A driving behavior recognition method and apparatus

Abstract

The application provides a driving behavior identification method and device, which extracts driving information of surrounding vehicles in a target time period from surrounding information of a target vehicle in the target time period, obtains actual driving conditions of the surrounding vehicles in the target time period according to the driving information of the surrounding vehicles in the target time period, and predicts expected driving conditions of the surrounding vehicles in the target time period by using a prediction module. The actual driving conditions and the expected driving conditions are compared, if they are inconsistent, the surrounding vehicles may have illegal behaviors, and the surrounding vehicles are marked as pre-illegal vehicles. Then, if the target vehicle is disturbed in the target time period, it is judged whether there is a preset behavior in the actual driving conditions, if yes, the pre-illegal vehicles are determined as actual illegal vehicles. In this way, the actual driving conditions in the preset time period are inconsistent with the expected driving conditions, the target vehicle is disturbed in the target time period, and the preset behavior exists in the actual driving conditions, so it is determined that the surrounding vehicles have illegal behaviors.

Description

Technical Field

[0001] This application relates to the field of automotive technology, and more particularly to a driving behavior recognition method and device. Background Technology

[0002] Intelligent vehicles are integrated systems that combine environmental perception, planning and decision-making, and multi-level assisted driving functions. They utilize technologies such as computers, modern sensing, information fusion, communication, artificial intelligence, and automatic control, and have become a research hotspot in the field of vehicle engineering worldwide and a new driving force for the growth of the automotive industry.

[0003] When intelligent vehicles are driving, they may encounter traffic violations by other vehicles. Such violations may lead to traffic accidents or adverse consequences. Timely identification of traffic violations by surrounding vehicles can effectively reduce the negative impact of such violations.

[0004] Therefore, how to identify traffic violations by surrounding vehicles is an urgent problem to be solved. Summary of the Invention

[0005] This application provides a driving behavior recognition method and device for identifying vehicles that have committed traffic violations.

[0006] Firstly, this application provides a driving behavior recognition method, including:

[0007] Extract the driving information of surrounding vehicles during the target time period from the surrounding information of the target vehicle during the target time period, and obtain the actual driving situation of the surrounding vehicles during the target time period based on the driving information of the surrounding vehicles during the target time period.

[0008] The prediction module is used to predict the expected driving conditions of the surrounding vehicles during the target time period.

[0009] The actual driving situation is compared with the expected driving situation. If they are inconsistent, the surrounding vehicles are marked as vehicles that are about to violate traffic rules.

[0010] If the target vehicle is interfered with during the target time period, it is determined whether there is a pre-set behavior in the actual driving situation. If so, the vehicle that was supposed to violate the rules is determined to be the vehicle that actually violated the rules.

[0011] Optionally, the method further includes:

[0012] Upon receiving a report instruction, the system identifies the identification information of surrounding vehicles and sends the identification information to the alarm server.

[0013] Optionally, the target time period includes the time period between the current moment and historical moments;

[0014] The step of extracting the driving information of surrounding vehicles during the target time period from the surrounding information of the target vehicle specifically includes:

[0015] Obtain the current surrounding information of the target vehicle at the current moment and the historical surrounding information at historical moments;

[0016] Extract the current driving information of surrounding vehicles at the current time from the current surrounding information, and extract the historical driving information of surrounding vehicles at the historical time from the historical surrounding information.

[0017] Optionally, obtaining the current surrounding information of the target vehicle at the current moment and the historical surrounding information at historical moments specifically includes:

[0018] Obtain current surrounding information of the target vehicle within a preset range at the current time, as well as historical surrounding information within the preset range at historical times.

[0019] Optionally, the surrounding information includes: lane-level environmental information, traffic light information, and traffic participant information;

[0020] The lane-level environmental information includes:

[0021] At least one of lane conditions, lane width, lane marking type, and lane speed limit, wherein the lane conditions include going straight and / or turning;

[0022] The traffic light information includes the traffic light information corresponding to each lane;

[0023] The traffic participant information includes the driving information of the surrounding vehicles.

[0024] Optionally, obtain the current surrounding information of the target vehicle at the current moment and the historical surrounding information at historical moments, specifically including:

[0025] The autonomous driving perception system or the driver assistance perception system acquires the current lane-level environmental information of the target vehicle at the current moment and the historical lane-level environmental information at historical moments; or,

[0026] Obtain the current location and historical location of the target vehicle, and query the current lane-level environmental information of the current location and the historical lane-level environmental information of the historical location on the high-precision map.

[0027] Optionally, obtaining the current surrounding information of the target vehicle within a first preset range in a first direction and a second preset range in a second direction, as well as historical surrounding information within the preset range at a historical time, specifically includes:

[0028] Obtain current surrounding information of the target vehicle within a first preset range in the first direction and a second preset range in the second direction at the current time, as well as historical surrounding information within the first preset range in the first direction and the second preset range in the second direction at historical times;

[0029] The first direction is the direction of travel of the target vehicle, and the second direction is the direction that intersects with the first direction;

[0030] When the target vehicle's travel distance within a preset time period is greater than the second preset range, the travel distance is taken as the first preset range; when the target vehicle's travel distance within a preset time period is less than the second preset range, the second preset range is taken as the first preset range. Optionally, the actual travel status of the surrounding vehicles during the target time period is obtained based on the driving information of the surrounding vehicles during the target time period, specifically including:

[0031] Based on the current driving information, obtain the current lane where the surrounding vehicles are currently located;

[0032] Based on the historical driving information, the historical lanes where the surrounding vehicles were located at historical moments are obtained.

[0033] If the current lane and the historical lane are inconsistent, then the actual driving situation of the surrounding vehicles during the target time period includes lane changes.

[0034] Optionally, the prediction module is used to predict the expected driving conditions of the surrounding vehicles during the target time period, specifically including:

[0035] The prediction module is used to predict driving information at multiple times within the target time period;

[0036] The expected driving conditions of surrounding vehicles within the target time period are obtained by combining the driving information from the multiple time points.

[0037] Optionally, the actual driving situation and the expected driving situation are compared. If they do not match, the surrounding vehicles are marked as vehicles about to commit a traffic violation. Specifically, this includes:

[0038] Compare the actual driving conditions with the expected driving conditions;

[0039] If the actual driving situation includes lane changing and the expected driving situation includes driving in the same lane, then the surrounding vehicles are marked as vehicles that are about to violate the rules.

[0040] Optionally, if the target vehicle is disturbed during the target time period, determining whether there is a preset behavior in the actual driving situation specifically includes:

[0041] Determine whether the target vehicle is interfered with during the target time period. If so, check whether there is at least one of the following in the actual driving situation based on the surrounding information during the target time period: changing lanes over a solid line, speeding, running a red light, or not driving in the designated lane.

[0042] Optionally, determining whether the target vehicle is interfered with during the target time period specifically includes:

[0043] Determine whether the target vehicle experiences a speed change greater than a threshold within the target time period.

[0044] Optionally, the method further includes:

[0045] A prompt voice is issued to indicate that the surrounding vehicles are actually violating traffic rules.

[0046] Secondly, this application provides a driving behavior recognition device, comprising:

[0047] The acquisition module is used to extract the driving information of surrounding vehicles in the target time period from the surrounding information of the target vehicle in the target time period, and to obtain the actual driving situation of the surrounding vehicles in the target time period based on the driving information of the surrounding vehicles in the target time period.

[0048] The prediction module is used to predict the expected driving conditions of the surrounding vehicles during the target time period.

[0049] The comparison module is used to compare the actual driving situation with the expected driving situation. If they are inconsistent, the surrounding vehicles are marked as vehicles that are about to violate traffic rules.

[0050] The judgment module is used to determine whether there is a preset behavior in the actual driving situation if the target vehicle is interfered with during the target time period. If so, the vehicle that is about to violate the rules is determined to be the vehicle that actually violates the rules.

[0051] Thirdly, this application provides an electronic device, including: a memory and a processor;

[0052] The memory is used to store instructions; the processor is used to invoke the instructions in the memory to execute the driving behavior recognition method in the first aspect and any possible design of the first aspect.

[0053] Fourthly, this application provides a computer-readable storage medium storing computer instructions, which, when executed by at least one processor of an electronic device, enable the electronic device to perform the driving behavior recognition method of the first aspect and any possible design of the first aspect.

[0054] Fifthly, this application provides a computer program product comprising computer instructions, wherein when at least one processor of an electronic device executes the computer instructions, the electronic device performs the driving behavior recognition method of the first aspect and any possible design of the first aspect.

[0055] The driving behavior recognition method provided in this application extracts the driving information of surrounding vehicles within a target time period from the surrounding information of the target vehicle within that time period. Based on this information, it obtains the actual driving conditions of the surrounding vehicles within the target time period and uses a prediction module to predict their expected driving conditions within that time period. Then, it compares the actual driving conditions with the expected driving conditions. If the actual driving conditions are inconsistent with the expected driving conditions, the surrounding vehicles may have committed traffic violations, and these vehicles are marked as potentially violating traffic rules. Furthermore, if the target vehicle is interfered with within the target time period, it determines whether a pre-set behavior exists in the actual driving conditions. If such a behavior exists, the potentially violating vehicle is identified as an actual violating vehicle. Thus, when the actual driving conditions within a preset time period are inconsistent with the predicted driving conditions, it indicates that surrounding vehicles may have committed traffic violations. Simultaneously, if the target vehicle is interfered with within the target time period and a pre-set behavior exists in its actual driving conditions, it further confirms that the surrounding vehicles have committed traffic violations, thereby accurately identifying vehicles with violations. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A schematic diagram of a scenario for a driving behavior recognition method provided in an embodiment of this application;

[0058] Figure 2 A flowchart of a driving behavior recognition method provided in an embodiment of this application;

[0059] Figure 3 A flowchart of a driving behavior recognition method provided in an embodiment of this application;

[0060] Figure 4 This is a schematic diagram of the structure of a driving behavior recognition device provided in an embodiment of this application;

[0061] Figure 5 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0063] When intelligent vehicles are in operation, they may encounter traffic violations by surrounding vehicles. Such violations can interfere with the normal driving of both the vehicle itself and other vehicles, potentially leading to traffic accidents or other adverse consequences. Timely identification and reporting of these violations can effectively reduce their negative impact. Therefore, how to identify traffic violations by surrounding vehicles is a crucial issue.

[0064] To address the aforementioned issues, this application proposes a driving behavior recognition method. This method extracts the driving information of surrounding vehicles within a target time period from the surrounding information of the target vehicle during that time period. Based on this information, it obtains the actual driving conditions of the surrounding vehicles within the target time period. A prediction module is used to predict the expected driving conditions of surrounding vehicles within the target time period. The actual driving conditions are compared with the predicted expected driving conditions. If the actual driving conditions and predicted driving conditions are inconsistent, the surrounding vehicles are marked as vehicles suspected of traffic violations. Furthermore, if the target vehicle is disturbed during the target time period, it is determined whether a pre-set behavior exists in the actual driving conditions during that time period. If such a behavior exists, the vehicle suspected of traffic violations is identified as a vehicle that has actually committed a traffic violation. In this way, by obtaining the actual driving conditions and predicting the expected driving conditions during the target time period, and when the actual and expected driving conditions are inconsistent, it indicates that surrounding vehicles may be committing traffic violations, and these surrounding vehicles are marked as vehicles suspected of traffic violations. If the target vehicle is disturbed during the target time period and a pre-set behavior exists in its actual driving conditions, then the vehicle suspected of traffic violations has committed a traffic violation, and is identified as a vehicle that has actually committed a traffic violation, thereby accurately identifying vehicles around the target vehicle that are committing traffic violations.

[0065] The technical solutions of this application will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0066] Figure 1This is a schematic diagram illustrating a driving behavior recognition method according to an embodiment of this application. The vehicle system of the target vehicle 10 acquires surrounding information of the target vehicle 10 within a target time period, extracts driving information of surrounding vehicles within the target time period from this information, and obtains the actual driving situation of the surrounding vehicles within the target time period based on this information. The vehicle system of the target vehicle 10 uses a prediction module to predict the expected driving situation of surrounding vehicles within the target time period, and compares the actual driving situation with the expected driving situation. If the actual driving situation is inconsistent with the expected driving situation, the vehicle system marks the surrounding vehicles as vehicles suspected of traffic violations. Then, the vehicle system of the target vehicle 10 determines whether the target vehicle 10 is interfered with within the target time period. If the target vehicle 10 is interfered with within the target time period, the vehicle system determines whether there is a preset behavior in the actual driving situation of the surrounding vehicles. If there is a preset behavior in the actual driving situation, the vehicle system marks the vehicles suspected of traffic violations as vehicles that have actually violated traffic rules, and issues a prompt voice to alert the user that there is a traffic violation by the surrounding vehicles. After receiving a report instruction sent by the user through client 30, the vehicle's infotainment system identifies the identification information of surrounding vehicles and sends the identification information to alarm server 20 and / or client 30. Upon receiving the identification information, alarm server 20 and / or client 30 can report the identification information to report the actual vehicle committing the traffic violation.

[0067] Figure 2 A flowchart illustrating a driving behavior recognition method according to an embodiment of this application is shown. Figure 1 Based on the illustrated embodiments, as Figure 2 As shown, with the vehicle's infotainment system as the executing entity, the method in this embodiment may include the following steps:

[0068] S101. Extract the driving information of surrounding vehicles during the target time period from the surrounding information of the target vehicle during the target time period, and obtain the actual driving situation of surrounding vehicles during the target time period based on the driving information of surrounding vehicles during the target time period.

[0069] In this embodiment of the application, the target vehicle can be an intelligent vehicle with autonomous driving function, the vehicle system can be an autonomous driving system or an assisted driving system, the autonomous driving system is, for example, a Level 4 autonomous driving system or a Level 5 autonomous driving system, the autonomous driving system can include an autonomous driving perception system, the assisted driving system is, for example, a Level 1 autonomous driving system, a Level 2 autonomous driving system or a Level 3 autonomous driving system, the assisted driving system can include an assisted driving perception system.

[0070] The vehicle-mounted system acquires the surrounding information of the target vehicle within a target time period, which can include the time interval between the current moment and historical moments. For example, if the start of the target time period is a historical moment or a point in time before the historical moment, and the end of the target time period is the current moment or a point in time after the current moment, then the target time period is the time interval between the historical moment and the current moment, or the target time period includes the time interval between the historical moment and the current moment. Therefore, the surrounding information of the target time period includes the current surrounding information of the current moment and the historical surrounding information of the historical moment. The current surrounding information of the current moment can be understood as the surrounding information of the target vehicle's current location, and the historical surrounding information of the historical moment can be understood as the surrounding information of the target vehicle's location at a past moment.

[0071] The target vehicle is equipped with multiple sensors, such as photoelectric sensors, image sensors, and speed sensors. These sensors continuously sense the surrounding environment and collect data as the vehicle moves. The vehicle's infotainment system acquires the detection data from these sensors to obtain information about the target vehicle's surroundings during a target time period. Specifically, it can obtain information about the target vehicle's surroundings at various moments within the target time period.

[0072] Surrounding information includes lane-level environmental information, traffic light information, and information on surrounding road users. Lane-level environmental information includes at least one of the following: lane conditions, lane width, lane marking type, and road speed limit. Lane conditions include straight and / or turning lanes, such as straight, left turn, right turn, straight and left turn, straight and right turn, left turn and right turn, U-turn, left turn and U-turn, etc. Lane marking types include solid white lines, single solid yellow lines, double solid yellow lines, dashed white lines, and single dashed yellow lines, etc. Road speed limits include 60km / h-120km / h, 60km / h-100km / h, 40km / h-80km / h, 30km / h-60km / h, 20km / h-60km / h, 40km / h-60km / h, 30km / h-50km / h, and 20km / h-40km / h, etc. Traffic light information includes traffic light information for each lane. Surrounding traffic participant information includes the driving information of surrounding vehicles, such as the lane they are in, their speed, position, and angle between them and the lane lines. Lanes include, for example, the inner lane, outer lane, and middle lane; speeds include, for example, 60 km / h and 90 km / h; positions include, for example, before or after a traffic light intersection; and angles between them and the lane lines include, for example, 10° and 30°. Therefore, the surrounding information for the target time period includes the target vehicle's current lane-level environmental information, current traffic light information, and current driving information of surrounding vehicles at the current moment. It also includes the target vehicle's historical lane-level environmental information, historical traffic light information, and historical driving information of surrounding vehicles at historical moments. For example, a historical moment could be the moment 8 seconds before the current moment; therefore, the time period between the current moment and a historical moment could be 8 seconds, and the target time period could be 8 seconds.

[0073] In some embodiments, the current lane-level environmental information of the target vehicle at the current moment and the historical lane-level environmental information at a previous moment are obtained through an autonomous driving perception system or an assisted driving perception system. Alternatively, the current location and historical location of the target vehicle are obtained, and the lane-level environmental information of the current location and the historical lane-level environmental information of the historical location at a previous moment are queried on a high-precision map. Traffic light information of the target vehicle at the current moment and the historical traffic light information are obtained through the autonomous driving perception system or the assisted driving perception system. Information on surrounding traffic participants is obtained through the autonomous driving perception system or the assisted driving perception system.

[0074] In some embodiments, the current surrounding information of the target vehicle at the current moment may include current surrounding information within a preset range at the current moment. The current surrounding information within the preset range may include, for example, current surrounding information within a first preset range in a first direction and current surrounding information within a second preset range in a second direction. The historical surrounding information of the target vehicle at a historical moment may include historical surrounding information within a preset range at that historical moment. The historical surrounding information within the preset range may include, for example, historical surrounding information within a first preset range in a first direction and historical surrounding information within a second preset range in a second direction.

[0075] The first direction is the direction of travel of the target vehicle, and the second direction is the direction of intersection with the first direction. For example, when the intersection is a two-way intersection, the two-way intersection is referred to as the first intersection and the second intersection. The first direction can be the direction of the first intersection, and the second direction can be the direction of the second intersection. The angle between the first direction and the second direction is the angle between the first intersection and the second intersection. The angle between the first direction and the second direction can be 90 degrees, then the first direction and the second direction are perpendicular.

[0076] When an intersection is a three-way intersection, it is designated as the first fork, the second fork, and the third fork. The first direction can be the direction of the first fork, and the second direction can be either the direction of the second fork or the direction of the third fork. When the second fork is the second direction, the angle between the first and second directions is the angle between the directions of the first and second forks; when the third fork is the third direction, the angle between the first and second directions is the angle between the directions of the first and third forks.

[0077] When an intersection is a four-way intersection, the four forks are designated as the first fork, the second fork, the third fork, and the fourth fork. The first direction can be the direction of the first fork, and the second direction can be any of the four directions. When the second fork direction is the second direction, the angle between the first and second directions is the same as the angle between the first and second fork directions. When the third fork direction is the third direction, the angle between the first and second directions is the same as the angle between the first and third fork directions. When the fourth fork direction is the second direction, the angle between the first and second directions is the same as the angle between the first and fourth fork directions.

[0078] The first preset range is the driving distance within the second preset range or a preset time period. When the second preset range is greater than the driving distance within the preset time period, the first preset range is equal to the second preset range; when the second preset range is less than the driving distance within the preset time period, the first preset range is equal to the driving distance within the preset time period. For the first preset range at the current moment, the current driving distance of the target vehicle within the preset time period can be obtained from the current moment as the starting point. It can be determined whether the current driving distance is greater than the second preset range. If the current driving distance is greater than the second preset range, the current driving distance is used as the first preset range at the current moment; if the current driving distance is less than the second preset range, the second preset range is used as the first preset range at the current moment. For the first preset range at a historical moment, the historical driving distance of the target vehicle within the preset range can be obtained from the historical moment as the starting point. It can be determined whether the historical driving distance is greater than the second preset range. If the historical driving distance is greater than the second preset range, the historical driving distance is used as the first preset range at the historical moment; if the historical driving distance is less than the second preset range, the second preset range is used as the first preset range at the historical moment. The second preset range can be 50 meters, and the preset time period can be a target time period, such as 8 seconds. Then, the driving distance within the preset time period is the product of the preset time and the vehicle speed.

[0079] After obtaining the surrounding information of the target vehicle within the target time period, the driving information of the surrounding vehicles of the target vehicle within the target time period is extracted from the surrounding information of the target vehicle within the target time period. For example, the driving information of the surrounding vehicles at the current moment is extracted from the current surrounding information of the target vehicle at the current moment, and the driving information of the surrounding vehicles at the historical moment is extracted from the historical surrounding information of the target vehicle at a historical moment.

[0080] The actual driving situation of surrounding vehicles during the target time period can be obtained by using the driving information of surrounding vehicles at the current time and the historical driving information of surrounding vehicles at the historical time.

[0081] One implementation method is to obtain the current lane position of surrounding vehicles based on current driving information, and the historical lane position of surrounding vehicles based on historical driving information. When the current lane position and the historical lane position are inconsistent, the actual driving situation of surrounding vehicles within the target time period includes lane changes. For example, if surrounding vehicles are currently in lane 1 and were historically in lane 2, then the actual driving situation within the target time period includes changing from lane 1 to lane 2.

[0082] As another implementation method, the current location of surrounding vehicles can be obtained based on the current driving information, and the historical location of surrounding vehicles can be obtained based on the historical driving information. For example, if surrounding vehicles are currently behind the traffic light intersection, and historically they are in front of the traffic light intersection, then the actual driving situation within the target time period includes passing through traffic lights.

[0083] As another implementation method, the current speed of surrounding vehicles can be obtained based on the current driving information, and the historical speed of surrounding vehicles can be obtained based on the historical driving information. For example, if the current speed of surrounding vehicles is 80 km / h and the historical speed is 40 km / h, then the actual driving situation during the target time period includes rapid acceleration.

[0084] In some embodiments, the driving information of surrounding vehicles at multiple times within a target time period can also be obtained, such as multiple times within a time period between the current time and a historical time. Based on the driving information at multiple times, the actual driving situation of surrounding vehicles within the target time period can be obtained, and the actual driving situation of surrounding vehicles within the target time period can be accurately obtained.

[0085] S102. Use the prediction module to predict the expected driving conditions of surrounding vehicles within the target time period.

[0086] The prediction module includes a prediction model trained on traffic rules. This model, based on the driving information of surrounding vehicles at a given moment and within the constraints permitted by traffic rules, predicts the high-probability driving situation at the next moment. In other words, the expected driving situation at the next moment is learned by the model based on traffic rules and the driving information from the previous moment. Here, "next moment" and "previous moment" are two adjacent moments, and the interval between them can be determined based on specific circumstances and is not limited here. Therefore, the prediction module can predict the expected driving information of surrounding vehicles at multiple moments within a target time period, thus obtaining the expected driving situation for that target time period.

[0087] In some embodiments, when the target time period is short, such as less than 5 seconds, a prediction model can be used to predict the expected driving conditions of surrounding vehicles within the target time period based on historical information about the target vehicle's surroundings. For example, if the lane markings of the lanes occupied by surrounding vehicles are obtained from the target vehicle's lane-level environmental information at historical times, such as a single solid yellow line, then the prediction model predicts that the expected driving conditions of surrounding vehicles within the target time period include driving in the current direction. Similarly, if the traffic light information of the lanes occupied by surrounding vehicles is green based on the target vehicle's historical traffic light information, then the prediction model predicts that the expected driving conditions of surrounding vehicles at the current time include proceeding through the traffic light.

[0088] In other embodiments, when the target time period is long, such as greater than 5 seconds, a prediction model can be used to predict the driving information of surrounding vehicles at multiple moments within the target time period, and the expected driving situation of surrounding vehicles within the target time period can be obtained based on the predicted driving information of surrounding vehicles at multiple moments. The target time period can be 8 seconds, and the interval between multiple moments can be 2 seconds. For example, the multiple moments include a first moment, a second moment, and a third moment. If the lane line type of the vehicle at the first moment is a single solid yellow line, then the prediction model predicts that the expected driving situation at the second moment includes driving in the current direction. If the lane line type of the vehicle at the second moment is a double solid yellow line, then the prediction model predicts that the expected driving situation at the third moment includes driving in the current direction. Therefore, the expected driving situation of the vehicle within the target time period includes driving in the current direction. For example, if the speed limit on the road where the vehicle is located at the first moment is 20 km / h-60 km / h, then the prediction model predicts that the expected driving situation at the second moment will include driving at a speed of 20 km / h-60 km / h. If the speed limit on the road where the vehicle is located at the second moment is 20 km / h-60 km / h, then the prediction model predicts that the expected driving situation at the third moment will also include driving at a speed of 20 km / h-60 km / h. Therefore, the expected driving situation of the vehicle in the target time period includes no acceleration. For example, if the traffic light information for the lane where the vehicle is located at the first moment is red and the red light time is greater than 60 seconds, the prediction model predicts that the expected driving situation at the second moment will include stopping. If the traffic light information for the lane where the vehicle is located at the second moment is still red and the red light time is greater than 30 seconds, the prediction model predicts that the expected driving situation at the third moment will include stopping. Therefore, the expected driving situation of the vehicle in the target time period includes stopping and stopping at traffic lights.

[0089] S103. Compare the actual driving situation with the expected driving situation. If they are inconsistent, mark the surrounding vehicles as vehicles that are about to violate the rules.

[0090] The actual driving conditions within the target time period are compared with the predicted driving conditions. If the actual driving conditions are inconsistent with the predicted driving conditions, it indicates that surrounding vehicles may not be following traffic rules and may be committing violations. These vehicles are then marked as potentially violating traffic rules. For example, if the actual driving conditions include lane changing while the predicted driving conditions include driving in the same lane, surrounding vehicles are marked as potentially violating traffic rules. Similarly, if the actual driving conditions include accelerating while the predicted driving conditions include not accelerating, surrounding vehicles are marked as potentially violating traffic rules. For example, if the actual driving conditions include proceeding through a traffic light while the predicted driving conditions include stopping at a traffic light, surrounding vehicles are marked as potentially violating traffic rules. Understandably, if the actual driving conditions are more aggressive than the predicted driving conditions, it indicates that surrounding vehicles may be committing violations.

[0091] S104. If the target vehicle is interfered with during the target time period, determine whether there is a pre-set behavior in the actual driving situation. If so, determine that the vehicle that was supposed to violate the rules is the actual vehicle that violated the rules.

[0092] After marking surrounding vehicles as vehicles likely to violate traffic rules, it is determined whether the target vehicle is interfered with during the target time period. When the target vehicle is interfered with, it indicates that vehicles around the target vehicle may be engaging in traffic violations that affect the target vehicle's normal driving. For example, if the target vehicle suddenly decelerates or brakes abruptly, it indicates that vehicles in front of the target vehicle may be driving abnormally. Therefore, it can be determined whether the target vehicle's speed changes exceed a threshold during the target time period. If so, it indicates that the target vehicle is interfered with during the target time period.

[0093] If the target vehicle is disturbed during the target time period, it is determined whether there is any pre-set behavior in the actual driving situation of the surrounding vehicles. Pre-set behavior may include at least one of the following: changing lanes over a solid line, speeding, running a red light, or not driving in the designated lane. If the pre-set behavior is found in the actual driving situation, the vehicle that was supposed to violate the rules is determined to be the actual vehicle that violated the rules, meaning that the surrounding vehicles of the target vehicle did indeed commit a violation.

[0094] The driving behavior recognition method provided in this application acquires the actual driving situation and the expected driving situation for a target time period. When the actual driving situation and the target driving situation are inconsistent, it indicates that surrounding vehicles may be committing traffic violations, and the surrounding vehicles are marked as vehicles that are about to commit violations. If the target vehicle is disturbed during the target time period and a preset behavior exists in the actual driving situation, then the vehicle that is about to commit a violation is determined to be the actual vehicle that committed a violation, thereby accurately identifying vehicles around the target vehicle that are committing violations.

[0095] Figure 3 A flowchart illustrating a driving behavior recognition method according to an embodiment of this application is shown. Figure 3 As shown, with the vehicle infotainment system as the executing entity, the method in this embodiment may include the following steps:

[0096] S201. Extract the driving information of surrounding vehicles during the target time period from the surrounding information of the target vehicle during the target time period, and obtain the actual driving situation of surrounding vehicles during the target time period based on the driving information of surrounding vehicles during the target time period.

[0097] S202. Use the prediction module to predict the expected driving conditions of surrounding vehicles within the target time period.

[0098] S203. Compare the actual driving situation with the expected driving situation. If they are inconsistent, mark the surrounding vehicles as vehicles that are about to violate the rules.

[0099] Among them, step S201 and Figure 2Step S101 in the embodiment is implemented in a similar manner, and step S202 is similar to... Figure 2 Step S102 in the embodiment is implemented in a similar manner, and step S203 is similar to... Figure 2 The implementation of step S103 in the embodiment is similar, and will not be repeated here.

[0100] S204. If the target vehicle is interfered with during the target time period, determine whether there is a pre-set behavior in the actual driving situation. If so, determine that the vehicle that was supposed to violate the rules is the actual vehicle that violated the rules.

[0101] To determine whether a target vehicle is interfered with within a target time period, the vehicle speed at any time T and the speed at time T-4 can be used to determine whether the target vehicle is interfered with within the target time period. If -2m / s exists within 4 seconds... 2 If the vehicle decelerates by 50% or to 30 km / h within 4 seconds, it may be considered that the target vehicle has been interfered with. For example, if the current speed is 5 km / h and the speed was 30 km / h 4 seconds ago, the target vehicle can be considered to have been interfered with during the target time period. When the target vehicle is interfered with during the target time period, the likelihood of surrounding vehicles committing traffic violations increases further. Then, it is determined whether the actual driving situation of surrounding vehicles includes at least one of the following: changing lanes over a solid line, speeding, running a red light, or not driving in the designated lane. If so, the vehicle suspected of committing a traffic violation is identified as the actual violating vehicle.

[0102] S205, Issue a prompt voice.

[0103] When a vehicle suspected of a traffic violation is confirmed to be actually in violation, the vehicle's infotainment system issues a voice prompt to alert surrounding vehicles that the vehicle has committed the violation. Therefore, after hearing the voice prompt, the user can submit a report through the app.

[0104] S206. Upon receiving a report instruction, identify the identification information of surrounding vehicles and send the identification information to the alarm server.

[0105] After receiving a report instruction, the vehicle-mounted system identifies the identification information of surrounding vehicles, such as license plate numbers, and sends this information to the alarm server or client. Upon receiving the identification information, the alarm server or client can then send the information, along with relevant data, to the traffic management system to report the traffic violations committed by the surrounding vehicles.

[0106] The driving behavior recognition method provided in this application can send the information of a vehicle that has committed a traffic violation to an alarm server or client after the vehicle has been identified, so as to report the vehicle and reduce the adverse consequences of the violation.

[0107] Figure 4This application provides a schematic diagram of the structure of a driving behavior recognition device according to an embodiment of the present application. Figure 4 As shown, the driving behavior recognition device 10 in this embodiment is used to implement the operation corresponding to the vehicle system in any of the above method embodiments. The driving behavior recognition device 10 in this embodiment includes:

[0108] The acquisition module 11 is used to extract the driving information of surrounding vehicles in the target time period from the surrounding information of the target vehicle in the target time period, and to obtain the actual driving situation of surrounding vehicles in the target time period based on the driving information of surrounding vehicles in the target time period.

[0109] Prediction module 12 is used to predict the expected driving conditions of surrounding vehicles within a target time period;

[0110] The comparison module 13 is used to compare the actual driving situation with the expected driving situation. If they are inconsistent, the surrounding vehicles are marked as vehicles that are about to violate the rules.

[0111] The judgment module 14 is used to determine whether there is a preset behavior in the actual driving situation if the target vehicle is disturbed during the target time period. If so, the vehicle that was supposed to violate the rules is determined to be the vehicle that actually violated the rules.

[0112] The driving behavior recognition device 10 provided in this application embodiment can execute the above method embodiment. Its specific implementation principle and technical effect can be found in the above method embodiment, and will not be repeated here.

[0113] Figure 5 A schematic diagram of the hardware structure of an electronic device according to an embodiment of this application is shown. Figure 5 As shown, the electronic device 20 in this embodiment may include: a memory 21, a processor 22, and a communication interface 23.

[0114] The memory 21 is used to store computer instructions. The memory 21 may include high-speed random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device, and may also be a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.

[0115] Processor 22 is used to execute computer instructions stored in memory to implement the driving behavior recognition method in the above embodiments. For details, please refer to the relevant descriptions in the foregoing method embodiments. The processor 22 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0116] Alternatively, the memory 21 can be either standalone or integrated with the processor 22.

[0117] The communication interface 23 can be connected to the processor 22. The processor 22 can control the communication interface 23 to realize the functions of receiving and sending information.

[0118] The electronic device provided in this embodiment can be used to execute the driving behavior recognition method described above. Its implementation method and technical effect are similar, and will not be described again here.

[0119] This application also provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, are used to implement the methods provided in the various embodiments described above.

[0120] This application also provides a computer program product including computer instructions stored in a computer-readable storage medium. At least one processor of the device can read the computer instructions from the computer-readable storage medium, and the at least one processor executes the computer instructions to cause the device to perform the methods provided in the various embodiments described above.

[0121] This application also provides a chip including a memory and a processor. The memory is used to store computer instructions, and the processor is used to call and execute the computer instructions from the memory, causing a device equipped with the chip to perform the methods described in the various possible embodiments described above.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A driving behavior recognition method, characterized in that, include: Extract the driving information of surrounding vehicles during the target time period from the surrounding information of the target vehicle during the target time period, and obtain the actual driving situation of the surrounding vehicles during the target time period based on the driving information of the surrounding vehicles during the target time period. The prediction module is used to predict the expected driving conditions of surrounding vehicles within the target time period. The prediction module includes a prediction model, which is trained based on traffic rules. When the target time period is short, the prediction model is used to predict the expected driving situation of surrounding vehicles within the target time period based on the surrounding information of the target vehicle at historical times; when the target time period is long, the prediction model is used to predict the driving information of surrounding vehicles at multiple times within the target time period, and the expected driving situation of surrounding vehicles within the target time period is obtained based on the predicted driving information of surrounding vehicles at multiple times. The actual driving situation is compared with the expected driving situation. If they are inconsistent, the surrounding vehicles are marked as vehicles that are about to violate traffic rules. Determine whether the target vehicle experiences a speed change greater than a threshold within the target time period. If so, check the surrounding information during the target time period to see if there is at least one of the following in the actual driving situation: changing lanes over a solid line, speeding, running a red light, or not driving in the designated lane. If so, determine that the vehicle suspected of violating traffic rules is actually violating traffic rules.

2. The method according to claim 1, characterized in that, The method further includes: Upon receiving a report instruction, the system identifies the identification information of surrounding vehicles and sends the identification information to the alarm server.

3. The method according to claim 1, characterized in that, The target time period includes the time period between the current moment and historical moments; The step of extracting the driving information of surrounding vehicles during the target time period from the surrounding information of the target vehicle specifically includes: Obtain the current surrounding information of the target vehicle at the current moment and the historical surrounding information at historical moments; Extract the current driving information of surrounding vehicles at the current time from the current surrounding information, and extract the historical driving information of surrounding vehicles at the historical time from the historical surrounding information.

4. The method according to claim 3, characterized in that, The acquisition of the target vehicle's current surrounding information at the current moment and its historical surrounding information at historical moments specifically includes: Obtain current surrounding information of the target vehicle within a preset range at the current time, as well as historical surrounding information within the preset range at historical times.

5. The method according to claim 3, characterized in that, The surrounding information includes: lane-level environmental information, traffic light information, and traffic participant information; The lane-level environmental information includes: At least one of lane conditions, lane width, lane marking type, and lane speed limit, wherein the lane conditions include going straight and / or turning; The traffic light information includes the traffic light information corresponding to each lane; The traffic participant information includes the driving information of the surrounding vehicles.

6. The method according to claim 5, characterized in that, Obtain the current surrounding information of the target vehicle at the current moment and the historical surrounding information at historical moments, specifically including: The autonomous driving perception system or the driver assistance perception system acquires the current lane-level environmental information of the target vehicle at the current moment and the historical lane-level environmental information at historical moments; or, Obtain the current location and historical location of the target vehicle, and query the current lane-level environmental information of the current location and the historical lane-level environmental information of the historical location on the high-precision map.

7. The method according to claim 4, characterized in that, The acquisition of the target vehicle's current surrounding information within a preset range at the current moment and its historical surrounding information within the preset range at historical moments specifically includes: Obtain current surrounding information of the target vehicle within a first preset range in the first direction and a second preset range in the second direction at the current time, as well as historical surrounding information within the first preset range in the first direction and the second preset range in the second direction at historical times; The first direction is the direction of travel of the target vehicle, and the second direction is the direction that intersects with the first direction; When the target vehicle travels a distance greater than the second preset range within a preset time, the travel distance is taken as the first preset range; when the target vehicle travels a distance less than the second preset range within a preset time, the second preset range is taken as the first preset range.

8. The method according to any one of claims 3-7, characterized in that, The actual driving situation of the surrounding vehicles during the target time period is obtained based on the driving information of the surrounding vehicles during the target time period, specifically including: Based on the current driving information, obtain the current lane where the surrounding vehicles are currently located; Based on the historical driving information, the historical lanes where the surrounding vehicles were located at historical moments are obtained. If the current lane and the historical lane are inconsistent, then the actual driving situation of the surrounding vehicles during the target time period includes lane changes.

9. The method according to any one of claims 1-7, characterized in that, The actual driving situation is compared with the expected driving situation. If they do not match, the surrounding vehicles are marked as vehicles that are about to violate traffic rules. Specifically, this includes: Compare the actual driving conditions with the expected driving conditions; If the actual driving situation includes lane changing and the expected driving situation includes driving in the same lane, then the surrounding vehicles are marked as vehicles that are about to violate the rules.

10. The method according to any one of claims 1-7, characterized in that, The method further includes: A prompt voice is issued to indicate that the surrounding vehicles are actually violating traffic rules.

11. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the driving behavior recognition method as described in any one of claims 1 to 10.

Images