A vehicle-road cooperative assisted driving method and system based on artificial intelligence

Abstract

The present invention discloses a vehicle-road cooperative assisted driving method and system based on artificial intelligence, which belongs to the field of assisted driving technology. It includes: a data acquisition module that senses road obstacles at different positions of the vehicle through radar sensors and generates sensing behavior feature labels; a data cloud management module that pre-processes the received sensing behavior feature labels and outputs sensing behavior data sets and radar sensing behavior graphs; a data analysis module that calculates feature label values, calibrates radar trigger values ​​based on radar sensing behavior graphs, and calculates the error values ​​between the first trigger objects; an assisted driving feedback module that determines whether there is a sensing cooperative relationship between radar sensors and outputs assisted driving decisions; the present invention can utilize the sensor resources installed on the vehicle itself to achieve interaction and coordination of vehicle sensors, and can adaptively adjust the operating frequency of the sensor to change the monitoring distance of the sensor, thereby improving the traffic efficiency and safety of the road.

Description

Technical Field

[0001] The present invention relates to the field of assisted driving technology, and specifically to a vehicle-road collaborative assisted driving method and system based on artificial intelligence. Background Art

[0002] Vehicle-road collaboration refers to the real-time information exchange and collaborative work between vehicles and road infrastructure through wireless communication technology. This technology aims to improve road traffic efficiency and safety and realize the construction and operation of intelligent transportation systems, including various forms of interaction and collaboration such as vehicle-to-vehicle, vehicle-to-infrastructure, vehicle-to-pedestrian, and vehicle-to-network.

[0003] In the existing technology, the functions of interaction and collaboration are generally achieved through a vehicle-road collaborative perception network method that combines "vehicle perception + roadside perception", including installing wide-area radar perception equipment on the roadside and vehicle head at intersections and curves, collecting real-time information of people and vehicles on regional roads in an all-round and three-dimensional manner, analyzing and judging the trajectories of pedestrians and vehicles, and interconnecting through wireless networks to achieve collision avoidance warnings for blind spots or adjacent vehicles, so that people, vehicles and roads can collaborate effectively. This method not only requires the installation of multiple sensors on vehicles, but also requires the widespread installation of perception equipment on the roads that can interact and collaborate with vehicle sensors. On the one hand, it increases the consumption of infrastructure, and on the other hand, the scenario of "vehicle perception + roadside perception" is more complex, which not only requires stable network resource support, but also requires the design of effective interaction methods, which increases the difficulty of interaction and collaboration and is not conducive to subsequent equipment maintenance. Summary of the invention

[0004] The purpose of the present invention is to provide a vehicle-road collaborative assisted driving method and system based on artificial intelligence to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] An artificial intelligence-based vehicle-road cooperative assisted driving system, the system comprises a data acquisition module, a data cloud management module, a data analysis module and an assisted driving feedback module connected in sequence;

[0007] The data acquisition module senses road obstacles at different positions of the vehicle through a radar sensor, and when a road obstacle is sensed, collects a sensing behavior feature label of the radar sensor;

[0008] The data cloud management module is used to build a vehicle-road cooperative data cloud, pre-process the sensing behavior feature labels received from each radar sensor, and output a sensing behavior data set and a radar sensing behavior graph;

[0009] The data analysis module is used to build a vehicle-road collaborative analysis center, calculate feature label values, calibrate radar trigger values ​​based on radar sensing behavior graphs, and calculate the error value between the first trigger objects;

[0010] The assisted driving feedback module is used to determine whether there is a sensing cooperative relationship between the radar sensors, and output an assisted driving decision based on the sensing cooperative relationship.

[0011] Further, the data acquisition module includes a sensor unit and a data acquisition unit;

[0012] The sensor unit is used to connect to radar sensors installed at different positions of the vehicle and adjust the operating frequency of the radar sensors;

[0013] The data acquisition unit is used to collect the operating frequency of the radar sensor and the distance of the road obstacle sensed by the radar sensor.

[0014] Furthermore, the data cloud management module includes a data statistics unit and a data graph characterization unit;

[0015] The data statistics unit is used to retrieve the sensing behavior feature labels recorded each time a sensing behavior is generated to generate a sensing behavior data set;

[0016] The data graph characterization unit is used to establish a radar sensing behavior graph model to generate a radar sensing behavior graph each time a sensing behavior is generated.

[0017] Further, the data analysis module includes a feature label quantization unit and a collaborative error quantization unit;

[0018] The feature label quantification unit establishes a sensing feedback model based on the sensing behavior data set and calculates the feature label value;

[0019] The collaborative error quantization unit selects a first trigger object based on the feature label value, calibrates the radar trigger value, and calculates the error value between the first trigger objects.

[0020] Furthermore, the auxiliary driving feedback module includes a collaborative relationship determination unit and a decision feedback unit;

[0021] The cooperative relationship determination unit determines whether there is a sensing cooperative relationship between the radar sensors based on the error value;

[0022] The decision feedback unit enables the radar sensors in a cooperative relationship to maintain the same operating frequency based on the cooperative induction relationship.

[0023] A vehicle-road cooperative assisted driving method based on artificial intelligence, the method comprising the following steps:

[0024] Step S1: connecting radar sensors installed at different positions of the vehicle, adjusting the operating frequencies of the radar sensors so that the operating frequencies of the radar sensors are coordinated, and the radar sensors are used to sense and provide feedback on whether there are road obstacles at different positions of the vehicle;

[0025] Step S2: When the radar sensors installed at different positions sense a road obstacle, the sensing behavior feature label of each radar sensor is recorded and uploaded to the vehicle-road cooperative data cloud. The sensing behavior feature label records the operating frequency of the radar sensor and the distance of the road obstacle sensed by the radar sensor.

[0026] Step S3: Based on the sensing behavior of the radar sensor, the vehicle-road cooperative data cloud pre-processes the received sensing behavior feature labels of each radar sensor, quantifies the feature label values ​​of the radar sensor, and obtains the radar sensing behavior graph when the sensing behavior is generated;

[0027] Step S4: Based on the radar sensing behavior diagram, calibrate the radar trigger value; analyze the sensing coordination relationship between radar sensors through the vehicle-road collaborative analysis center, and output the assisted driving decision based on the sensing coordination relationship.

[0028] Furthermore, the generation method of the sensing behavior feature label includes:

[0029] The radar sensors are uniformly numbered and the i-th radar sensor is recorded as , where a radar sensor is installed at each position;

[0030] When the sensing behavior occurs, the operating frequency of the radar sensor and the distance of the road obstacle sensed by the radar sensor are obtained, and the radar sensor is formed. The sensing behavior feature label is denoted as .

[0031] Furthermore, the vehicle-road cooperative data cloud preprocesses the received sensing behavior feature labels of each radar sensor, including:

[0032] Retrieve the characteristic tag of the sensing behavior recorded when the sensing behavior occurs for the xth time , and generate a sensing behavior dataset, denoted as ,in, represents the i-th radar sensor The operating frequency, represents the i-th radar sensor The distance of the sensed road obstacle, I represents the total number of radar sensors;

[0033] Establishing a radar sensing behavior graph model, wherein the radar sensing behavior graph model is a regular polygon graph, and the total number of edges of the regular polygon graph is equal to the total number of radar sensors, wherein one radar sensor corresponds to one edge of the polygon;

[0034] The center point of the regular polygon is marked, and rays are drawn to each edge with the center point as the vertex. The length of the ray is the quantized feature label value. The end points of each ray are connected in sequence to form a radar sensing behavior diagram for the xth sensing behavior, which is recorded as .

[0035] Furthermore, the quantification method of the feature label value includes:

[0036] Retrieving the sensor behavior dataset , establish an induction feedback model and calculate the feature label value:

[0037] ;

[0038] In the formula, Indicates radar sensor The feature label value of represents the average value of each radar sensor when the sensing behavior occurs for the xth time, and , represents the average distance of the road obstacles sensed by each radar sensor when the sensing behavior is generated for the xth time, and , represents the standard deviation of the operating frequency, and , represents the standard deviation of the distance, and ;

[0039] According to the above method, the characteristic tag value is quantified based on the area method. The working frequency of the sensor affects its effective sensing range. Different working frequencies trigger different sensing effects. The larger the area value, the larger the tag value. However, the larger the working frequency, the better. An excessively large working frequency will result in a longer identification distance, and more invalid sensing will be triggered. In the process of collaborative sensing of sensors, the present invention selects the minimum characteristic tag value, studies and controls the sensor to be triggered within an appropriate working frequency, so that different radar sensors can maintain the same frequency of working frequencies.

[0040] Furthermore, the specific implementation process of step S4 includes:

[0041] Select the i-th radar sensor corresponding to the smallest feature label value , as the first trigger object of the x-th sensing behavior; the radar sensing behavior map of the x-th sensing behavior The area value of the first trigger object is recorded as ;

[0042] In all sensing behaviors, obtain radar sensors separately and radar sensors As the average of the radar trigger values ​​when the first trigger object is generated, calculate the error value between the two means , where Indicates the radar sensor that generates sensing behavior for the xth time Radar trigger value when it is the first trigger object;

[0043] If the error value is less than or equal to the error threshold, then the radar sensor and radar sensors Assisted driving collaboration occurs between them, otherwise, no assisted driving collaboration occurs;

[0044] Radar sensors for driver assistance and radar sensors , if the radar sensor When a road obstacle is detected, the radar sensor With radar sensor Keep the operating frequency the same.

[0045] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention provides an artificial intelligence-based vehicle-road cooperative assisted driving method and system, including: a data acquisition module that senses road obstacles at different positions of the vehicle through radar sensors to generate sensing behavior feature labels; a data cloud management module that preprocesses the received sensing behavior feature labels and outputs sensing behavior data sets and radar sensing behavior graphs; a data analysis module that calculates feature label values, calibrates radar trigger values ​​based on radar sensing behavior graphs, and calculates error values ​​between the first triggering objects; an assisted driving feedback module that determines whether there is a sensing cooperative relationship between radar sensors and outputs assisted driving decisions; the present invention can utilize the sensor resources installed on the vehicle itself to achieve interaction and coordination of vehicle sensors, and can adaptively adjust the operating frequency of the sensor to change the monitoring distance of the sensor, thereby improving road traffic efficiency and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0047] Figure 1 It is a schematic diagram of the steps of a vehicle-road collaborative assisted driving method based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] In the first embodiment of the present invention, a vehicle-road cooperative assisted driving system based on artificial intelligence is provided, the system comprising: a data acquisition module, a data cloud management module, a data analysis module and an assisted driving feedback module connected in sequence;

[0050] The data acquisition module senses road obstacles at different positions of the vehicle through a radar sensor, and when a road obstacle is sensed, collects a sensing behavior feature label of the radar sensor;

[0051] The data acquisition module includes a sensor unit and a data acquisition unit;

[0052] The sensor unit is used to connect to radar sensors installed at different positions of the vehicle and adjust the operating frequency of the radar sensors;

[0053] The data acquisition unit is used to collect the operating frequency of the radar sensor and the distance of the road obstacle sensed by the radar sensor.

[0054] The data cloud management module is used to build a vehicle-road cooperative data cloud, pre-process the sensing behavior feature labels received from each radar sensor, and output a sensing behavior data set and a radar sensing behavior graph;

[0055] The data cloud management module includes a data statistics unit and a data graph characterization unit;

[0056] The data statistics unit is used to retrieve the sensing behavior feature labels recorded each time a sensing behavior is generated to generate a sensing behavior data set;

[0057] The data graph characterization unit is used to establish a radar sensing behavior graph model to generate a radar sensing behavior graph each time a sensing behavior is generated.

[0058] The data analysis module is used to build a vehicle-road collaborative analysis center, calculate feature label values, calibrate radar trigger values ​​based on radar sensing behavior graphs, and calculate the error value between the first trigger objects;

[0059] The data analysis module includes a feature label quantization unit and a collaborative error quantization unit;

[0060] The feature label quantification unit establishes a sensing feedback model based on the sensing behavior data set and calculates the feature label value;

[0061] The collaborative error quantization unit selects a first trigger object based on the feature label value, calibrates the radar trigger value, and calculates the error value between the first trigger objects.

[0062] The assisted driving feedback module is used to determine whether there is a sensing cooperative relationship between the radar sensors, and output an assisted driving decision based on the sensing cooperative relationship;

[0063] The auxiliary driving feedback module includes a collaborative relationship determination unit and a decision feedback unit;

[0064] The cooperative relationship determination unit determines whether there is a sensing cooperative relationship between the radar sensors based on the error value;

[0065] The decision feedback unit enables the radar sensors in a cooperative relationship to maintain the same operating frequency based on the cooperative induction relationship.

[0066] See also Figure 1 In the second embodiment, a vehicle-road cooperative assisted driving method based on artificial intelligence is provided to better apply a vehicle-road cooperative assisted driving system based on artificial intelligence of the present invention, and the method includes the following steps:

[0067] Step S1: connecting radar sensors installed at different positions of the vehicle, adjusting the operating frequencies of the radar sensors so that the operating frequencies of the radar sensors are coordinated, and the radar sensors are used to sense and provide feedback on whether there are road obstacles at different positions of the vehicle;

[0068] Step S2: When the radar sensors installed at different positions sense a road obstacle, the sensing behavior feature label of each radar sensor is recorded and uploaded to the vehicle-road cooperative data cloud. The sensing behavior feature label records the operating frequency of the radar sensor and the distance of the road obstacle sensed by the radar sensor.

[0069] Exemplarily, the generation method of the sensing behavior feature tag includes:

[0070] The radar sensors are uniformly numbered and the i-th radar sensor is recorded as , where a radar sensor is installed at each position;

[0071] When the sensing behavior occurs, the operating frequency of the radar sensor and the distance of the road obstacle sensed by the radar sensor are obtained, and the radar sensor is formed. The sensing behavior feature label is denoted as .

[0072] Step S3: Based on the sensing behavior of the radar sensor, the vehicle-road cooperative data cloud pre-processes the received sensing behavior feature labels of each radar sensor, quantifies the feature label values ​​of the radar sensor, and obtains the radar sensing behavior graph when the sensing behavior is generated;

[0073] Exemplarily, the vehicle-road cooperative data cloud preprocesses the received sensing behavior feature labels of each radar sensor including:

[0074] Retrieve the characteristic tag of the sensing behavior recorded when the sensing behavior occurs for the xth time , and generate a sensing behavior dataset, denoted as ,in, represents the i-th radar sensor The operating frequency, represents the i-th radar sensor The distance of the sensed road obstacle, I represents the total number of radar sensors;

[0075] Establishing a radar sensing behavior graph model, wherein the radar sensing behavior graph model is a regular polygon graph, and the total number of edges of the regular polygon graph is equal to the total number of radar sensors, wherein one radar sensor corresponds to one edge of the polygon;

[0076] The center point of the regular polygon is marked, and rays are drawn to each edge with the center point as the vertex. The length of the ray is the quantized feature label value. The end points of each ray are connected in sequence to form a radar sensing behavior diagram for the xth sensing behavior, which is recorded as ;

[0077] Exemplarily, the quantification method of the feature label value includes:

[0078] Retrieving the sensor behavior dataset , establish an induction feedback model and calculate the feature label value:

[0079] ;

[0080] In the formula, Indicates radar sensor The feature label value of represents the average value of each radar sensor when the sensing behavior occurs for the xth time, and , represents the average distance of the road obstacles sensed by each radar sensor when the sensing behavior is generated for the xth time, and , represents the standard deviation of the operating frequency, and , represents the standard deviation of the distance, and .

[0081] Step S4: based on the radar sensing behavior graph, calibrate the radar trigger value; analyze the sensing coordination relationship between the radar sensors through the vehicle-road cooperative analysis center, and output the assisted driving decision based on the sensing coordination relationship;

[0082] For example, the i-th radar sensor corresponding to the smallest feature label value is selected , as the first trigger object of the x-th sensing behavior; the radar sensing behavior map of the x-th sensing behavior The area value of the first trigger object is recorded as ;

[0083] In all sensing behaviors, obtain radar sensors separately and radar sensors As the average of the radar trigger values ​​when the first trigger object is generated, calculate the error value between the two means , where Indicates the radar sensor that generates sensing behavior for the xth time Radar trigger value when it is the first trigger object;

[0084] If the error value is less than or equal to the error threshold, then the radar sensor and radar sensors Assisted driving collaboration occurs between them, otherwise, no assisted driving collaboration occurs;

[0085] Radar sensors for driver assistance and radar sensors , if the radar sensor When a road obstacle is detected, the radar sensor With radar sensor Keep the operating frequency the same.

[0086] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0087] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A vehicle-road cooperative assisted driving method based on artificial intelligence, characterized in that: The method comprises the following steps: Step S1: connecting radar sensors installed at different positions of the vehicle, adjusting the operating frequencies of the radar sensors so that the operating frequencies of the radar sensors are coordinated, and the radar sensors are used to sense and provide feedback on whether there are road obstacles at different positions of the vehicle; Step S2: When the radar sensors installed at different positions sense a road obstacle, the sensing behavior feature label of each radar sensor is recorded and uploaded to the vehicle-road cooperative data cloud. The sensing behavior feature label records the operating frequency of the radar sensor and the distance of the road obstacle sensed by the radar sensor. Step S3: Based on the sensing behavior of the radar sensor, the vehicle-road cooperative data cloud pre-processes the received sensing behavior feature labels of each radar sensor, quantifies the feature label values ​​of the radar sensor, and obtains the radar sensing behavior graph when the sensing behavior is generated; Step S4: based on the radar sensing behavior graph, calibrate the radar trigger value; analyze the sensing coordination relationship between the radar sensors through the vehicle-road cooperative analysis center, and output the assisted driving decision based on the sensing coordination relationship; The vehicle-road cooperative data cloud preprocesses the received sensing behavior feature labels of each radar sensor, including: Retrieve the characteristic tag of the sensing behavior recorded when the sensing behavior occurs for the xth time , and generate a sensing behavior dataset, denoted as ,in, represents the i-th radar sensor The operating frequency, represents the i-th radar sensor The distance of the sensed road obstacle, I represents the total number of radar sensors; Establishing a radar sensing behavior graph model, wherein the radar sensing behavior graph model is a regular polygon graph, and the total number of edges of the regular polygon graph is equal to the total number of radar sensors, wherein one radar sensor corresponds to one edge of the polygon; The center point of the regular polygon is marked, and rays are drawn to each edge with the center point as the vertex. The length of the ray is the quantized feature label value. The end points of each ray are connected in sequence to form a radar sensing behavior diagram for the xth sensing behavior, which is recorded as ; The quantification method of the feature label value includes: Retrieving the sensor behavior dataset , establish an induction feedback model and calculate the feature label value: ； In the formula, Indicates radar sensor The feature label value of represents the average value of each radar sensor when the sensing behavior occurs for the xth time, and , represents the average distance of the road obstacles sensed by each radar sensor when the sensing behavior is generated for the xth time, and , represents the standard deviation of the operating frequency, and , represents the standard deviation of the distance, and ; The specific implementation process of step S4 includes: Select the i-th radar sensor corresponding to the smallest feature label value , as the first trigger object of the x-th sensing behavior; the radar sensing behavior map of the x-th sensing behavior The area value of the first trigger object is recorded as ; In all sensing behaviors, obtain radar sensors separately and radar sensors As the average of the radar trigger values ​​when the first trigger object is generated, calculate the error value between the two means , where Indicates the radar sensor that generates sensing behavior for the xth time Radar trigger value when it is the first trigger object; If the error value is less than or equal to the error threshold, then the radar sensor and radar sensors Assisted driving collaboration occurs between them, otherwise, no assisted driving collaboration occurs; Radar sensors for driver assistance and radar sensors , if the radar sensor When a road obstacle is detected, the radar sensor With radar sensor Keep the operating frequency the same.

2. The vehicle-road cooperative assisted driving method based on artificial intelligence according to claim 1, characterized in that: The generation method of the sensing behavior feature label includes: The radar sensors are uniformly numbered and the i-th radar sensor is recorded as , where a radar sensor is installed at each position; When the sensing behavior occurs, the operating frequency of the radar sensor and the distance of the road obstacle sensed by the radar sensor are obtained, and the radar sensor is formed. The sensing behavior feature label is denoted as .

3. A vehicle-road cooperative assisted driving system based on artificial intelligence, executing a vehicle-road cooperative assisted driving method based on artificial intelligence as claimed in claim 1, characterized in that: The system comprises a data acquisition module, a data cloud management module, a data analysis module and an assisted driving feedback module which are connected in sequence; The data acquisition module senses road obstacles at different positions of the vehicle through a radar sensor, and when a road obstacle is sensed, collects a sensing behavior feature label of the radar sensor; The data cloud management module is used to build a vehicle-road cooperative data cloud, pre-process the sensing behavior feature labels received from each radar sensor, and output a sensing behavior data set and a radar sensing behavior graph; The data analysis module is used to build a vehicle-road collaborative analysis center, calculate feature label values, calibrate radar trigger values ​​based on radar sensing behavior graphs, and calculate the error value between the first trigger objects; The assisted driving feedback module is used to determine whether there is a sensing cooperative relationship between the radar sensors, and output an assisted driving decision based on the sensing cooperative relationship.

4. The artificial intelligence-based vehicle-road cooperative assisted driving system according to claim 3, characterized in that: The data acquisition module includes a sensor unit and a data acquisition unit; The sensor unit is used to connect to radar sensors installed at different positions of the vehicle and adjust the operating frequency of the radar sensors; The data acquisition unit is used to collect the operating frequency of the radar sensor and the distance of the road obstacle sensed by the radar sensor.

5. The artificial intelligence-based vehicle-road cooperative assisted driving system according to claim 3, characterized in that: The data cloud management module includes a data statistics unit and a data graph characterization unit; The data statistics unit is used to retrieve the sensing behavior feature labels recorded each time a sensing behavior is generated to generate a sensing behavior data set; The data graph characterization unit is used to establish a radar sensing behavior graph model to generate a radar sensing behavior graph each time a sensing behavior is generated.

6. The artificial intelligence-based vehicle-road cooperative assisted driving system according to claim 3, characterized in that: The data analysis module includes a feature label quantization unit and a collaborative error quantization unit; The feature label quantification unit establishes a sensing feedback model based on the sensing behavior data set and calculates the feature label value; The collaborative error quantization unit selects a first trigger object based on the feature label value, calibrates the radar trigger value, and calculates the error value between the first trigger objects.

7. The artificial intelligence-based vehicle-road cooperative assisted driving system according to claim 3, characterized in that: The auxiliary driving feedback module includes a collaborative relationship determination unit and a decision feedback unit; The cooperative relationship determination unit determines whether there is a sensing cooperative relationship between the radar sensors based on the error value; The decision feedback unit enables the radar sensors in a cooperative relationship to maintain the same operating frequency based on the cooperative induction relationship.

Images