A new energy automobile parking space safety early warning system

Abstract

The present application relates to the technical field of automobile safety warning, in particular to a new energy automobile parking space safety warning system. It comprises: an environment perception unit, which collects parking space surrounding environment data in real time; a positioning and navigation unit, which obtains the relative position and dynamic information of the vehicle and the parking space based on the surrounding environment data; a data processing and analysis unit, which analyzes the information obtained by the positioning and navigation unit, and analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model. The optimization of the slope environment and the vehicle attitude is introduced, the optimized kernel function adds the adaptability function of the slope inclination angle and the vehicle pitch angle, which can accurately calculate the distance between the vehicle and the parking space, considering the attitude change caused by the slope, thereby improving the adaptability of the model to the slope parking environment and ensuring the safety warning ability of the system when parking on the slope.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automobile safety warning technology, in particular, to a new energy automobile parking space safety warning system. BACKGROUND

[0002] With the popularization of new energy vehicles (such as electric vehicles), the demand for parking spaces not only continues to grow in quantity, but also needs to meet the requirements of charging, intelligent management, diversified use scenarios, etc. The new energy vehicle parking space safety warning system uses various sensors (such as cameras, radars, ultrasonic sensors, etc.) to monitor the environment around the vehicle in real time, and combines positioning and navigation technology (such as GPS, IMU inertial measurement unit), as well as data analysis and machine learning algorithms, to assess potential risks during parking and timely warn the driver or take automatic measures to avoid collisions or other accidents, as part of the advanced driver assistance system (ADAS), the parking space safety warning system provides important support for achieving higher levels of automated driving, especially in the last kilometer of low-speed driving scenarios.

[0003] For users, the new energy vehicle parking space safety warning system can assist the car owner to park more accurately in a small or complex layout of the home garage, reducing the risk of scratches and other damage; however, in a complex or non-standard parking space, especially in a parking environment with dynamic obstacles or slopes, the vehicle needs to overcome the influence of gravity and accurately control the braking and driving force, and the wrong perception of sliding or attitude angle on the slope will increase the difficulty of parking, therefore, a new energy automobile parking space safety warning system is designed. SUMMARY

[0004] The purpose of the present application is to provide a new energy automobile parking space safety warning system to solve the problem of overcoming the influence of gravity and accurately controlling the braking and driving force in a complex or non-standard parking space, especially in a parking environment with dynamic obstacles or slopes, and the wrong perception of sliding or attitude angle on the slope will increase the difficulty of parking.

[0005] To achieve the above purpose, the present application provides a new energy automobile parking space safety warning system, comprising:

[0006] An environment perception unit, which acquires parking space surrounding environment data in real time;

[0007] A positioning and navigation unit, which obtains the relative position and dynamic information of the vehicle and the parking space based on the surrounding environment data;

[0008] A data processing and analysis unit analyzes the information acquired by the positioning and navigation unit, and analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model.

[0009] The decision-making and warning unit performs risk assessments and triggers early warning mechanisms based on the analysis results.

[0010] As a further improvement to this technical solution, the environmental data surrounding the parking space includes the location of the parking space markings. , The relative position of parking spaces , .

[0011] As a further improvement to this technical solution, the positioning and navigation unit (2) acquires the relative position and dynamic information of the vehicle and parking space based on surrounding environmental data, including the following steps:

[0012] S1.1 Preprocess the collected environmental data around the parking space and collect the original coordinates of environmental information points through sensors. ;

[0013] S1.2 Based on the vehicle's own positioning sensors and environmental data, based on the original coordinates Determine the vehicle's current absolute position. and attitude angle , ,in, Indicates that the vehicle is in velocity components in the direction, Indicates that the vehicle is in Velocity component in the direction;

[0014] S1.3 Calculate the position of the parking space markings and the current absolute position of the vehicle. relative position : ;

[0015] S1.4 Calculate the vehicle speed using vehicle sensor data. and acceleration ;

[0016] S1.5, Based on the absolute position of the parking space Calculate the relative position of the vehicle and the parking space. : .

[0017] As a further improvement to this technical solution, the data processing and analysis unit includes a data fusion processing module and an intelligent algorithm module;

[0018] The data fusion processing module integrates the relative position of the vehicle and the parking space output by the positioning and navigation unit , the relative position of the parking space , vehicle dynamic information and environmental data to form a surrounding environment model

[0019] The intelligent algorithm module analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model

[0020] As a further improvement of the technical solution, the data fusion processing module integrates the relative position of the vehicle and the parking space output by the positioning and navigation unit , the relative position of the parking space , vehicle dynamic information and environmental data to form a surrounding environment model, including the following steps:

[0021] S2.1, basic processing is performed on the input relative position of the vehicle and the parking space , the relative position of the parking space , vehicle dynamic information and environmental data

[0022] S2.2, convert sensor data to a unified coordinate system : , wherein represents a rotation matrix, represents a translation vector

[0023] S2.3, the detection results of the same target are fused by weighted average

[0024] S2.4, a surrounding environment model is established based on the fused data

[0025] As a further improvement of the technical solution, in S2.4, a surrounding environment model is established based on the fused data, including the following steps:

[0026] S2.41, a vehicle position model is constructed ;

[0027] S2.42, a vehicle direction model is constructed : , wherein represents the angular velocity of the vehicle

[0028] S2.43, a relative position model of the vehicle and the parking space is constructed , wherein represents the position of the parking space

[0029] S2.44, the above models are integrated to form a surrounding environment model

[0030] As a further improvement of the technical solution, in S2.44, the surrounding environment model is:

[0031] ;

[0032] wherein, represents the surrounding environment anomaly evaluation model; represents the relative speed difference between the vehicle and the parking space in time ; represents the weight coefficient of the relative position; represents the weight coefficient of the vehicle heading angle; represents the weight coefficient of the relative position; represents the geometric relationship between the vehicle heading angle and the parking space;

[0033] Optimize the surrounding environment model for the vehicle's motion trajectory and the relative angle between the vehicle and the parking space:

[0034] ;

[0035] wherein, represents the optimized surrounding environment anomaly evaluation model; represents the relative angular velocity change, , represents the relative heading angle change between the parking space and the vehicle; represents the weight coefficient of the relative angular velocity change;

[0036] Introduce the vehicle's future motion trajectory prediction model into the optimized surrounding environment model:

[0037] ;

[0038] ;

[0039] wherein, represents the surrounding environment anomaly evaluation model after introducing the vehicle's future motion trajectory prediction model; represents the vehicle motion trajectory prediction model.

[0040] As a further improvement of the technical solution, the intelligent algorithm module analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model through the support vector machine model, including the following steps:

[0041] S3.1, the established surrounding environment model is input as an input feature, the input feature includes the relative position, speed, acceleration, heading angle of the vehicle, etc., and the input feature is input as a data set;

[0042] S3.2, use the dataset to train the support vector machine model, and use the kernel function to adjust the hyperparameters of the support vector machine model;

[0043] S3.3, using the trained support vector machine model to analyze the real-time collected parking space surrounding environment data;

[0044] S3.4, according to the output of the support vector machine, the abnormal state evaluation value of the vehicle and the parking space surrounding environment is obtained.

[0045] As a further improvement of the technical solution, in the S3.2, the kernel function is:

[0046] ;

[0047] Wherein, The distance between the vehicle and the parking space is represented by d; The state feature vector of the vehicle is represented by x; The state feature vector of the parking space is represented by y; The scale parameter in the kernel function is represented by σ;

[0048] Optimize the kernel function for the influence of the ramp parking environment on the vehicle posture:

[0049] ;

[0050] ;

[0051] ;

[0052] Wherein, The distance between the vehicle and the parking space after optimization is represented by d; The terrain adaptability function is represented by f, which considers the influence of the slope angle on the relative position of the vehicle and the parking space; The vehicle posture function is represented by g, which considers the influence of the pitch angle of the vehicle on the relative position of the vehicle and the parking space; The slope angle is represented by θ; The pitch angle of the vehicle is represented by φ; The weight coefficient for adjusting the influence of the terrain on the kernel function is represented by w; The weight coefficient for controlling the influence of the pitch angle on the kernel function is represented by w.

[0053] As a further improvement of the technical solution, the decision and warning unit performs risk assessment and triggers the early warning mechanism according to the analysis results, including the following steps:

[0054] S4.1, receiving the abnormal evaluation results from the data processing and analysis unit;

[0055] S4.2. Set medium-risk and high-risk levels to determine whether there are safety hazards in the environment around the parking space. Low risk.

[0056] S4.3. Trigger a warning when there is a medium risk in the environment around the parking space, and trigger an emergency warning when there is a high risk.

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0058] 1. In this new energy vehicle parking space safety early warning system, the traditional support vector machine model is optimized for slope environment and vehicle attitude. In order to cope with the slope environment, the optimized kernel function adds an adaptive function for slope tilt angle and vehicle pitch angle, which can more accurately calculate the distance between the vehicle and the parking space. Taking into account the attitude changes caused by the slope, the model's adaptability to the slope parking environment is improved, ensuring the system's safety early warning capability when parking on a slope.

[0059] 2. In this new energy vehicle parking space safety early warning system, by introducing a vehicle future motion trajectory prediction model, the system can predict the possible position and state of the vehicle in the future based on the vehicle's current position, speed, acceleration and other information. For parking on a slope, the system can predict the vehicle's deviation caused by changes in the slope terrain and identify potential risks in a timely manner, such as predicting whether the distance between the vehicle and the parking space boundary is too close, thereby issuing a warning in advance or triggering automatic braking.

[0060] Predicting future movement trajectories helps the system accurately predict the vehicle's movement on a slope as it approaches a parking space, thus avoiding accidental collisions or parking failures caused by changes in the slope angle. Attached Figure Description

[0061] Figure 1 This is an overall flowchart of the present invention;

[0062] The meanings of the labels in the diagram are as follows:

[0063] 1. Environmental sensing unit;

[0064] 2. Positioning and navigation unit;

[0065] 3. Data processing and analysis unit; 31. Data fusion processing module; 32. Intelligent algorithm module;

[0066] 4. Decision-making and warning unit. Detailed Implementation

[0067] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0068] Embodiment: Please refer to Figure 1 As shown in the figure, a new energy vehicle parking space safety warning system is provided, which comprises an environment perception unit 1, and the environment perception unit 1 collects parking space surrounding environment data in real time.

[0069] The parking space surrounding environment data comprises a marking line position of the parking space (i.e. the boundary or center position of the parking space), a relative position of the parking space (i.e. the specific position of the parking space in or around the parking space), .

[0070] In the embodiment, the new energy vehicle parking space safety warning system further comprises a positioning and navigation unit 2, and the positioning and navigation unit 2 obtains the relative position and dynamic information (attitude, speed, acceleration) of the vehicle and the parking space based on the surrounding environment data.

[0071] The positioning and navigation unit 2 obtains the relative position and dynamic information (attitude, speed, acceleration) of the vehicle and the parking space based on the surrounding environment data, comprising the following steps:

[0072] S1.1, pre-process the collected parking space surrounding environment data, and collect the original coordinates of the environment information points through the sensor The original coordinates are the position coordinates of the environment information points collected by the sensor, which represent the original data in the local coordinate system of the sensor.

[0073] S1.2, based on the self-positioning sensor (such as GPS, IMU) of the vehicle and the environment data, determine the current absolute position of the vehicle and the attitude angle based on the original coordinates , wherein, represents the velocity component of the vehicle in the direction, represents the velocity component of the vehicle in the direction.

[0074] S1.3, calculate the relative position between the marking line position of the parking space and the current absolute position of the vehicle: ;

[0075] S1.4, calculating the speed of the vehicle by the sensor data of the vehicle and acceleration ;

[0076] S1.5, calculating the relative position of the vehicle to the parking space based on the absolute position of the parking space : .

[0077] The new energy vehicle parking safety warning system further comprises a data processing and analysis unit 3, which analyzes the information obtained by the positioning and navigation unit 2, and analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model;

[0078] The data processing and analysis unit 3 comprises a data fusion processing module 31 and an intelligent algorithm module 32.

[0079] The data fusion processing module 31 integrates the relative position of the vehicle to the parking space, the relative position of the parking space, the vehicle dynamic information (attitude, speed, acceleration) and the environmental data output by the positioning and navigation unit 2 to form a surrounding environment model. The intelligent algorithm module 32 analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model.

[0080] Further, the data fusion processing module 31 integrates the relative position of the vehicle to the parking space, the relative position of the parking space, the vehicle dynamic information (attitude, speed, acceleration) and the environmental data output by the positioning and navigation unit 2 to form a surrounding environment model, comprising the following steps:

[0081] S2.1, performing basic processing on the input relative position of the vehicle to the parking space, the relative position of the parking space, the vehicle dynamic information (attitude, speed, acceleration) and the environmental data, the basic processing comprising time synchronization (synchronizing the data collected by the environmental perception unit and the positioning and navigation unit according to the time stamp to ensure data consistency), unit and format unification (converting data from different sources to a unified coordinate system, unit (such as metric) and data format), and abnormality detection and elimination (checking for and eliminating or repairing abnormal values (such as sensor noise or invalid data) in the data); S2.2, converting the sensor data to a unified coordinate system

[0082] S2.1, performing basic processing on the input relative position of the vehicle to the parking space, the relative position of the parking space, the vehicle dynamic information (attitude, speed, acceleration) and the environmental data, the basic processing comprising time synchronization (synchronizing the data collected by the environmental perception unit and the positioning and navigation unit according to the time stamp to ensure data consistency), unit and format unification (converting data from different sources to a unified coordinate system, unit (such as metric) and data format), and abnormality detection and elimination (checking for and eliminating or repairing abnormal values (such as sensor noise or invalid data) in the data); S2.2, converting the sensor data to a unified coordinate system

[0083] S2.2, converting the sensor data to a unified coordinate system : ,​​​​ Represents the rotation matrix. Represents the translation vector;

[0084] S2.3 Detection results for the same target are fused using a weighted average:

[0085] S2.4. Based on the fused data, establish a surrounding environment model, which is used to reflect the dynamic and static information around the vehicle.

[0086] In S2.4 of this embodiment, the surrounding environment model is established based on the fused data, including the following steps:

[0087] S2.41 Constructing a vehicle position model ;

[0088] S2.42, Constructing the vehicle orientation model : ,in, Indicates the angular velocity of the vehicle;

[0089] S2.43. Construct a relative position model of vehicles and parking spaces: ,in, Indicates the location of the parking space;

[0090] S2.44. Integrate the above models to form a surrounding environment model. The surrounding environment model is used to perceive, analyze and predict static and dynamic information in parking spaces and their surrounding environment in real time, providing data support for safety early warning and auxiliary decision-making.

[0091] In S2.44 of this embodiment, the surrounding environment model is as follows:

[0092] ;

[0093] in, This represents an assessment model for anomalies in the surrounding environment. This indicates the time of the vehicle and parking space. The relative speed difference; Weighting coefficients representing relative positions; The weighting coefficient representing the vehicle's heading angle; Weighting coefficients representing relative positions; This indicates the geometric relationship between the vehicle's heading angle and the parking space. It calculates the angle between the vehicle's orientation and the parking space markings. If the angle between the vehicle and the parking space deviates significantly (i.e., the vehicle may deviate from the parking space), the risk is greater.

[0094] The surrounding environment model is optimized based on the vehicle's trajectory and the relative angle between the vehicle and the parking space.

[0095] By analyzing the real driving path of the vehicle, it can be more accurately understood how the vehicle moves to the current position, which is crucial for identifying abnormal driving behaviors (such as sudden turns or sudden braking) that may indicate impending risks; the relative angle between the vehicle and the parking space reflects the way the vehicle enters the parking space, and if the angle is not appropriate, it may cause parking difficulties or increased collision risks. Therefore, accurately capturing this angle can help the system better assess whether the current parking operation is safe; by combining the actual motion trajectory of the vehicle and the relative angle, the system can build a more detailed surrounding environment model, which not only includes static information (such as parking space boundaries), but also includes dynamic changes (such as how the vehicle approaches the parking space), such a model can provide more rich context information for risk assessment, making the decision-making process more intelligent; different drivers have different driving habits, and different parking spaces have different geometric characteristics, by considering the actual motion trajectory and the relative angle, the system can better adapt to various parking scenarios, even in complex or non-standard parking spaces, it can maintain high performance and reliability;

[0096] ;

[0097] wherein, represents the optimized surrounding environment anomaly evaluation model; represents the relative angular velocity change, , represents the relative heading angle change between the parking space and the vehicle; represents the weight coefficient of the relative angular velocity change;

[0098] In the optimized surrounding environment model, a future motion trajectory prediction model of the vehicle is introduced: ;

[0099] ;

[0100] wherein, represents the surrounding environment anomaly evaluation model after introducing the future motion trajectory prediction model of the vehicle; represents the vehicle motion trajectory prediction model.

[0101] By predicting the possible position and state of the vehicle in the future, the system can identify potential risks before an accident occurs. The system can predict the possible position and state of the vehicle in the future based on the current position, speed, acceleration, etc. of the vehicle. For hill parking, the system can predict the deviation of the vehicle due to the change of the hill terrain, identify potential risks in time, such as predicting whether the distance between the vehicle and the parking boundary is too close, and thus issuing a warning or triggering an automatic brake in advance; for example, if the system predicts that the vehicle will collide with an obstacle or deviate from the parking space, it can issue an early warning to give the driver enough time to take corrective action; the prediction model provides important information input for the system, based on which the system can make more intelligent decisions, such as adjusting the speed, direction or starting the emergency brake of the vehicle, to ensure safe parking; the introduction of future motion trajectory prediction helps the system better adapt to the dynamically changing parking environment, even if there are moving objects or other uncertain factors around, the system can adjust the strategy in time according to the latest prediction results, and maintain high safety and reliability;

[0102] The intelligent algorithm module 32 analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model through the support vector machine model, including the following steps:

[0103] S3.1, input the established surrounding environment model as input features, which include the relative position, speed, acceleration, heading angle, etc. of the vehicle, and input the input features as a data set;

[0104] S3.2, use the data set to train the support vector machine (SVM) model, which can handle nonlinear problems (such as the complex relationship between the vehicle and the parking space angle, position, etc.) by mapping data to a high-dimensional feature space, and use the kernel function (RBF) to adjust the hyperparameters (such as the penalty coefficient and the parameters of the kernel function) of the support vector machine (SVM) model to improve the classification accuracy;

[0105] S3.3, use the trained support vector machine model to analyze the real-time collected parking space surrounding environment data, and the SVM model will output the prediction result (normal or abnormal) according to the input surrounding environment data;

[0106] S3.4, according to the output of the support vector machine, get the abnormal state evaluation value of the vehicle and the parking space surrounding environment.

[0107] Wherein, in the S3.2, the kernel function is:

[0108] ;

[0109] Wherein, represents the distance between the vehicle and the parking space; State feature vector representing the vehicle (position, velocity, acceleration, attitude, etc.); State feature vector representing the parking space (marking position of the parking space, relative position of the parking space, position of obstacles near the parking space, etc.); Scale parameter in the kernel function, controlling the sensitivity of similarity calculation between the vehicle and the parking space;

[0110] Optimizing the kernel function for the impact of hill parking environment on vehicle attitude:

[0111] In the hill parking environment, the attitude of the vehicle (such as the pitch angle) will change due to the ground slope, which directly affects the relative position and angle relationship between the vehicle and the parking space; the pitch angle of the vehicle will affect the perception range and accuracy of its sensors, for example, when going uphill or downhill, the distance sensing in front and behind the vehicle may be affected by the change of attitude; in the hill environment, the traditional kernel function may not fully consider the impact of vehicle attitude and terrain, resulting in inaccurate risk assessment, by optimizing the kernel function, the influence of these factors on vehicle behavior can be captured more finely, so as to make more reasonable risk prediction; in some extreme cases, such as steep hill parking, the attitude change of the vehicle may be very significant, the optimized kernel function can ensure that the system still maintains high accuracy and response speed in such cases, avoiding potential safety hazards;

[0112] ;

[0113] ;

[0114] ;

[0115] wherein, represents the distance between the optimized vehicle and the parking space; represents the terrain adaptability function, considering the impact of the slope angle of the hill on the relative position of the vehicle and the parking space, correcting the relative position error caused by the terrain slope; represents the vehicle attitude function, considering the impact of the pitch angle of the vehicle on the relative position of the vehicle and the parking space; represents the slope angle; represents the pitch angle of the vehicle; represents the weight coefficient adjusting the influence intensity of the terrain on the kernel function; represents the weight coefficient controlling the influence of the pitch angle on the kernel function.

[0116] The new energy vehicle parking space safety warning system further comprises a decision and warning unit 4, which performs risk assessment and triggers the warning mechanism according to the analysis result;

[0117] In the present embodiment, the decision and warning unit 4 performs risk assessment according to the analysis results and triggers the early warning mechanism, including the following steps:

[0118] S4.1, receiving the abnormality assessment results from the data processing and analysis unit 3;

[0119] S4.2, setting risk levels to determine whether there is a security risk in the parking space environment; according to the relative position and dynamic information (such as speed, acceleration, attitude, etc.) of the vehicle and the parking space to determine whether it is in the normal range;

[0120] Among them, the medium risk (the dynamic information of the vehicle shows some unusual trends, such as the vehicle deviating from the parking space, the speed being too fast or the acceleration being abnormal);

[0121] High risk (the relative position of the vehicle and the parking space deviates significantly, there is a collision danger or the distance between the obstacles around the parking space is too close, and the speed, angle change of the vehicle is abnormal);

[0122] S4.3, triggering a warning reminder (prompting the driver: through the vehicle display screen, sound system or mobile phone application, reminding the driver to pay attention to the safety of the parking space; displaying warning information: such as "the parking space is slightly deviated, please operate carefully") when there is a medium risk in the parking space environment, and triggering an emergency warning (sending an emergency alarm sound; the automatic parking assistance system starts, automatically adjusts the vehicle speed, lane deviation, etc., to avoid collision; send a notification: notify the vehicle owner, parking management personnel or automatic system, so as to handle in time) when there is a high risk.

[0123] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only preferred examples of the present application and are not intended to limit the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application.

Claims

1. A new energy vehicle parking space safety early warning system, characterized in that, Comprise: An environment perception unit (1) that collects parking space surrounding environment data in real time; A positioning and navigation unit (2) that obtains the relative position and dynamic information of the vehicle and the parking space based on the surrounding environment data; A data processing and analysis unit (3) that analyzes the information obtained by the positioning and navigation unit (2), and analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model; A decision and warning unit (4) that performs risk assessment and triggers the early warning mechanism according to the analysis result; The data processing and analysis unit (3) comprises a data fusion processing module (31) and an intelligent algorithm module (32); The data fusion processing module (31) integrates the relative position of the vehicle and the parking space output by the positioning and navigation unit (2) , the relative position of the parking space , the vehicle dynamic information and the environmental data to form a surrounding environment model; The intelligent algorithm module (32) analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model; The intelligent algorithm module (32) analyzes the abnormal state of the vehicle around the parking space based on the surrounding environment model through the support vector machine model, comprising the following steps: S3.1, taking the established surrounding environment model as the input feature, and taking the input feature as the data set; S3.2, using the data set to train the support vector machine model, and using the kernel function to adjust the hyperparameters of the support vector machine model; S3.3, using the trained support vector machine model to analyze the real-time collected parking space surrounding environment data; S3.4, according to the output of the support vector machine, the abnormal state evaluation value of the vehicle and the parking space surrounding environment is obtained; In the S3.2, the kernel function is: ； wherein, denotes the distance between the vehicle and the parking space; denotes the state feature vector of the vehicle; denotes the state feature vector of the parking space; denotes a scale parameter in the kernel function; Optimize the kernel function for the influence of the ramp parking environment on the vehicle posture: ； ； ； wherein, represents the distance between the vehicle and the parking space after optimization; represents a terrain adaptability function; represents a vehicle posture function; represents a ramp angle; represents a pitch angle of the vehicle; represents a weight coefficient for adjusting the influence strength of the terrain on the kernel function; represents a weight coefficient for controlling the influence of the pitch angle on the kernel function.

2. The new energy vehicle parking space safety warning system according to claim 1, characterized in that: The parking space surrounding environment data includes a marking position of the parking space , , an absolute position of the parking space , .

3. The new energy vehicle parking space safety warning system according to claim 2, characterized in that: The positioning and navigation unit (2) obtains the relative position and dynamic information of the vehicle and the parking space based on the surrounding environment data, comprising the following steps: S1.1, preprocessing the collected parking space surrounding environment data, and collecting the original coordinates of the environment information points through the sensor ; S1.2, based on the vehicle's own positioning sensors and environmental data, based on raw coordinates determining the current absolute position of the vehicle and the angle of pose , wherein denotes the velocity component of the vehicle in the direction, denotes the velocity component of the vehicle in the direction; S1.3, calculating the position of the marking of the parking space and the current absolute position of the vehicle the relative position : ; S1.

4. Calculate the speed of the vehicle by sensor data of the vehicle and acceleration ; S1.5, absolute position of the parking space based , calculating the relative position of the vehicle to the parking space : .

4. The new energy vehicle parking space safety warning system according to claim 1, characterized in that: The data fusion processing module (31) integrates the relative position of the vehicle and the parking space output by the positioning and navigation unit (2) , the relative position of the parking space , the vehicle dynamic information and the environmental data to form a surrounding environment model, including the following steps: S2.1, performing a basic processing of the input vehicle-to-parking space relative position , the vehicle-to-parking space relative position , the vehicle dynamic information and the environmental data S2.2, converting sensor data to a unified coordinate system : wherein denotes a rotation matrix, denotes a translation vector; S2.3, the detection results of the same target are fused by weighted average: S2.4, a surrounding environment model is established based on the fused data.

5. The new energy vehicle parking space safety warning system according to claim 4, characterized in that: In the S2.4, the surrounding environment model is established based on the fused data, comprising the following steps: S2.41, building a vehicle position model ; S2.42, building a vehicle direction model : wherein, denotes the angular velocity of the vehicle; S2.43, constructing a relative position model of the vehicle and the parking space: wherein, denotes the absolute position of the parking space; S2.44, integrate the above models to form a surrounding environment model.

6. The new energy vehicle parking space safety warning system according to claim 5, characterized in that: In the S2.44, the surrounding environment model is: ； wherein, represents a surrounding environment anomaly evaluation model; represents a relative speed difference between the vehicle and the parking space at time represents a weight coefficient of the relative position; represents a weight coefficient of the vehicle heading angle; represents a weight coefficient of the relative position; represents a weight coefficient of the relative position; represents a geometric relationship between the heading angle of the vehicle and the parking space; Optimize the surrounding environment model for the motion trajectory of the vehicle and the relative angle between the vehicle and the parking space: ； wherein, represents the optimized peripheral environment anomaly evaluation model; represents a relative angular velocity change; represents a weight coefficient of the relative angular velocity change; Introduce the future motion trajectory prediction model of the vehicle in the optimized surrounding environment model: ； wherein, represents a surrounding environment anomaly evaluation model after a future motion trajectory prediction model of the vehicle is introduced; represents a vehicle motion trajectory prediction model.

7. The new energy vehicle parking space safety warning system according to claim 1, characterized in that: The decision and warning unit (4) performs risk assessment and triggers the early warning mechanism according to the analysis result, comprising the following steps: S4.1, receive the abnormal evaluation result from the data processing and analysis unit (3); S4.2, set the medium risk and high risk levels to judge whether there is a safety hidden danger in the parking space surrounding environment low risk; S4.3, trigger the warning reminder when there is medium risk in the parking space surrounding environment, and trigger the emergency warning when there is high risk.

Images