A vehicle situation awareness method and system based on experience synchronization

Abstract

The present invention provides a vehicle situation awareness method and system based on experience synchronization, which obtain vehicle information on the road, wherein the vehicle information includes vehicle position obtained through GPS and vehicle speed, acceleration, steering angle and vehicle type obtained through in-vehicle sensors; the vehicle information is uploaded to an information processing system through a communication channel; data is collected again, and vehicle information on the road is obtained again through radar speed measurement and range cameras, and the vehicle information is pre-processed to eliminate erroneous data; data of normally traveling vehicles and driving data of accident vehicles that have been involved in accidents are collected, and the vehicle information is compared with the normally traveling vehicle data and the accident vehicle driving data to perform situation prediction; based on the results of the situation prediction, the information processing system notifies the vehicle by voice through the communication channel to reduce the occurrence of accidents.

Description

Technical Field

[0001] The present invention relates to the field of vehicle safety technology, and in particular to a vehicle situation awareness method and system based on experience synchronization. Background Art

[0002] With the continuous increase in car ownership in recent years, the number of traffic accidents has also continued to rise, and the proportion of traffic accidents caused by improper driving is increasing. If a vehicle can comprehensively perceive surrounding vehicles and the driving environment while driving and infer whether driving behavior is abnormal, many traffic accidents caused by improper driving can be avoided.

[0003] The existing warning systems or methods based on the Internet of Vehicles do not take into account the impact of past experience information on situational awareness, so it also has a significant impact on the judgment of whether there is a safety risk. Summary of the Invention

[0004] The present invention proposes a vehicle situation perception method and system based on experience synchronization to solve the problem of situation prediction based on experience information.

[0005] To solve the above technical problems, the present invention provides a vehicle situation awareness method based on experience synchronization, which is characterized by comprising the following steps:

[0006] Step S1: Acquire vehicle information on the road, wherein the vehicle information includes vehicle position acquired through GPS and vehicle speed, acceleration, steering angle, and vehicle type acquired through in-vehicle sensors;

[0007] Step S2: uploading the vehicle information to the information processing system via a communication channel;

[0008] Step S3: Secondary data collection: obtaining vehicle information on the road again through radar speed measurement and distance cameras, and pre-processing the vehicle information to eliminate erroneous data;

[0009] Step S4: collecting normal driving vehicle data and accident vehicle driving data, and comparing the vehicle information with the normal driving vehicle data and accident vehicle driving data to perform situation prediction;

[0010] Step S5: Based on the result of the situation prediction, the information processing system notifies the vehicle via a communication channel to reduce the occurrence of accidents.

[0011] Preferably, the communication channel includes long-range wireless communication and short-range wireless communication.

[0012] Preferably, the short-range wireless communication interacts with the information processing system via a wireless hotspot pole or a Bluetooth hotspot pole installed on the road.

[0013] Preferably, the method for preprocessing in step S3 includes:

[0014] 1) Compare the vehicle information obtained by the radar speed measurement and distance camera with the vehicle information uploaded by the vehicle. If the difference is greater than the set threshold, the corresponding data will be excluded;

[0015] 2) Obtaining the maximum speed of vehicles of the vehicle type according to the obtained vehicle type, thereby excluding data with a speed greater than the maximum speed;

[0016] 3) Exclude data with a steering angle greater than or less than a steering angle range through the steering angle information.

[0017] Preferably, the method for performing situation prediction in step S4 includes the following steps:

[0018] Step S41: Modeling the normal driving vehicle data and vehicle information using a time series model;

[0019] Step S42: Calculating the model fitting degree. When the model fitting degree is less than the set normal driving judgment threshold, it is determined that the vehicle has an accident risk.

[0020] Step S43: When there is a risk of an accident occurring, a time series model is used to model the driving data of the accident vehicle, and a degree of model fit between the driving data model of the accident vehicle and the vehicle information model is calculated. The degree of model fit represents the probability of an accident occurring.

[0021] Step S44: Obtain the accident probability of all surrounding vehicles, set weights based on vehicle distance to calculate the accident probability of the current vehicle, and perform situation prediction.

[0022] Preferably, the time series model adopts the autoregressive integrated moving average (ARIMA) model.

[0023] Preferably, step S5 further includes: obtaining traffic flow information on the road through satellite remote sensing technology, and when the traffic flow information is too large, notifying vehicles of the necessary congestion through a communication channel to further reduce the possibility of accidents.

[0024] The present invention also provides a vehicle situation awareness system based on experience synchronization, characterized in that: the system includes a data acquisition module, a communication module, a data processing module and a situation prediction module;

[0025] The data acquisition module is used to obtain vehicle information on the road, the vehicle information including vehicle position obtained through GPS and vehicle speed, acceleration, steering angle and vehicle type obtained through in-vehicle sensors;

[0026] The communication module is used for communication between the vehicle and the data processing module;

[0027] The data processing module is used to obtain vehicle information on the road again through radar speed measurement and distance camera, and pre-process the vehicle information to eliminate erroneous data;

[0028] The situation prediction module is used to collect normal driving vehicle data and accident vehicle driving data, and compare the vehicle information with the normal driving vehicle data and accident vehicle driving data to perform situation prediction.

[0029] Preferably, the method for eliminating erroneous data in the data processing module includes:

[0030] 1) Compare the vehicle information obtained by the radar speed measurement and distance camera with the vehicle information uploaded by the vehicle. If the difference is greater than the set threshold, the corresponding data will be excluded;

[0031] 2) Obtaining the maximum speed of vehicles of the vehicle type according to the obtained vehicle type, thereby excluding data with a speed greater than the maximum speed;

[0032] 3) Exclude data with a steering angle greater than or less than a steering angle range through the steering angle information.

[0033] Preferably, the situation prediction module performs situation prediction in the following manner: modeling the normal driving vehicle data and vehicle information using a time series model; calculating the degree of model fitting, and when the degree of model fitting is less than a set normal driving judgment threshold, judging that the vehicle is at risk of an accident; when the vehicle is at risk of an accident, modeling the accident vehicle driving data using a time series model, and calculating the degree of model fitting of the accident vehicle driving data model and the vehicle information model, wherein the degree of model fitting is the probability of an accident; obtaining the probability of an accident for all surrounding vehicles, and setting weights based on vehicle distance to calculate the probability of an accident for the current vehicle, so as to perform situation prediction.

[0034] The beneficial effects of the present invention are: through secondary data collection and preprocessing, erroneous data is eliminated, making the predicted data more accurate, and using normal driving vehicle data and accident vehicle driving data as empirical data for situation prediction, further improving the accuracy of situation prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1Flowchart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following is a clear and complete description of the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0037] like Figure 1 As shown, an embodiment of the present invention provides a vehicle situation awareness method based on experience synchronization, comprising the following steps:

[0038] Step S1: Acquire vehicle information on the road, including vehicle location acquired through GPS and vehicle speed, acceleration, steering angle, and vehicle type acquired through in-vehicle sensors;

[0039] Communication channels include long-range wireless communication and short-range wireless communication. Among them, short-range wireless communication interacts with the information processing system through wireless hotspot poles or Bluetooth hotspot poles installed on the road. The embodiment of the present invention complements long-range wireless communication and short-range wireless communication. Wireless hotspot poles or Bluetooth hotspot poles are installed on key roads. When the long-range wireless communication signal is poor, information can be transmitted through short-range wireless communication, ensuring the stability of information transmission.

[0040] Step S2: Uploading vehicle information to the information processing system through a communication channel;

[0041] Step S3: Secondary data collection: Vehicle information on the road is acquired again through radar speed measurement and distance cameras, and the vehicle information is pre-processed to eliminate erroneous data;

[0042] In the embodiment of the present invention, the purpose of preprocessing is to eliminate data with acquisition errors, including:

[0043] 1) Compare the vehicle information obtained by the radar speed measurement and distance camera with the vehicle information uploaded by the vehicle. If the difference is greater than the set threshold, the corresponding data will be excluded;

[0044] 2) Obtain the maximum speed of the vehicle type by the acquired vehicle type to exclude data with a speed greater than the maximum speed;

[0045] 3) Based on the steering angle information, exclude data with steering angles greater than or less than the steering angle range.

[0046] Step S4: collecting normal driving vehicle data and accident vehicle driving data, and comparing the vehicle information with the normal driving vehicle data and the accident vehicle driving data to perform situation prediction;

[0047] Step S41: Modeling normal driving vehicle data and vehicle information using a time series model;

[0048] Step S42: Calculating the model fitting degree. When the model fitting degree is less than the set normal driving judgment threshold, it is determined that the vehicle has an accident risk.

[0049] Step S43: When the vehicle is at risk of an accident, the time series model is used to model the accident vehicle driving data, and the degree of model fit between the accident vehicle driving data model and the vehicle information model is calculated. The degree of model fit is the probability of the accident;

[0050] Step S44: Obtain the accident probability of all surrounding vehicles, set weights based on vehicle distance to calculate the accident probability of the current vehicle, and perform situation prediction.

[0051] In the embodiment of the present invention, by sensing all surrounding vehicles, the probability of the vehicle having an accident is corrected according to the probability of their accidents, so that the result is more accurate.

[0052] Based on the accuracy of positioning, the present invention selects all vehicles within a radius of 15 meters for judgment and calculates the probability of accident occurrence P by the following formula s :

[0053]

[0054]

[0055] Where N is the number of vehicles within the range, P i represents the probability of an accident occurring to the i-th vehicle, ω i represents the weight of the influence of the i-th vehicle on the vehicle, d i represents the distance between the i-th vehicle and the vehicle, and D represents the sum of the distances between the vehicle and all vehicles within the range.

[0056] The time series model in the embodiment of the present invention adopts the differential integrated moving average autoregressive model ARIMA. The differential integrated moving average autoregressive model ARIMA is a prior art and will not be described in detail again.

[0057] Step S5: Based on the situation prediction results, the information processing system notifies the vehicle via a communication channel to reduce the occurrence of accidents.

[0058] In an embodiment of the present invention, traffic flow information on the road is also obtained through satellite remote sensing technology. When the traffic flow information is too large, voice notification of unnecessary congestion is sent through communication channels to further reduce the possibility of accidents.

[0059] The present invention also provides a vehicle situation awareness system based on experience synchronization, the system includes a data acquisition module, a communication module, a data processing module and a situation prediction module;

[0060] The data acquisition module is used to obtain vehicle information on the road. The vehicle information includes vehicle location obtained through GPS and vehicle speed, acceleration, steering angle and vehicle type obtained through in-vehicle sensors;

[0061] Communication module, used for communication between the vehicle and the data processing module;

[0062] The data processing module is used to obtain vehicle information on the road again through radar speed measurement and distance camera, and pre-process the vehicle information to eliminate erroneous data;

[0063] The situation prediction module is used to collect the normal driving vehicle data and the accident vehicle driving data, and compare the vehicle information with the normal driving vehicle data and the accident vehicle driving data to perform situation prediction.

[0064] Preferably, the data processing module includes the following steps to eliminate erroneous data:

[0065] 1) Compare the vehicle information obtained by the radar speed measurement and distance camera with the vehicle information uploaded by the vehicle. If the difference is greater than the set threshold, the corresponding data will be excluded;

[0066] 2) Obtain the maximum speed of the vehicle type by the acquired vehicle type, so as to exclude data with a speed greater than the maximum speed;

[0067] 3) Based on the steering angle information, exclude data with steering angles greater than or less than the steering angle range.

[0068] Preferably, the situation prediction module performs situation prediction in the following manner: using a time series model to model normal driving vehicle data and vehicle information; calculating the degree of model fitting, and when the degree of model fitting is less than a set normal driving judgment threshold, judging that the vehicle is at risk of an accident; when the vehicle is at risk of an accident, using a time series model to model the accident vehicle driving data, and calculating the degree of model fitting of the accident vehicle driving data model and the vehicle information model, the degree of model fitting is the probability of an accident; obtaining the probability of an accident for all surrounding vehicles, setting weights based on vehicle distance to calculate the probability of an accident for the current vehicle, and performing situation prediction.

[0069] The technical features of the above embodiments may be combined in any manner. To simplify the description, not all possible combinations of the technical features in the above embodiments are described. Only preferred embodiments of the present invention are presented. While the description is relatively specific and detailed, it should not be construed as limiting the scope of the present invention. As long as there are no contradictions in the combination of these technical features, they should be considered to be within the scope of this specification.

[0070] It should be noted that those skilled in the art may make various modifications and improvements without departing from the scope of the present invention, and these modifications and improvements fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be based on the appended claims.

Claims

1. A vehicle situation awareness method based on experience synchronization, characterized by: The following steps are involved: Step S1: Acquire vehicle information on the road, wherein the vehicle information includes vehicle position acquired through GPS and vehicle speed, acceleration, steering angle, and vehicle type acquired through in-vehicle sensors; Step S2: uploading the vehicle information to the information processing system via a communication channel; Step S3: Secondary data collection: obtaining vehicle information on the road again through radar speed measurement and distance cameras, and pre-processing the vehicle information to eliminate erroneous data; Step S4: collecting normal driving vehicle data and accident vehicle driving data, and comparing the vehicle information with the normal driving vehicle data and accident vehicle driving data to perform situation prediction; Step S5: Based on the situation prediction result, the information processing system notifies the vehicle via a communication channel to reduce the occurrence of accidents; The method for performing situation prediction in step S4 includes the following steps: Step S41: Modeling the normal driving vehicle data and vehicle information using a time series model; Step S42: Calculating the model fitting degree. When the model fitting degree is less than the set normal driving judgment threshold, it is determined that the vehicle has an accident risk. Step S43: When there is a risk of an accident occurring, a time series model is used to model the driving data of the accident vehicle, and a degree of model fit between the driving data model of the accident vehicle and the vehicle information model is calculated. The degree of model fit represents the probability of an accident occurring. Step S44: Obtain the accident probability of all surrounding vehicles, set weights based on vehicle distance to calculate the accident probability of the current vehicle, and perform situation prediction; The time series model adopts the difference integrated moving average autoregressive model ARIMA.

2. The vehicle situation awareness method based on experience synchronization according to claim 1, characterized in that: The communication channels include long-range wireless communication and short-range wireless communication.

3. The vehicle situation awareness method based on experience synchronization according to claim 2, characterized in that: The short-range wireless communication interacts with the information processing system through wireless hotspots or Bluetooth hotspots installed on the road.

4. The vehicle situation awareness method based on experience synchronization according to claim 1, characterized in that: The method for preprocessing in step S3 includes: 1) Compare the vehicle information obtained by the radar speed measurement and distance camera with the vehicle information uploaded by the vehicle. If the difference is greater than the set threshold, the corresponding data will be excluded; 2) Obtaining the maximum speed of vehicles of the vehicle type based on the obtained vehicle type, thereby excluding data with a speed greater than the maximum speed; 3) Exclude data with a steering angle greater than or less than a steering angle range through the steering angle information.

5. The vehicle situation awareness method based on experience synchronization according to claim 1, characterized in that: Step S5 also includes: obtaining traffic flow information on the road through satellite remote sensing technology. When the traffic flow information is too large, voice notification of necessary congestion is sent to vehicles through communication channels to further reduce the possibility of accidents.

6. A vehicle situation awareness system based on experience synchronization, characterized by: The system includes a data acquisition module, a communication module, a data processing module and a situation prediction module; The data acquisition module is used to obtain vehicle information on the road, the vehicle information including vehicle position obtained through GPS and vehicle speed, acceleration, steering angle and vehicle type obtained through in-vehicle sensors; The communication module is used for communication between the vehicle and the data processing module; The data processing module is used to obtain vehicle information on the road again through radar speed measurement and distance camera, and pre-process the vehicle information to eliminate erroneous data; The situation prediction module is used to collect normal driving vehicle data and accident vehicle driving data, and compare the vehicle information with the normal driving vehicle data and accident vehicle driving data to perform situation prediction.

7. The vehicle situation awareness system based on experience synchronization according to claim 6, characterized in that: The method for eliminating erroneous data in the data processing module includes: 1) Compare the vehicle information obtained by the radar speed measurement and distance camera with the vehicle information uploaded by the vehicle. If the difference is greater than the set threshold, the corresponding data will be excluded; 2) Obtaining the maximum speed of vehicles of the vehicle type based on the obtained vehicle type, thereby excluding data with a speed greater than the maximum speed; 3) Exclude data with a steering angle greater than or less than a steering angle range through the steering angle information.

8. The vehicle situation awareness system based on experience synchronization according to claim 6, characterized in that: The situation prediction module performs situation prediction by: using a time series model to model the normal driving vehicle data and vehicle information; calculating the degree of model fit, and when the degree of model fit is less than a set normal driving judgment threshold, determining that the vehicle has a risk of an accident; When a vehicle is at risk of an accident, a time series model is used to model the accident vehicle driving data, and the model fit of the accident vehicle driving data model and the vehicle information model is calculated. The model fit degree is the probability of an accident; the accident probability of all surrounding vehicles is obtained, and a weight is set based on the vehicle distance to calculate the accident probability of the current vehicle for situation prediction.

Images