

[0049] ;

[0050] Where AIS1, AIS2, AIS3 are the highest AIS values of the three most severe injury regions selected from all injured regions.

[0051] To unify the AIS and ISS scoring criteria, both of them are normalized to obtain the AIS-ISS combined scoring criteria (CTS), and the specific formula is as follows:

[0052] CTS = max(AIS 最高 , );

[0053] Take the highest AIS value (AIS highest, range 1-6 points), calculate the ISS standardized value (, range 0.07-5), and the comprehensive score is the larger value of the two (range 1-6 points, consistent with AIS classification).

[0054] In the CTS evaluation process, the priority principle is adopted: if any part AIS = 6 points, it is directly determined that CTS = 6 points (ISS does not need to be calculated); if ISS ≥ 75 points, even if the highest AIS ≤ 5 points, it is still determined that CTS = 5 points. The specific evaluation criteria are as follows:

[0055] CTS≤1 point; AIS highest score ≤1 point; ISS ≤8 points; determined as mild injury, only simple treatment is needed, no need for hospitalization; CTS≤2 points; AIS highest score ≤2 points; ISS ≤15 points; determined as moderate injury, no life danger; CTS≤3 points; AIS highest score ≤3 points; ISS ≤24 points; determined as more severe injury, short-term vital signs stable; CTS≤4 points; AIS highest score ≤4 points; ISS ≤49 points; determined as severe injury, life danger exists; CTS≤5 points; AIS highest score ≤5 points; ISS ≤74 points; determined as critical injury, mortality significantly increased; CTS = 6 points; AIS highest score = 6 points or ISS = 75 points; determined as extremely critical injury, survival rate is extremely low.

[0056] Quantify the occupant injury of each different standard sitting position under the accident morphology feature into different CTS scores.

[0057] Replace the traditional subjective judgment with the quantitative scoring system, which greatly improves the accuracy and comparability of the evaluation results.

[0058] Identify the accident morphology features and occupant injury severity of the high-risk working condition data set, and obtain the standard posture data set under the high-risk working condition based on the accident morphology features and occupant injury severity, which is specifically:

[0059] Identify the accident morphology characteristics and occupant injury severity of the high-risk working condition dataset, adopt CTS representation, form n samples, take the accident morphology characteristic parameters (including occupant posture parameters) of each sample as input, and take the occupant injury severity, input these data into the fitting model after quantization, and obtain a visual occupant injury severity model. The visual occupant injury severity model is used for prediction.

[0060] Based on the existing accident morphology characteristics and occupant injury severity, a standard posture dataset under high-risk working conditions is obtained. The standard posture refers to the joint parameters of the occupant being standard posture parameters.

[0061] Based on the standard posture dataset, the occupant posture parameters (i.e., the joint parameters of the occupant) are re-considered, and the simulation is re-modeled, taking the accident morphology characteristic parameters (including occupant posture parameters) of each sample as input and the injury severity as output. After quantization, input these data into the fitting model to operate, and re-obtain a visual occupant injury severity model. The re-obtained visual occupant injury severity model is used for prediction.

[0062] Based on the accident morphology characteristics and occupant injury severity, a simulation dataset of the coupling of collision parameters and non-standard posture parameters under high-risk working conditions is obtained, and a global posture database under high-risk working conditions is obtained by fusing the standard posture dataset and the simulation dataset. The non-standard posture refers to the joint parameters of the occupant being non-standard posture parameters, such as lying flat.

[0063] The coupling simulation dataset of collision parameters and non-standard posture parameters under high-risk working conditions is constructed, specifically:

[0064] 1) Consider the collision parameters, including: collision speed (20-120 km / h, with an interval of 5 km / h), collision position (front collision, offset collision, side collision, rear collision), collision angle (0°-90°, with an interval of 15°), and introduce vehicle stiffness parameters (force-displacement curves of key structures such as front longitudinal beam and door sill beam), as shown in Figure 2 .

[0065] 2) Consider the occupant posture parameters, including: seat orientation (standard / 10° reclined / 15° tilted forward), seat angle (backrest angle 15°-35°), occupant posture (standard / 20° tilted sideways / legs stretched forward / arms crossed); consider the safety belt wearing state (normal / relaxed / unfastened) and airbag triggering timing (0-50 ms delay) as shown in Figure 2 .

[0066] 3) Based on orthogonal experimental design or Latin hypercube sampling method, generate not less than 1000 groups of parameter combinations to ensure that all working conditions are covered as much as possible.

[0067] 4) Use existing software to establish a dummy model to simulate the collision scene under each parameter combination;

[0068] 5) Output key injury indicators such as occupant head injury criterion (HIC) and thoracic compression (ThPC).

[0069] 6) Normalize the different evaluation criteria of the nine parts of AIS and output the comprehensive evaluation score, which is the same as the CTS evaluation level of the first aspect.

[0070] 7) Output the comprehensive evaluation score of 1000 different combinations of collision parameters and occupant posture parameters, which is beneficial to the establishment of subsequent data sets.

[0071] Expand the database to a sufficient number. Fuse the standard sitting posture data set (i.e. the standard posture data set under high-risk working conditions) with the non-standard sitting posture data set (i.e. the simulation data set of coupling collision parameters and non-standard posture parameters under high-risk working conditions) to form a multi-dimensional injury database (i.e. a global sitting posture database under high-risk working conditions).

[0072] Through systematic parameter combination and large-scale simulation, the coverage of the non-standard posture injury model is significantly improved; fusion of simulation and real data solves the problem of insufficient actual samples; the accuracy of the subsequent neural network prediction model for occupant injury severity in the cabin is significantly improved, and quantitative evaluation of occupant collision injury under non-standard posture is realized.

[0073] Through the combination of deep learning, data-driven and evolutionary computing, an efficient occupant injury prediction algorithm is proposed, a mapping model between each collision parameter and occupant injury is constructed, and the accuracy requirement is met.

[0074] Further, to improve the diversity and coverage of the data set, the present application uses data enhancement technology, including applying Gaussian noise disturbance to the original collision parameters to generate multiple near-neighbor samples, and using a generative adversarial network (GAN) to generate a model to enhance the coupling simulation data set of collision parameters and standard and non-standard posture parameters under high-risk working conditions, expand the size of the training data set, and improve the adaptability of the model to different collision scenarios.

[0075] As shown in Figure 3 , the data enhancement method includes: applying Gaussian noise disturbance to the original collision parameters to generate multiple near-neighbor samples, selecting a generative adversarial network as a data derivation matrix, and the inputs of the generator (G) and the discriminator (D) will contain class labels and standard and non-standard posture type data. Based on the original data set, the features of the data sources in the original data set are combined and operated by the generative adversarial network to complete the enhancement of the training data set.

[0076] Further, a deep neural network-based occupant injury prediction model is constructed using the above data set processed by data augmentation. This model uses a fully connected neural network to effectively learn the complex nonlinear mapping between collision parameters and occupant injury indicators, and thus achieve high-precision prediction of occupant injury.

[0077] Further, evolutionary computation methods such as genetic algorithm (GA) or particle swarm optimization (PSO) are used to optimize the hyperparameters of the deep learning model, such as learning rate, number of hidden layer nodes, regularization coefficient, network structure, etc. Through iterative search for the optimal hyperparameter combination, the prediction accuracy of the model is improved.

[0078] Further, the training data is divided into training set and validation set, and the training set is used to train the model, and the validation set is used to evaluate the model performance. Cross-validation and other methods are used to ensure the stability and generalization ability of the model.

[0079] Further, through the trained deep learning model, the mapping relationship between collision parameters and occupant injury is established. This model can quickly and accurately predict occupant injury based on input collision parameters, thereby providing data support for vehicle safety design, crash testing and intelligent simulation.

[0080] The training model includes a fully connected neural network, in which each neuron is connected to all neurons of the previous and next layers, forming a dense connection structure. Collision parameters (collision speed, collision position and collision angle) and occupant posture parameters (seat orientation, seat angle and member sitting posture) are used as Figure 3 inputs of the fully connected neural network, and the fully connected neural network outputs the member injury severity of the high-risk working condition intelligent cabin non-standard sitting posture.

[0081] This example can construct a high-risk working condition data set under an intelligent cabin non-standard sitting posture based on real traffic accident cases and simulation data in China, and establish an occupant injury severity prediction model through multidisciplinary fusion algorithms, with collision severity, injury risk and protection effectiveness as targets, detect and evaluate occupant injury indicators under different sitting posture conditions, realize data modeling, analysis and prediction of the whole cycle of collision safety performance, and complete safety evaluation and protection system optimization tasks.

[0082] Specifically, it includes:

[0083] In the first aspect, this example can collect data and extract features of the non-standard sitting posture to be evaluated, with collision parameters, sitting posture parameters and injury indicators as targets, and establish a typical high-risk working condition library and an occupant injury severity model through statistical analysis and intelligent optimization algorithms, and quantitatively evaluate the injury risk under different working conditions.

[0084] In a second aspect, the present example can construct a coupling simulation dataset of collision parameters and non-standard sitting posture parameters, and establish a sitting posture-collision condition mapping relationship model through multi-body dynamics simulation and finite element analysis, and accurately calibrate the damage threshold under each condition, aiming at the biomechanical response, damage mechanism and protection strategy.

[0085] In a third aspect, the present example can propose an efficient occupant injury prediction algorithm by combining deep learning, data-driven and evolutionary computation, and establish an intelligent mapping model between collision parameters and occupant injury, aiming at prediction accuracy, computational efficiency and generalization ability, and meeting the precision requirements of engineering applications.

[0086] Embodiment 2

[0087] The present embodiment provides a damage severity prediction system for intelligent cockpit non-standard sitting posture occupants, comprising:

[0088] The data acquisition module is configured to collect real traffic accident case data, and to enrich the data by simulating the working conditions using multi-body dynamics to obtain a high-risk working condition dataset;

[0089] The standard posture dataset construction module is configured to identify the accident morphology features and occupant injury severity of the high-risk working condition dataset, and to obtain a standard posture dataset under high-risk working conditions based on the accident morphology features and occupant injury severity;

[0090] The global sitting posture database construction module is configured to obtain a simulation dataset of coupling of collision parameters and non-standard sitting posture parameters under high-risk working conditions based on the standard posture dataset and considering the collision parameters and occupant posture parameters, and to obtain a global sitting posture database under high-risk working conditions by fusing the standard posture dataset and the simulation dataset;

[0091] The model training and prediction module is configured to train a neural network using the global sitting posture database under high-risk working conditions to obtain a trained prediction model, and to use the trained prediction model to predict the occupant injury severity.

[0092] It should be noted that the above modules correspond to the steps described in Embodiment 1, and the above modules have the same examples and application scenarios as the corresponding steps, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules as part of the system can be executed in a computer system such as a set of computer executable instructions.

[0093] In more embodiments, the following are also provided:

[0094] An electronic device includes a memory and a processor and computer instructions stored on the memory and running on the processor, when the computer instructions are run by the processor, the method described in embodiment 1 is completed. For the sake of brevity, it will not be repeated here.

[0095] It should be understood that in the embodiments, the processor can be a central processing unit CPU, and the processor can also be other general-purpose processors, digital signal processors DSPs, application-specific integrated circuits ASICs, ready-to-program gate arrays FPGA or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0096] The memory can include read-only memory and random access memory, and provide instructions and data to the processor, and a part of the memory can also include non-volatile random access memory. For example, the memory can also store device type information.

[0097] A computer readable storage medium for storing computer instructions, when the computer instructions are executed by the processor, the method described in embodiment 1 is completed.

[0098] The method in embodiment 1 can be directly embodied as hardware processor execution completion, or executed by a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, register, etc. The storage medium is located in the memory, and the processor reads the information in the memory, and combines the hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.

[0099] A computer program product includes a computer program, which is executed by the processor to realize the method described in embodiment 1.

[0100] The present application also provides at least one computer program product tangibly stored on a non-transitory computer readable storage medium. The computer program product includes computer executable instructions, such as instructions included in program modules, which are executed in devices on real or virtual processors of targets to perform processes / methods as described above. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functions of the program modules can be combined or divided as needed among the program modules. Machine executable instructions for program modules can be executed within local or distributed devices. In distributed devices, program modules can be located in local and remote storage media.

[0101] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages. The computer program code can execute entirely on a computer, a special purpose computer, or other programmable apparatus to produce the functions / acts specified in the flow diagrams and / or block diagrams. The program code can execute entirely on a computer, a special purpose computer, or other programmable apparatus, as a stand-alone software package, partly on the computer and partly on a remote computer, or entirely on the remote computer or server.

[0102] In the context of the present application, the computer program code or related data can be carried by any suitable carrier to enable the device, apparatus or processor to perform the various processes and operations described above. Examples of carriers include signals, computer readable media, and the like. Examples of signals can include electrical, optical, radio, sound or other forms of propagated signals, such as carrier waves, infrared signals, and the like.

[0103] Those skilled in the art can understand that the units and algorithm steps of the examples described in conjunction with the embodiments can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software manner depends on the specific application and design constraints of the technical solutions. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0104] Although the specific embodiments of the present application have been described in conjunction with the accompanying drawings, the description is not intended to limit the scope of the present application. Those skilled in the art should understand that various modifications or variations can be made to the technical solutions of the present application without departing from the scope of the present application, and such modifications or variations are still within the scope of the present application.

Claims

1. A method for predicting injury severity of a non-standard sitting occupant in an intelligent cockpit, characterized in that, The method comprises the following steps: Collect real traffic accident case data, and use multi-body dynamics to simulate the working condition reconstruction, enrich the data, and obtain a high-risk working condition data set; Identify the accident morphology characteristics and occupant injury severity of the high-risk working condition data set, and obtain a standard posture data set under the high-risk working condition based on the accident morphology characteristics and occupant injury severity; Based on the standard posture data set, considering the collision parameters and occupant posture parameters, a simulation data set of the coupling of the collision parameters and the non-standard sitting posture parameters under the high-risk working condition is obtained, the standard posture data set and the simulation data set are fused, and a high-risk working condition global sitting posture database is obtained; The high-risk working condition global sitting posture database is used to train a neural network, and a trained prediction model is obtained, and the trained prediction model is used for occupant injury severity prediction; The coupling simulation data set of the collision parameters and the non-standard sitting posture parameters under the high-risk working condition is constructed, specifically: The collision parameters are considered, including: collision speed, collision position, collision angle and vehicle stiffness parameter; The occupant posture parameters are considered, including: seat orientation, seat angle, occupant sitting posture; considering the safety belt wearing state and airbag triggering time sequence; Based on the orthogonal test design or Latin hypercube sampling method, not less than 1000 groups of parameter combinations are generated to ensure that all working conditions are covered as much as possible; Existing software is used to establish a dummy model to simulate the collision scene under each parameter combination; Output the key injury indicators such as occupant head injury criteria and chest compression amount; The different evaluation criteria of the nine parts of AIS are normalized and output to obtain a comprehensive evaluation score, which is the same as the CTS evaluation level of the first aspect; The comprehensive evaluation score of 1000 different collision parameter and occupant posture parameter combinations is output, which is beneficial to the establishment of subsequent data sets; Expand to a sufficient number of databases; the standard posture data set under the high-risk working condition and the simulation data set of the coupling of the collision parameters and the non-standard sitting posture parameters under the high-risk working condition are fused to form a multi-dimensional injury database. 2.The smart cockpit facing non-standard sitting posture occupant injury severity prediction method of claim 1, wherein, It also includes dividing the finally predicted occupant injury severity into risk levels according to the AIS-ISS scoring standard. 3.The smart cockpit facing non-standard sitting posture occupant injury severity prediction method of claim 1, wherein, The accident morphology characteristics include accident frequency, severity, collision direction, collision position, collision angle, collision speed, occupant injury, and collision object. 4.The smart cockpit facing non-standard sitting posture occupant injury severity prediction method of claim 1, wherein, The occupant injury severity includes personnel casualty and abbreviated injury scale.

5. The smart cockpit facing non-standard sitting posture occupant injury severity prediction method of claim 1, wherein, The collision parameters include collision speed, collision position, collision angle, and vehicle stiffness parameter.

6. The smart cockpit facing non-standard sitting posture occupant injury severity prediction method of claim 1, wherein The occupant posture parameters include seat orientation, seat angle, occupant sitting posture, safety belt wearing state, and airbag triggering time sequence.

7. A system for predicting injury severity of a non-standard seated occupant in an intelligent cabin, characterized in that, It comprises: The data acquisition module is configured to collect real traffic accident case data, and use multi-body dynamics to simulate the working condition reconstruction, enrich the data, and obtain a high-risk working condition data set; The standard posture data set construction module is configured to identify the accident morphology characteristics and occupant injury severity of the high-risk working condition data set, and obtain a standard posture data set under the high-risk working condition based on the accident morphology characteristics and occupant injury severity; The global sitting posture database construction module is configured to obtain a simulation data set of coupling of the collision parameter and the non-standard sitting posture parameter under the high-risk working condition based on the standard posture data set, considering the collision parameter and the occupant posture parameter, fuse the standard posture data set and the simulation data set, and obtain the global sitting posture database under the high-risk working condition; The model training and prediction module is configured to train the neural network by using the global sitting posture database under the high-risk working condition, obtain a trained prediction model, and perform occupant injury severity prediction by using the trained prediction model; The coupling simulation data set of the collision parameter and the non-standard sitting posture parameter under the high-risk working condition is constructed, and specifically: The collision parameter is considered, including collision speed, collision position, collision angle, and vehicle stiffness parameter is introduced; The occupant posture parameter is considered, including seat orientation, seat angle, occupant sitting posture, safety belt wearing state, and airbag triggering time sequence; Based on the orthogonal test design or Latin hypercube sampling method, not less than 1000 groups of parameter combinations are generated to ensure that all working conditions are covered as much as possible; An existing software is used to establish a dummy model to simulate the collision scene under each parameter combination; Key injury indicators such as occupant head injury criteria and chest compression are output; Different evaluation criteria of the nine parts of AIS are normalized to output a comprehensive evaluation score, which is the same as the CTS evaluation level of the first aspect; The comprehensive evaluation score of 1000 groups of different collision parameter and occupant posture parameter combinations is output, which is beneficial to the establishment of subsequent data sets; The database is expanded to a sufficient number; the standard posture data set under the high-risk working condition and the simulation data set of coupling of the collision parameter and the non-standard sitting posture parameter under the high-risk working condition are fused to form a multi-dimensional injury database.

8. An electronic device, comprising: The computer program product comprises a memory and a processor, and computer instructions stored in the memory and run on the processor, when the computer instructions are run by the processor, the method of any one of claims 1-6 is completed.

9. A computer-readable storage medium, characterized in that, A computer program product for storing computer instructions, when the computer instructions are executed by a processor, the method of any one of claims 1-6 is completed.

10. A computer program product, characterised in that, A computer program product for storing computer instructions, when the computer instructions are executed by a processor, the method of any one of claims 1-6 is completed.

Images