TWI788173B - Vehicle fleet management system - Google Patents

Abstract

A vehicle fleet management system, including an edge device and a cloud device, the edge device is set on a vehicle, including a computing module, a proximity event detection module, a post-event detection module, and an image capture module, a GPS module and a mobile status module, the cloud device is connected to the edge device in communication, the cloud device includes a driving behavior analysis module and a database, and the approach event detection module judges that the vehicle satisfies When the approaching event condition is met, an approaching event warning is issued to a driver of the vehicle. When the post-event detection module judges that the vehicle meets the post-event condition, it is determined that a violation event has occurred. The driving behavior analysis module according to the The proximity event, the violation event, and a driving record information generate an information assessing a safety of the driver.

Description

Vehicle Fleet Management System

The present invention relates to a vehicle fleet management system, and in particular to a vehicle fleet management system capable of detecting and warning dangerous driving in advance and effectively evaluating driver behavior.

Vehicle fleet management is widely used in vehicles such as cars or trucks driving on the road or land vehicles (such as forklift trucks or pallet trucks) driving in a specific field. Existing Most of the systems and methods focus on using positioning to manage the driving route of the vehicle and analyzing the acquired images, so as to manage the vehicle and/or driving.

For example, Chinese Invention Patent Publication No. CN112802344A, U.S. Invention Publication No. US10643477B2, Publication No. US20140114502A1, Publication No. US10671869B2, Publication No. US10719725B2, etc.

However, in the prior art, it is impossible to detect or judge dangerous driving behaviors or violations, thus increasing the risk of accidents. For fleet managers, if they want to effectively manage and control, they must monitor the driving recorder on the vehicle all the time and make judgments. This will consume a lot of manpower and money costs, and it is extremely difficult to implement . Therefore, the existing vehicle fleet management system needs to be improved.

The main purpose of the present invention is to solve the problem that the existing vehicle fleet management system cannot effectively detect or judge dangerous driving behaviors or illegal behaviors, thus resulting in difficult risk management and control.

In order to achieve the above object, the present invention provides a vehicle fleet management system, including: an edge device installed on a vehicle, including a computing module, a proximity event detection module, a post-event detection module, and an image An acquisition module, a GPS module and a movement status module, the computing module is configured to obtain at least one vehicle surrounding image and determine at least one traffic instruction according to the vehicle surrounding image; a cloud device, connected to the The edge device, the cloud device includes a driving behavior analysis module and a database; wherein, the proximity event detection module obtains a proximity event condition according to the traffic instruction, and the proximity event detection module obtains a proximity event condition from the moving state module Obtain at least one movement status information of the vehicle in a first time interval and obtain at least one GPS information of the vehicle in the first time interval from the GPS module, and determine whether the vehicle satisfies the approach event condition, and when the approach is satisfied When the event condition is determined as a proximity event, the proximity event detection module sends a proximity event warning to a driver of the vehicle through an output module arranged on the vehicle; wherein, the post-event detection module according to The traffic instruction obtains a post-event condition, and the post-event detection module obtains at least one moving state information of the vehicle in the first time interval and at least one moving state information of the vehicle in a second time interval from the moving state module And obtain at least one GPS information of the vehicle in the first time interval and at least one GPS information of the second time interval from the GPS module, and judge whether the vehicle satisfies the post-event condition, when the post-event condition is satisfied determined as a violation event; and, wherein the database is configured to store the approach event, the violation event and a driving record information of the driver, the driving behavior analysis module according to the approach event, the violation event and the The driving record information generates an information for evaluating a safety of the driver.

In order to achieve the above purpose, the present invention also provides a vehicle fleet management system, including: an edge device installed on a vehicle, including a computing module, a proximity event detection module, a post-event detection module, a An image capture module, a GPS module and a movement status module, the computing module is configured to obtain at least one vehicle surrounding image and determine at least one target according to the vehicle surrounding image; a cloud device, connected to the The edge device, the cloud device includes a driving behavior analysis module and a database; wherein, the proximity event detection module obtains a proximity event condition according to the target, and the proximity event detection module obtains from the movement state module At least one moving state information of the vehicle in a first time interval and obtain at least one GPS information of the vehicle in the first time interval from the GPS module, and according to the moving state information and the GPS in the first time interval The information judges whether the vehicle is traveling on a possible collision route with the target to confirm whether an approach event condition is met, and when the approach event condition is met, it is determined as an approach event, and the approach event detection module is set on the An output module on the vehicle sends an approach event warning to a driver of the vehicle; wherein, the post-event detection module obtains a post-event condition according to the target, and the post-event detection module obtains a post-event condition from the moving state The module obtains at least one moving state information of the vehicle in the first time interval and at least one moving state information in a second time interval, and obtains the at least one GPS information and the GPS module in the first time interval At least one GPS information of the vehicle in the second time interval, and determine whether the vehicle meets the post-event condition, and when the post-event condition is satisfied, the post-event detection module determines that it is a violation event; and, wherein the The database is configured to store the approach event, the violation event and a driving record information of the driver, and the driving behavior analysis module generates and evaluates the driver's safety according to the approach event, the violation event and the driving record information sexual information.

Herein, although some embodiments perform steps in a particular order, the steps may be performed in another reasonable order. Certain features described below may be substituted or eliminated for different embodiments. It should be understood that some additional operations may be performed before, during, or after the recited method, and that certain operations may be replaced or omitted in other embodiments of the method.

The terminology used in describing various embodiments herein is for the purpose of describing particular examples only and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" also include the plural forms unless the context clearly dictates otherwise, or the number of elements is intentionally limited.

The present invention discloses a vehicle fleet management system, referring to "Fig. 1", in one embodiment, the system includes an edge device 10, a cloud device 20 and an external data source 30, the edge device 10 is a set in a vehicle In-vehicles device (in-vehicles device), the edge device 10 includes a processing unit 11, a mobile state module 12, an image capture module 13, a GPS module 14, an output module 15, a A first wireless communication module 16 and a memory unit 17 . In one embodiment, the processing unit 11 can be realized by a processor, and the processing unit 11 includes an operation module 111, an almost (or nearing) event detection module 112 and a post-event ( After event) detection module 113, the mobile status module 12 can be realized by one or more sensors, including a gyroscope 121, an accelerometer 122 and an ignition status detection module (Ignition Status monitor module) 123 Wait.

The cloud device 20 can be realized by a server or a computer, the cloud device 20 includes a processing unit 21, a database 22 and a second wireless communication module 23, the processing unit 21 can be realized by a processor , the processing unit 21 includes a driving behavior analysis module 211 and a driving scoring module 212 . The first wireless communication module 16 and the second wireless communication module 23 can establish a wireless communication connection through the Internet.

In the present invention, an event refers to a collision, accident, accident, or violation of traffic laws, traffic regulations or traffic signs of the vehicle. The proximity event detection module 112 and the post-event detection module 113 respectively obtain a plurality of information about the vehicle in a driving or non-driving state from a plurality of modules, and use these information as parameters to determine whether a specified condition is satisfied. conditions of. The approach event detection module 112 detects the vehicle before the event may occur (that is, the time point before the event occurs), in other words, the judgment condition of the approach event detection module 112 is based on the vehicle's The situation at that time may be the basis of the design that the event may be about to occur, that is, the event did not occur but may constitute danger or risk; the detection module 113 after the event detects that the event has been produced (after the event) for the vehicle In other words, the judgment condition of the post-event detection module 113 is based on the fact that the event has occurred in the vehicle at that time, that is, the event has occurred.

It can be understood that the proximity event detection module 112 and the post-event detection module 113 perform detection according to the information at different times. The approach event detection module 112 collects the information of the vehicle in a first time interval, and the post-event detection module 113 collects the information of the vehicle in the first time interval and a second time interval respectively In some embodiments, the first time interval and the second time interval are consecutive times, such as before or after the vehicle passes through a single intersection. For example, if it is 5 seconds before the vehicle passes through the intersection, the second time interval is 5 seconds after the vehicle passes through the intersection. The approach event detection module 112 judges that an approach event condition is satisfied when the vehicle is passing through the intersection, then the approach event detection module 112 obtains and based on the information of the first time interval; when the event After the detection module 113 judges that a post-event condition is satisfied 5 seconds after the vehicle passes through the intersection, the detection module 113 after the event obtains and uses the first time interval and the second time interval The information can also be understood as the information between the first time interval and the second time interval (from 5 seconds before passing the intersection to 5 seconds after passing the intersection). For example, the post-event detection module 113 needs to obtain at least one surrounding image of the vehicle in the first time interval to identify at least one traffic instruction or at least one target, and must be based on the vehicle in the first time interval and (or to ) at least one moving state information and at least one GPS information in the second time interval, in order to know the state of the vehicle passing the traffic instruction or the target at or around the time, so as to determine whether the post-event condition is satisfied.

The moving state module 12 is configured to continuously or periodically obtain at least one moving state information of the vehicle, for example, the moving state information includes obtaining a three-axis angular velocity of the vehicle from the gyroscope 121, and obtaining from the accelerometer 122 obtains a three-axis acceleration of the vehicle, and detects from the ignition state detection module 123 that the engine of the vehicle is in a start state or an off state, etc. In one example, the mobile state module 12 is coupled to For the image capture module 13, the moving state information further includes obtaining one or more surrounding images of the vehicle from the image capture module 13. The GPS module 14 is configured to obtain at least one piece of GPS information of the vehicle, for example, the GPS information such as a position of the vehicle, a moving direction of the vehicle and a speed of the vehicle.

The image capturing module 13 can be realized by using one or more cameras. The image capturing module 13 obtains one or more surrounding images of the vehicle. In one example, the surrounding image of the vehicle is an image in front of the vehicle. In an example, the surrounding image of the vehicle may also be (or further include) a rear image or a side image of the vehicle. The image capturing module 13 transmits the surrounding image of the vehicle to the processing unit 11 , and the computing module 111 can use data mining to determine any one of at least one traffic instruction or at least one target from the surrounding image of the vehicle. The traffic instruction is at least any one of a traffic sign, a traffic light, a ground sign or a ground marking, and the target is a pedestrian, an animal, a land vehicle (such as a motorcycle, a bicycle, a car, at least any one of a truck or a bus, etc.) or a moving or fixed obstacle.

The output module 15 can be realized by a horn, a display screen or through a steering wheel, which is configured to send an approach event alert to a driver of the vehicle. The approach event alert can be a visual alert, an audio alert, A tactile alert or any combination of the foregoing.

The memory unit 17 pre-stores a data, which may include a plurality of traffic instruction information, a plurality of target information, a plurality of approach event conditions corresponding to the traffic instruction information and the target information, and corresponding to the traffic instructions Information and a plurality of post-event conditions of the target information. In an example, the data may be presented by a plurality of rules. For example, the data is configured to be used by the processing unit 11 to control the vehicle The surrounding image performs a rule-based judgment on the traffic instruction, the target, the approaching event condition, the post-event condition, and the like. It can be understood that one of the traffic instruction information (or one of the target information) may correspond to one or more of the approaching event condition/the post-event condition. For example, when the traffic instruction information is that the speed limit must not exceed 50km/h, the approach event condition is that the speed of the vehicle is between 40km/h and 50km/h and the acceleration is between 1m/s ² to 2m/ Between s and ² , the post-event condition is that the speed of the vehicle is greater than 50km/h; or, when the target information is a pedestrian, the approach event condition is that the pedestrian is located within 15 degrees of the expected moving direction of the vehicle In one range, the distance between the vehicle and the pedestrian is less than 20 meters, the speed of the vehicle is greater than 40 km/h, and the acceleration of the vehicle is greater than -5m/s ² .

During execution, the image capturing module 13 obtains the surrounding image of the vehicle, the moving state module 12 obtains the moving state information of the vehicle, and the GPS module 14 obtains the GPS information of the vehicle. When it is determined from the surrounding image of the vehicle that the traffic indication appears around the vehicle, the approach event detection module 112 obtains the corresponding approach event condition according to the traffic indication, and the approach event detection module 112 judges the vehicle's Whether the moving state information and the GPS information satisfy the proximity event condition. When the proximity event condition is satisfied (for example, according to the movement status information and the GPS information of the vehicle in the first time interval), the proximity event detection module 112 determines that a proximity event occurs, and is configured on the vehicle The output module 15 sends out the proximity event warning.

On the other hand, when it is determined from the surrounding image of the vehicle that the target appears around the vehicle, the approach event detection module 112 obtains the corresponding approach event condition according to the target. Then, the proximity event detection module 112 judges whether the vehicle is driving on a possible collision route with the target according to the target in the vehicle peripheral image, the moving state information of the vehicle and the GPS information, and if so, The approach event detection module 112 further determines whether the moving state information and the GPS information of the vehicle in the first time interval satisfy the approach event condition. When the proximity event condition is met, the proximity event detection module 112 issues the proximity event warning through the output module 15 disposed on the vehicle. In some embodiments, in addition to the movement status information and the GPS information in the first time interval, the approach event detection module 112 also needs to obtain the surrounding image of the vehicle in the first time interval to determine Whether the vehicle is eligible for the approach event.

The post-event detection module 113 obtains the post-event condition according to the traffic instruction and/or the target, and the post-event detection module 113 obtains the moving state information and the GPS information (such as the first time interval) of the vehicle and the movement state information and the GPS information of the second time interval), and determine whether the movement state information and the GPS information of the vehicle meet the post-event condition, and when the post-event condition is met, the post-event detection The module determines that a violation has occurred. In some embodiments, in addition to the movement status information and the GPS information of the first time interval and the second time interval, the post-event detection module 113 also needs to obtain the first time interval and/or The surrounding image of the vehicle during the second time interval to determine whether the vehicle meets the post-event condition (for example, confirm whether the vehicle crosses the vehicle surrounding image displayed between the first time interval and the second time interval red light).

In other embodiments, the proximity event detection module 112 or the post-event detection module 113 can further obtain at least one vehicle-related auxiliary information to assist judgment in addition to obtaining the moving state information and the GPS information , such as obtaining the rotation direction and range of a steering wheel or the information of a direction light or a brake light from an on-vehicle diagnostic system (OBD or OBD-II).

[FIG. 2] illustrates the judgment process according to an embodiment of the present invention. The proximity event detection module 112 is based on the traffic instruction and/or the target information (block 100), the moving state information (block 200) and the GPS Information (block 300) determines whether the approach event occurs (blocks 400, 500); after the event, the detection module 113 is based on the traffic instruction and/or the target information (block 100), the moving state information (block 200) and the GPS information (block 300) to determine whether the violation event occurs (blocks 600, 700).

When the proximity event detection module 112 detects the proximity event and the post-event detection module 113 detects the violation event, the proximity event detection module 112 and the post-event detection module 113 respectively The proximity event and the violation event are stored in the memory unit 17 and/or sent to the database 22 of the cloud device 20 . [Fig. 3] shows that a fleet includes multiple target vehicles 40a, 40b, 40c, each target vehicle 40a, 40b, 40c is equipped with the edge device 10a, 10b, 10c respectively, the edge device 10a, 10b, 10c are respectively connected to the cloud device 20 by communication, and the cloud device 20 collects the approach event and the violation event from the edge devices 10a, 10b, 10c, and collects a driving record information about the driver from the external data source 30 , the driving record information includes a ticket record of the driver, including the number of tickets and the type of the driver. Accordingly, the driving behavior analysis module 211 can analyze a driving behavior of the driver according to the data 800 of the approach event, the data 810 of the violation event, the data 820 of the ticket and a correlation 830 of the aforementioned data.

Further, the driving behavior analysis module 211 can analyze the driving behavior of the driver according to any one or more of the following parameters: 1. A count of the proximity event; 2. A type of the proximity event; 3. The number of violations; 4. A type of the violation; 5. The number of times the ticket was issued; 6. A type of the ticket; 7. A correlation between the approach event and the breach event; and 8. A correlation between the approach, the violation, and the ticket.

The driving scoring module 212 judges a safety information of the driver according to a result generated by the driving behavior analysis module 211 . It can be understood that when judging the relevance, the approach event, the violation event and the ticket in adjacent time intervals can be used as the basis for judgment, for example. If driver A triggers the approach event before the intersection, but does not trigger the violation event after the approach event (passing through the intersection), and does not receive the ticket; driver B triggers the approach event before the intersection , the violation event is also triggered after the approach event (passing through the intersection), then the driving behavior of driver A is judged to be better than that of driver B, because driver A takes corresponding measures after receiving the warning of the approach event. measures, such as slowing down.

The following table is an example of the records of multiple drivers in a convoy. According to the data, the numbers 900 and 903 are drivers with a medium risk level. Compared with the number 903, the number 900 has a better response to the approach event warning ; Number 901 is a high-risk driver; number 902 is a low-risk driver; number 904 is a good driver. driver Proximity Events Number of violations relevance Number of tickets No reaction Responsive 900 4 0 0 4 0 901 6 4 4 2 2 902 2 0 0 0 0 903 3 2 2 1 1 904 0 0 0 0 0

Whether there is a reaction in the above table refers to whether the driver immediately corrects the driving behavior at that time after receiving the warning of the approaching event, so as to avoid the triggering of the violation event.

A few examples are listed below to illustrate the judgment of the approaching event and the violation event. [Fig. The relationship between speed (and acceleration) and driving distance. In these figures, the thick solid line represents the threshold of the approaching event condition and the post-event condition, the dotted line represents the satisfaction of the approaching event condition and the post-event condition, and the thin solid line represents The proximity event condition and the post-event condition are not met.

Traffic directions are traffic lights:

In Fig. 4, when the traffic indication is a traffic light 42 showing a red light, the target vehicle 40 approaches the traffic light 42, if the target vehicle 40 arrives at the traffic light 42, the target vehicle 40 If the speed (dotted line) does not decelerate below a threshold (thick solid line), the approach event condition (point A1) is met; , that is, the proximity event condition is not met.

When the traffic light 42 is still displayed as a red light, and the target vehicle 40 passes (over) the traffic light 42 (that is, the dotted line is higher than the thick solid line), it is judged that the target vehicle 40 meets the post-event condition (point B1); otherwise , if the target vehicle 40 does not pass the traffic light 42 but stops before the traffic light 42 , the post-event condition is not satisfied.

Traffic instructions are ground markings:

In "Fig. 5", when the traffic indication is identified as an intersection without traffic lights according to the ground markings, the target vehicle 40 approaches the intersection. If the target vehicle 40 arrives at the intersection, the target vehicle 40 The speed (dotted line) of the target vehicle 40 does not decelerate below a threshold (thick solid line), that is, the approach event condition (point A2) is met; line), that is, the proximity event condition is not met.

When the target vehicle 40 passes (crosses) the crossing without decelerating, it is judged that the target vehicle 40 satisfies the post-event condition (point B2); The post-event condition.

[FIG. 6] shows that the road 41 is provided with a stop sign 43a of a bus stop and an area 43b of a bus stop. The area 43b can be recognized from the surrounding image of the vehicle to determine that the traffic instruction is the ground marking and a road section where parking is prohibited, and the stop sign 43a can also be recognized from the peripheral image of the vehicle to determine that the traffic instruction is the traffic sign And it is a road section where parking is prohibited. Here, a bus stop is used as an example. In other examples, red lines, slotted lines, or reticulated lines may also be included. If the engine of the target vehicle 40 is in the start state, and the position of the target vehicle 40 is to stay in the area 43b longer than a first stop threshold, the approach event condition is met, or, if the target vehicle 40 If the engine is in the off state, the proximity event condition is also satisfied; on the contrary, if the engine of the target vehicle 40 is in the start state, and the position of the target vehicle 40 is that the time of staying in the area 43b is shorter than the first stay threshold , that is, the proximity event condition is not met.

If the target vehicle 40 is in the activated state, and the position of the target vehicle 40 is to continue to stay in the area 43b for longer than a second stay threshold (greater than the first stay threshold), the post-event condition is satisfied, or , if the engine of the target vehicle 40 is in the off state and exceeds the second stay threshold, the post-event condition is also satisfied; otherwise, if the position of the target vehicle 40 is to stay in the area 43b for a time between the first stay threshold And between the second stay threshold, that is, the approach event condition is not met.

Traffic instructions are traffic signs:

In "Fig. 7", the traffic instruction is a traffic sign showing the speed limit (i.e. a speed limit sign 44), if the speed and acceleration of the target vehicle 40 at that time are respectively higher than a vehicle speed threshold (below the speed limit sign 44 speed limit) and an acceleration threshold, that is, the approaching event condition (point A3) is met; ) and the acceleration threshold, that is, the approach event condition is not met.

If the speed of the target vehicle 40 at that time has been higher than the speed limit of the speed limit sign 44, it is judged that the target vehicle 40 meets the post-event condition (point B3); otherwise, if the speed of the target vehicle 40 at that time is lower than the speed limit The speed limit marked 44, that is, the post-event condition is not met.

Referring to "Fig. 8", or, for example, the target vehicle 40 is close to a traffic sign 45 for pedestrians or children, if the target vehicle 40 arrives at the traffic sign 45, the speed of the target vehicle 40 is not decelerated When the target vehicle 40 falls below a threshold, the approach event condition (point A4 ) is met; otherwise, if the target vehicle 40 decelerates below the threshold, the approach event condition is not met. [Figure 8] also shows another situation, the thin solid line shows that the target vehicle 40 meets the approach event condition at point A4', the driver slows down in time after receiving the approach event warning, and arrives at the traffic sign 45 stopped with the brakes.

The example of "Fig. 9A" and "Fig. 9B" is that the traffic instruction is a traffic sign 46 that prohibits turning left. In "Fig. 9A", the target vehicle 40 approaches the traffic sign 46 (before arriving at the traffic sign 46) , if it is judged that the target vehicle 40 may or is about to turn left according to the moving state information of the target vehicle 40, the GPS information and the on-board diagnostic system, it is determined that the approach event condition is satisfied; in "Fig. 9B", After the target vehicle 40 crosses the traffic sign 46, if it is judged that the target vehicle 40 has turned left or is turning left according to the moving state information of the target vehicle 40, the GPS information and the on-board diagnostic system, it is determined that the event is met. postcondition.

In the example of "Fig. 10", the ground markings are identified from the surrounding image of the vehicle, and the lanes 41a, 41b and the driving direction 41c to be followed are identified according to the ground markings. If the moving state information of the target vehicle 40 and The GPS information judges that the target vehicle 40 may or is going to travel in the lane 41a in a wrong direction, that is, it is determined that the approach event condition is met; if the target vehicle 40 is judged to have been It is judged that the post-event condition is satisfied when the vehicle travels on the lane 41a in the wrong direction.

Target vehicle:

[Figure 11] shows that there is a surrounding vehicle 50 in front of the target vehicle 40, and [Figure 12] shows that there is a pedestrian 51 in front of the target vehicle 40, and the surrounding vehicle 50 and the pedestrian 51 are judged as the target. The position of the target (the surrounding vehicle 50 and the pedestrian 51 ) in the image, the moving state information of the target vehicle 40 and the GPS information determine whether the target vehicle 40 is driving on the possible collision route, and if so, the approaching The event detection module 112 further judges whether the moving state information and the GPS information of the target vehicle 40 meet the approach event condition; after meeting the approach event condition, if according to the moving state information and/or If the GPS information determines that the target vehicle 40 has collided with the target, it is determined that the post-event condition is met.

In the present invention, the approach event detection module is used to predict before a collision, accident, accident, or violation of traffic laws, traffic regulations or traffic signs, so as to identify dangerous driving behaviors that have not yet reached the event, and issue the Proximity events are alerted and logged. In this way, the system of the present invention can analyze driving behavior more comprehensively and objectively, compared with the driving behavior analysis technology of existing fleet vehicles that only judges violations that have occurred. In addition, when the approach event is detected before the event occurs, it is possible to further track whether the driver has triggered the post-event condition (that is, the occurrence of the violation event) after the approach event, so as to judge the driver according to the correlation After receiving the warning, whether there is a corresponding correction in his driving behavior, it is possible to analyze the driver's mentality more deeply, and to understand the driving behavior of each driver in the fleet in detail. On the other hand, the present invention uses the proximity event detection module to warn the driver before the event occurs, which can greatly reduce and avoid the occurrence of accidents and violations, thereby improving driving safety and benefiting the fleet management.

10, 10a, 10b, 10c: Edge devices 11: Processing unit 111: Operation module 112:Proximity event detection module 113: Post-event detection module 12: Mobile state module 121: Gyroscope 122: Accelerometer 123:Ignition status detection module 13: Image capture module 14:GPS module 15: Output module 16: The first wireless communication module 17: Memory unit 20: Cloud device 21: Processing unit 211:Driving Behavior Analysis Module 212:Driving scoring module 22: Database 23: Second wireless communication module 30: External sources of information 40, 40a, 40b, 40c: target vehicle 41: road 41a, 41b: Lane 41c: Driving direction 50: Surrounding vehicles 51: Pedestrian 42: Traffic light 43a: stop sign 43b: area 44:Speed limit sign 45:Traffic signs 46: Traffic signs 800:Proximity event information 810: Information on violations 820: Information about fines 830: Relevance 900, 901, 902, 903, 904: number

[FIG. 1] illustrates a schematic diagram of a system architecture according to an embodiment of the present invention. [Fig. 2] illustrates the judgment flow according to an embodiment of the present invention. [FIG. 3] illustrates a schematic diagram of a system architecture according to an embodiment of the present invention. [FIG. 4] to [FIG. 12] illustrate schematic diagrams of judging approach events and violation events according to different examples of the present invention.

10: Edge devices

11: Processing unit

111: Operation module

112:Proximity event detection module

113: Post-event detection module

12: Mobile state module

121: Gyroscope

122: Accelerometer

123:Ignition status detection module

13: Image capture module

14:GPS module

15: Output module

16: The first wireless communication module

17: Memory unit

20: Cloud device

21: Processing unit

211:Driving Behavior Analysis Module

212:Driving scoring module

22: Database

23: Second wireless communication module

30: External sources of information

Claims (14)

A vehicle fleet management system, comprising: an edge device set on a vehicle, including a computing module, a proximity event detection module, a post-event detection module, an image capture module, and a GPS module A group and a moving state module, the computing module is configured to obtain at least one surrounding image of the vehicle and determine at least one traffic instruction according to the surrounding image of the vehicle; a cloud device is connected to the edge device through communication, and the cloud device includes a A driving behavior analysis module and a database; wherein, the approach event detection module acquires an approach event condition according to the traffic instruction, and the approach event detection module acquires the vehicle at a first time from the moving state module At least one movement status information of the interval and obtain at least one GPS information of the vehicle in the first time interval from the GPS module, and judge whether the vehicle satisfies the approach event condition, and determine as an approach event when the approach event condition is satisfied , the proximity event detection module sends a proximity event warning to a driver of the vehicle through an output module arranged on the vehicle; wherein, the post-event detection module obtains a post-event condition according to the traffic instruction , the post-event detection module obtains at least one piece of moving state information of the vehicle in the first time interval and at least one piece of moving state information of a second time interval from the moving state module, and obtains the moving state information from the GPS module At least one GPS information of the vehicle in the first time interval and at least one GPS information of the second time interval to determine whether the vehicle meets the post-event condition, and determine as a violation event when the post-event condition is met; and wherein , the database is configured to store the approach event warning, the violation event and a driving record information of the driver, and the driving behavior analysis module generates and evaluates the driver according to the approach event, the violation event and the driving record information A security information. The vehicle fleet management system according to claim 1, wherein the traffic instruction is at least any one of a traffic sign, a traffic light, a ground sign or a ground marking. The vehicle fleet management system as claimed in claim 1, wherein the driving behavior analysis module generates the information for evaluating the safety of the driver according to a correlation between the approach event, the violation event and the driving record information . The vehicle fleet management system as described in Claim 1, wherein the driving behavior analysis module generates and evaluates the safety of the driver according to the number of the approaching event, the number of the violation event, and the number of tickets issued by the driving record information of this information. The vehicle fleet management system as claimed in claim 1, wherein the proximity event warning is a visual warning, an audio warning, a tactile warning or any combination thereof. The vehicle fleet management system as claimed in claim 1, wherein the first time interval is earlier than the second time interval and adjacent to each other. The vehicle fleet management system as claimed in claim 1, wherein the driving record information at least includes a ticket number and a ticket type. A vehicle fleet management system, comprising: an edge device set on a vehicle, including a computing module, a proximity event detection module, a post-event detection module, an image capture module, and a GPS module A group and a moving state module, the computing module is configured to obtain at least one vehicle surrounding image and determine at least one target according to the vehicle surrounding image; a cloud device, connected to the edge device in communication, the cloud device includes a driver Behavior analysis module and a database; wherein, the approach event detection module obtains an approach event condition according to the target, and the approach event detection module obtains the vehicle's position in a first time interval from the moving state module at least one move State information and obtain at least one GPS information of the vehicle in the first time interval from the GPS module, and judge whether the vehicle is driving in a direction with the target according to the movement state information and the GPS information in the first time interval on the possible collision route to confirm whether a proximity event condition is met, and when the proximity event condition is met, it is determined as a proximity event, and the proximity event detection module sends a proximity event warning through an output module arranged on the vehicle For a driver of the vehicle; wherein, the post-event detection module obtains a post-event condition according to the target, and the post-event detection module obtains the vehicle's position in the first time interval from the moving state module at least one piece of movement state information and at least one piece of movement state information in a second time interval, and obtain at least one piece of GPS information of the vehicle in the first time interval and at least one GPS information in the second time interval from the GPS module , and determine whether the vehicle meets the post-event condition, and when the post-event condition is satisfied, it is determined as a violation event; and wherein the database is configured to store the approach event warning, the violation event, and a driving of the driver record information, the driving behavior analysis module generates information for evaluating the safety of the driver according to the approach event, the violation event and the driving record information. The vehicle fleet management system according to claim 8, wherein the target is at least any one of a pedestrian, an animal, a bicycle, a land vehicle, or a moving or fixed obstacle. The vehicle fleet management system as claimed in claim 8, wherein the driving behavior analysis module generates the information for evaluating the safety of the driver according to a correlation between the approach event, the violation event and the driving record information . The vehicle fleet management system as described in Claim 8, wherein the driving behavior analysis module generates and evaluates the safety of the driver according to the number of the approach event, the number of the violation event, and the number of tickets issued by the driving record information of this information. The vehicle fleet management system as claimed in claim 8, wherein the proximity event warning is a visual warning, an audio warning, a tactile warning or any combination thereof. The vehicle fleet management system as claimed in claim 8, wherein the first time interval is earlier than the second time interval and adjacent to each other. The vehicle fleet management system as claimed in claim 8, wherein the driving record information at least includes a ticket number and a ticket type.

Images