A vehicle-mounted danger warning system and method

Abstract

The present application belongs to the field of vehicle-mounted danger warning technology, and discloses a vehicle-mounted danger warning system and method; in the driving area of a vehicle, a monitoring area is set for the side with the risk of harassment, the monitoring area is divided into n sub-areas along the length direction of the vehicle body, and each sub-area is numbered in turn; videos in each sub-area are continuously collected, and whether a rod-shaped object exists is judged; whether a deceleration instruction or an automatic driving instruction is generated in turn is determined according to the number of rod-shaped objects entering the sub-area; traffic condition information of the vehicle is collected in real time; the driving environment of the vehicle is evaluated according to the traffic condition information, and a road condition evaluation index is generated; the driving safety and the life and property of passengers are comprehensively ensured from being infringed.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-mounted hazard warning technology, and more particularly, to a vehicle-mounted hazard warning system and method. Background Art

[0002] Public buses, thanks to their convenience and affordability, have become the preferred mode of transportation for many. However, with this growth in passenger traffic, some difficult issues have arisen. For example, some passengers, due to inadvertently missing their stops or rushing to board the wrong bus, can easily get into arguments with the driver, which can even escalate into physical altercations. In recent years, traffic accidents caused by passengers grabbing the steering wheel have been reported frequently in the news, leaving people heartbroken. While certain protective measures have been implemented in the driving area, existing response measures remain insufficient when such emergencies occur.

[0003] The act of grabbing the steering wheel usually occurs without warning and in an instant, making it difficult to defend against. It is difficult for the driver to react in an instant and unable to stabilize the steering wheel while ensuring the safe stop of the vehicle.

[0004] In view of this, the present invention proposes a vehicle-mounted hazard warning system and method to solve the above-mentioned problem. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned objectives, the present invention provides a vehicle-mounted hazard warning method, comprising:

[0006] Within the vehicle's driving area, a monitoring area is set for the side with a harassment risk. The monitoring area is divided into n sub-areas along the length of the vehicle body, where n is a positive integer greater than 1, and each sub-area is numbered sequentially;

[0007] Continuously capture video in each sub-area to determine whether there are rod-shaped objects. Depending on the number of rod-shaped objects entering the sub-area, it determines whether to generate deceleration instructions or automatic driving instructions.

[0008] Collect the vehicle's traffic condition information in real time;

[0009] Evaluate the vehicle driving environment based on traffic condition information and generate road condition evaluation indicators;

[0010] When a deceleration instruction is generated, the vehicle is controlled to decelerate. If an automatic driving instruction is generated, whether to execute the automatic driving instruction is determined based on the road condition assessment index at that time. If the automatic driving instruction is executed, the connection between the steering wheel and the steering mechanism is disconnected, and the vehicle is switched to automatic driving mode. If the automatic driving instruction is not executed, the connection state between the steering wheel and the steering mechanism remains unchanged, and the vehicle is controlled to perform braking operations on the basis of decelerating the vehicle.

[0011] Furthermore, the side with harassment risk is the direction in which the cab is easily invaded by pole-shaped objects; the monitoring area is divided within the monitoring range of the vehicle-mounted camera, and the sub-area is divided within the monitoring area.

[0012] Furthermore, the traffic condition information includes the number of nearby vehicles, the distance between the vehicle on the right and the vehicle, and the width of the road at the vehicle's location; the number of nearby vehicles and the distance between the vehicle on the right and the vehicle are measured in real time by the on-board radar; the width of the road at the vehicle's location is obtained through the on-board navigation system network; the number of nearby vehicles is the sum of the number of vehicles in front, behind, left and right within the detection range of the on-board radar of the vehicle; the distance between the vehicle on the right and the vehicle is the distance between the right side of the vehicle's body and the vehicle on the right, and the width of the road at the vehicle's location is the sum of the widths of lanes traveling in the same direction of the road section where the vehicle is traveling.

[0013] Furthermore, the method of sequentially generating a deceleration instruction or an automatic driving instruction includes:

[0014] The sub-area numbers increase from the side close to the steering wheel. When the stick-shaped object enters the preset r-th sub-area from the side away from the steering wheel, r∈n, a deceleration command is generated. When the stick-shaped object passes through the r-th sub-area, an automatic driving command is generated.

[0015] Furthermore, the method for identifying the rod-shaped object sequentially entering the preset r-th sub-area from the side away from the steering wheel includes:

[0016] The trained rod-shaped object recognition model is loaded into the on-board computer, which performs real-time recognition of the video of the monitored area and outputs whether there are rod-shaped objects in the sub-areas. If the number corresponding to the rod-shaped object in the identified sub-area decreases and enters the r-th sub-area, it is determined that a rod-shaped object has entered the preset r-th sub-area from the side away from the steering wheel.

[0017] If the numbers corresponding to the rod-shaped objects in the identified sub-areas increase sequentially and enter the r-th sub-area, it is not determined that there is a rod-shaped object entering the preset r-th sub-area sequentially from the side away from the steering wheel.

[0018] Furthermore, the training method of the recognition model includes:

[0019] During the experimental phase, p groups of videos of rods invading the driving area from the sub-area are collected as training data, where p is an integer greater than 1. The training data with a preset ratio is used as the training set, and the training data with the difference between 1 and the preset ratio is used as the validation set. The training set is used to train the recognition model, and the validation set is used to validate the trained recognition model. When the training reaches the convergence condition, the trained recognition model is output. The recognition model is YOLOv8.

[0020] Furthermore, the road condition evaluation index is generated based on a weighted integration of the dimensionless number of nearby vehicles, the distance between the vehicle on the right and the vehicle, and the width of the road at the location of the vehicle.

[0021] Furthermore, the method for determining whether to execute the automatic driving instruction includes:

[0022] If the road condition evaluation index is greater than or equal to the preset road condition evaluation index threshold, the automatic driving instruction is not executed and the vehicle braking is controlled;

[0023] If the road condition assessment index is less than the preset road condition assessment index threshold, the automatic driving instruction is executed.

[0024] Furthermore, the control component that controls the disconnection or connection of the steering wheel and the steering mechanism includes a first electromagnetic ring, a first friction disc, a second friction disc, a second electromagnetic ring, a groove and a prism. The first electromagnetic ring and the first friction disc are fixedly mounted on the bottom of the second transmission shaft, and the second friction disc and the second electromagnetic ring are fixedly mounted on the top of the first transmission shaft. The first friction disc and the second friction disc are opposite and adjacent to each other. The groove is opened at the bottom of the first transmission shaft, and the top of the prism is movably arranged in the groove. The bottom of the prism is fixedly connected to the power input end of the vehicle's front wheel steering mechanism; the top of the second transmission shaft is fixedly connected to the steering wheel. When the control component is not disconnected, the prism is used to transmit the steering wheel rotational force to control the steering of the vehicle's front wheels.

[0025] A vehicle-mounted hazard warning system, comprising:

[0026] A setting module is used to set a monitoring area for the side of the vehicle with a harassment risk within the driving area of ​​the vehicle, divide the monitoring area into n sub-areas along the length of the vehicle body, where n is a positive integer greater than 1, and number each sub-area in sequence;

[0027] The acquisition module is used to continuously collect video in each sub-area to determine whether there are rod-shaped objects; based on the number of rod-shaped objects entering the sub-area, it determines whether to generate deceleration instructions or automatic driving instructions in sequence;

[0028] The collection module is used to collect the vehicle's traffic condition information in real time.

[0029] An analysis module is used to evaluate the vehicle driving environment based on traffic condition information and generate road condition evaluation indicators;

[0030] The control module is used for controlling the vehicle to decelerate when the deceleration instruction is generated, determining whether to execute the automatic driving instruction according to the road condition evaluation index at this time if the automatic driving instruction is generated, disconnecting the connection between the steering wheel and the steering mechanism and switching the vehicle to the automatic driving mode if the automatic driving instruction is executed, and keeping the connection state between the steering wheel and the steering mechanism unchanged if the automatic driving instruction is not executed, and controlling the vehicle to brake on the basis of deceleration.

[0031] The vehicle-mounted danger warning system and method have the following technical effects and advantages:

[0032] 1. Real-time monitoring of the side of the cab where the risk of harassment exists during the driving of the vehicle, identification of the rod-shaped object intruding into the cab through regional division of multiple sub-regions and identification models, early prediction and handling of possible snatch steering events, avoidance of safety accidents caused thereby, and acquisition of valuable buffer time for subsequent disposal; when the arms or other foreign objects are identified and warned to interfere with the normal driving of the driver, the different road conditions and different road sections are assessed in real time, the current state of the vehicle is deeply researched and judged whether it meets the strict requirements of automatic driving, and the coping decision is determined; meanwhile, the vehicle is controlled to decelerate, and then the current automatic driving conditions of the vehicle are determined through the evaluation of the road condition evaluation index, and the vehicle is seamlessly switched to the automatic driving mode as soon as the conditions are met, and at the same time, the physical connection between the steering wheel and the steering mechanism is decisively cut off to prevent the vehicle from deviating from the intended trajectory due to uneven force on the steering wheel in the subsequent possible fierce struggle, and to comprehensively ensure the safety of driving and the life and property of passengers.

[0033] 2. Furthermore, if it is determined that the automatic driving conditions are not met, the system will immediately focus on the vehicle braking link, and firmly prevent the automatic driving function from being forcibly started, thereby building another solid defense line for the safety of vehicle driving. It needs to be emphasized that whether the vehicle braking operation is triggered or the automatic driving mode is successfully switched, both are important prerequisites for the vehicle to have realized stable deceleration in advance. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 The figure is a schematic diagram of the system module of the present application;

[0035] Figure 2 The figure is a schematic diagram of the side of the present application where the risk of harassment exists;

[0036] Figure 3 The figure is a flow chart of the method of the present application;

[0037] Figure 4 The figure is a schematic diagram of the disconnection structure of the control assembly of the present application;

[0038] Figure 5 This is a schematic diagram of the structure of the control component of the present invention when it is not disconnected.

[0039] Explanation of the reference numerals: 20, control assembly; 201, first electromagnetic ring; 202, first friction disc; 203, second friction disc; 204, second electromagnetic ring; 205, prism; 206, prism; 2021, groove; 2022, bump; 30, first transmission shaft; 40, second transmission shaft. DETAILED DESCRIPTION

[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0041] Example 1

[0042] See also Figure 1 As shown, the vehicle-mounted hazard warning system described in this embodiment includes a setting module, an acquisition module, a collection module, an analysis module and a control module.

[0043] The setting module is used to set a monitoring area for the side with harassment risk in the driving area of ​​the vehicle, and divide the monitoring area into n sub-areas along the length direction of the vehicle body, where n is a positive integer greater than 1, and number each sub-area in sequence. The side with harassment risk is the direction in the cab that is easily invaded by pole-shaped objects, such as the area between the steering wheel and the co-pilot in a taxi. Figure 2 As shown, Figure 2 The steering wheel is on the left side, and the side with harassment risk is on the right side. In Hong Kong or other regions, the steering wheel is on the right side, and the side with harassment risk is on the left side. In buses, the area between the steering wheel and the front door is the side with harassment risk. Rod-shaped objects can easily enter the driving area from the side with harassment risk, affecting safe driving. Rod-shaped objects such as hands can enter the driving area from the side with harassment risk and grab the steering wheel.

[0044] The monitoring area is divided within the monitoring range of the vehicle-mounted camera, and the sub-areas are divided within the monitoring area; the purpose is to facilitate the subsequent analysis of the changes of the rod-shaped objects in multiple sub-areas and more accurately determine whether there is an intrusion risk.

[0045] The acquisition module is used to continuously collect video in each sub-area to determine whether there are rod-shaped objects. The number of rod-shaped objects entering the sub-area determines whether to generate deceleration instructions or automatic driving instructions.

[0046] The method of sequentially generating a deceleration instruction or an automatic driving instruction includes:

[0047] The sub-area numbers increase from the side closest to the steering wheel, such as Figure 2 As shown, when the stick-shaped object sequentially enters the preset r-th sub-area from the side away from the steering wheel, r∈n, a deceleration instruction is generated, and when the stick-shaped object passes through the r-th sub-area, an automatic driving instruction is generated.

[0048] The rth sub-area is set by the technicians themselves, and can be set specifically according to the probability of the vehicle's steering wheel being snatched. The greater the probability, the larger the value of the rth sub-area, and vice versa.

[0049] The method for identifying a rod-shaped object sequentially entering a preset r-th sub-area from a side away from the steering wheel includes:

[0050] The trained rod-shaped object recognition model is loaded into the on-board computer, and the video of the monitored area is recognized in real time to output whether there are rod-shaped objects in the sub-area. If the corresponding numbers of the rod-shaped objects in the identified sub-area decrease in sequence and enter the r-th sub-area, it is determined that a rod-shaped object has entered the preset r-th sub-area from the side away from the steering wheel. At this time, there is a rod-shaped intrusion, that is, there is a risk of grabbing the steering wheel.

[0051] If the numbers corresponding to the rod-shaped objects in the identified sub-areas increase in sequence and enter the r-th sub-area, it is not determined that a rod-shaped object has entered the preset r-th sub-area from the side away from the steering wheel. This corresponds to the situation where the driver's hand accidentally enters the r-th sub-area. At this time, there is no rod-shaped object intrusion, that is, there is no risk of grabbing the steering wheel.

[0052] The rod-shaped object recognition model can be a convolutional neural network model. During the experimental stage, p groups of videos of rod-shaped objects invading the driving area from the sub-area are collected as training data, where p is an integer greater than 1. 70% of the training data is used as a training set, and 30% of the training data is used as a validation set. The training set is used to train the recognition model, and the validation set is used to validate the trained recognition model. The trained recognition model is output when the training reaches convergence conditions. The recognition model can be YOLOv8, which can quickly and accurately identify objects in the video, or other suitable models such as CNN. Rod-shaped objects include arms, umbrellas, and crutches.

[0053] The collection module is used to collect the vehicle's traffic condition information in real time. The traffic condition information includes the number of nearby vehicles, the distance between the vehicle on the right and the vehicle, and the width of the road where the vehicle is located; the number of nearby vehicles and the distance between the vehicle on the right and the vehicle are measured in real time by the on-board radar; the width of the road where the vehicle is located can be obtained through the on-board navigation system. The number of nearby vehicles will affect the avoidance function and path planning of automatic driving, the distance between the vehicle on the right and the vehicle and the width of the road where the vehicle is located will affect the avoidance ability, vehicle reaction time and fault tolerance space of path planning. The above data will affect the decision-making ability of automatic driving and affect the safety of automatic driving. The number of nearby vehicles is the sum of the number of vehicles in front, behind, left and right within the detection range of the vehicle-mounted radar of the vehicle; the distance between the vehicle on the right and the vehicle is the distance between the right side of the vehicle and the vehicle on the right, and the width of the road where the vehicle is located is the sum of the widths of lanes traveling in the same direction of the road section where the vehicle is traveling.

[0054] The analysis module evaluates the vehicle's driving environment based on traffic information and generates a road condition evaluation index. Specifically, the road condition evaluation index is generated based on the standardized number of nearby vehicles, the distance between the vehicle on the right and the vehicle, and the width of the road where the vehicle is located. The example expression is as follows:

[0055]

[0056] Where Cls is the number of nearby vehicles (reflecting traffic density), Ycj is the distance between the vehicle on the right and the vehicle (reflecting the safety margin), Cdk is the road width at the vehicle's location (reflecting road conditions), and Lkx is a road condition assessment indicator. The numerator emphasizes the direct impact of the number of nearby vehicles on risk (the more vehicles, the higher the potential danger); the denominator combines the distance to the vehicle on the right (the closer the distance, the higher the risk) and the road width (the narrower the width, the higher the risk), which together suppress the risk value.

[0057] ε1, ε2, and ε3 are preset weights (all greater than 0, used to adjust the influence intensity of each parameter). The weights can be determined through regression analysis. A multivariate regression model is established using historical accident data or simulation data, and the weight ratio is determined by parameter significance. Principal component analysis can also be used to extract the principal components that have the greatest impact on the risk of executing autonomous driving and assign corresponding weights (this is existing technology and will not be elaborated here).

[0058] The above design reflects the dynamic balance between vehicle density and road space safety through a ratio, then compresses the numerical range through logarithmic transformation, and finally outputs a standardized risk indicator. Through weights and mathematical structures, it realizes the integration of multi-dimensional road condition data and risk quantification, providing a safe prerequisite for judging whether to execute the subsequent automatic driving instructions.

[0059] The control module is used to control the vehicle to decelerate when a deceleration instruction is generated. If an automatic driving instruction is generated, it is determined whether to execute the automatic driving instruction based on the road condition assessment index at that time. If the automatic driving instruction is executed, the connection between the steering wheel and the steering mechanism is disconnected, and the vehicle is switched to the automatic driving mode and drives according to the predetermined navigation path. If the automatic driving instruction is not executed, the connection state between the steering wheel and the steering mechanism remains unchanged, and the vehicle is controlled to perform braking operations on the basis of decelerating the vehicle.

[0060] The method for determining whether to execute the automatic driving instruction includes:

[0061] If the road condition evaluation index is greater than or equal to the preset road condition evaluation index threshold, the automatic driving instruction is not executed and the vehicle braking is controlled;

[0062] If the road condition evaluation index is less than the preset road condition evaluation index threshold, the automatic driving instruction is executed;

[0063] The preset road condition assessment index threshold is set by technical personnel. In an experimental environment, the size of the data contained in the traffic condition information is continuously changed. When the vehicle switches to automatic driving mode, if a collision, scratch or behavior that obviously does not comply with driving regulations occurs, the corresponding groups of traffic condition information containing data are taken, and the average of the road condition assessment indicators corresponding to the groups of traffic condition information is taken, and preset as the road condition assessment index threshold.

[0064] See also Figure 4 As shown, the control component 20 for controlling the disconnection or connection of the steering wheel and the steering mechanism includes a first electromagnetic ring 201, a first friction disc 202, a second friction disc 203, a second electromagnetic ring 204, a prism 205 and a prism 206. The first electromagnetic ring 201 and the first friction disc 202 are both fixedly sleeved on the bottom of the second transmission shaft 40, and the second friction disc 203 and the second electromagnetic ring 204 are both fixedly sleeved on the top of the first transmission shaft 30. The first friction disc 202 and the second friction disc 203 are arranged opposite and adjacent to each other. The prism 205 is opened at the bottom of the first transmission shaft 30, and the top of the prism 206 is movably arranged in the prism 205. The bottom of the prism 206 is fixedly connected to the power input end of the vehicle's front wheel steering mechanism (not shown in the figure, which is an existing structure in the prior art); the top of the second transmission shaft 40 is fixedly connected to the steering wheel (not shown in the figure). When the control component 20 is not disconnected, the prism 206 is used to transmit the steering wheel rotation force to control the steering of the vehicle's front wheels.

[0065] When in use, the first electromagnetic ring 201 and the second electromagnetic ring 204 are energized. When the bottom magnetic pole of the first electromagnetic ring 201 and the top magnetic pole of the second electromagnetic ring 204 are opposite, suction is generated, and the first friction disk 202 and the second friction disk 203 are in close contact. Figure 5As shown, the first friction disc 202 transmits the rotational force of the second transmission shaft 40 to the second friction disc 203. When the bottom magnetic pole of the first electromagnetic ring 201 is the same as the top magnetic pole of the second electromagnetic ring 204, a repulsive force is generated, and the first friction disc 202 and the second friction disc 203 are separated. Figure 4 As shown, the steering wheel and the steering mechanism are disconnected. During the movement of the second friction disc 203, under the action of the prism 205, the second friction disc 203 contacts and separates from the first friction disc 202 in a fixed path to avoid misalignment. At the same time, the prism 206 and the prism 205 both have the function of transmitting rotational force.

[0066] The bottom surface of the first friction disc 202 is provided with grooves 2021 evenly distributed in a circular shape around its center, and the top surface of the second friction disc 203 is fixed with protrusions 2022 corresponding to the positions of the grooves 2021. In this way, under the action of the grooves 2021 and the protrusions 2022, the connection strength between the first friction disc 202 and the second friction disc (203) is further improved.

[0067] This embodiment monitors the side of the cab where there is a risk of harassment while the car is moving in real time. By dividing the area into multiple sub-areas and using a recognition model to identify rod-shaped objects intruding into the cab, it aims to predict and handle possible steering wheel grabbing incidents in advance to avoid safety accidents caused by them. When it identifies and warns that an arm or other foreign object is about to interfere with the driver's normal driving, it conducts corresponding assessments of different road conditions and different road sections in real time and determines response decisions. At the same time, it first controls the vehicle to slow down. Based on the deceleration, it then evaluates the road condition assessment indicators to determine whether the vehicle currently meets the conditions for autonomous driving. If so, the vehicle switches to autonomous driving mode. At the same time, it decisively cuts off the physical connection between the steering wheel and the steering mechanism to prevent the vehicle from deviating from the established driving trajectory due to uneven force on the steering wheel in the subsequent fierce competition that may occur, thereby comprehensively ensuring driving safety and protecting the lives and property of passengers.

[0068] Furthermore, if it is determined that the conditions for autonomous driving are not met, the system will immediately focus on braking the vehicle, resolutely preventing the forced activation of the autonomous driving function, thus building another solid line of defense for vehicle driving safety. It is important to emphasize that both triggering the vehicle's braking operation and smoothly switching to autonomous driving mode require the vehicle to have achieved smooth deceleration in the early stage as a key prerequisite.

[0069] Example 2

[0070] See also Figure 3 As shown, for the parts not described in detail in this embodiment, please refer to the description of embodiment 1. A vehicle-mounted hazard warning method is provided, including:

[0071] In the driving area of the vehicle, a monitoring area is set for the side with the risk of harassment, the monitoring area is divided into n sub-areas along the length direction of the vehicle body, n is a positive integer greater than 1, and each sub-area is numbered in turn;

[0072] The video in each sub-area is continuously collected to determine whether there is a rod-shaped object. Whether to generate a deceleration instruction or an automatic driving instruction in turn is determined according to the number of rod-shaped objects entering the sub-area;

[0073] The traffic condition information of the vehicle is collected in real time;

[0074] The driving environment of the vehicle is evaluated according to the traffic condition information to generate a road condition evaluation index;

[0075] When the deceleration instruction is generated, the vehicle is controlled to decelerate. If the automatic driving instruction is generated, it is determined whether to execute the automatic driving instruction according to the road condition evaluation index at this time. If the automatic driving instruction is executed, the connection between the steering wheel and the steering mechanism is disconnected, and the vehicle is switched to an automatic driving mode. If the automatic driving instruction is not executed, the connection state of the steering wheel and the steering mechanism remains unchanged. On the basis of decelerating the vehicle, the vehicle is controlled to brake.

[0076] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0077] Finally: the above is only a preferred embodiment of the present application and is not used to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included within the protection scope of the present application.

Claims

1. A vehicle-mounted hazard warning method, characterized in that: include: Within the vehicle's driving area, a monitoring area is set for the side with a harassment risk. The monitoring area is divided into n sub-areas along the length of the vehicle body, where n is a positive integer greater than 1, and each sub-area is numbered sequentially; Continuously collect video in each sub-area to determine whether there are rod-shaped objects. Based on the number of rod-shaped objects entering the sub-area, decide whether to generate deceleration instructions or automatic driving instructions in sequence; The method of sequentially generating a deceleration instruction or an automatic driving instruction includes: The sub-area numbers increase from the side close to the steering wheel. When the stick-shaped object enters the preset r-th sub-area from the side away from the steering wheel, , generating a deceleration command, and when the rod passes through the rth sub-area, generating an automatic driving command; Collect the vehicle's traffic condition information in real time; Evaluate the vehicle driving environment based on traffic condition information and generate road condition evaluation indicators; When a deceleration instruction is generated, the vehicle is controlled to decelerate; if an automatic driving instruction is generated, whether to execute the automatic driving instruction is determined based on the road condition assessment index at that time. If the automatic driving instruction is executed, the connection between the steering wheel and the steering mechanism is disconnected, and the vehicle is switched to automatic driving mode; if the automatic driving instruction is not executed, the connection state between the steering wheel and the steering mechanism remains unchanged, and the vehicle is controlled to perform braking operations on the basis of decelerating the vehicle.

2. The vehicle-mounted danger warning method according to claim 1, characterized in that: The side with harassment risk is the direction in the cab that is easily invaded by pole-shaped objects; the monitoring area is divided within the monitoring range of the vehicle-mounted camera, and the sub-areas are divided within the monitoring area.

3. The vehicle-mounted danger warning method according to claim 2, characterized in that: The traffic condition information includes the number of nearby vehicles, the distance between the vehicle on the right and the vehicle, and the width of the road at the vehicle's location; the number of nearby vehicles and the distance between the vehicle on the right and the vehicle are measured in real time by the on-board radar; the width of the road at the vehicle's location is obtained through the on-board navigation system network; the number of nearby vehicles is the sum of the number of vehicles in front, behind, left and right within the detection range of the on-board radar of the vehicle; the distance between the vehicle on the right and the vehicle is the distance between the right side of the vehicle's body and the vehicle on the right, and the width of the road at the vehicle's location is the sum of the widths of lanes traveling in the same direction of the road section where the vehicle is traveling.

4. The vehicle-mounted danger warning method according to claim 3, characterized in that: The method for identifying a rod-shaped object sequentially entering a preset r-th sub-area from a side away from the steering wheel includes: The trained rod-shaped object recognition model is loaded into the on-board computer, which performs real-time recognition of the video of the monitored area and outputs whether there are rod-shaped objects in the sub-areas. If the number corresponding to the rod-shaped object in the identified sub-area decreases and enters the r-th sub-area, it is determined that a rod-shaped object has entered the preset r-th sub-area from the side away from the steering wheel. If the numbers corresponding to the rod-shaped objects in the identified sub-areas increase sequentially and enter the r-th sub-area, it is not determined that there is a rod-shaped object entering the preset r-th sub-area sequentially from the side away from the steering wheel.

5. The vehicle-mounted danger warning method according to claim 4, characterized in that: The training method of the recognition model includes: Collect p groups of videos of rod-shaped objects invading the driving area from the sub-area as training data, where p is an integer greater than 1. Use a preset ratio of training data as the training set, and use the training data with the difference between 1 and the preset ratio as the validation set. Use the training set to train the recognition model, and use the validation set to validate the trained recognition model. When the training reaches convergence conditions, output the trained recognition model. The recognition model is YOLOv8.

6. The vehicle-mounted danger warning method according to claim 5, characterized in that: The road condition evaluation index is generated based on a weighted integration of the dimensionless number of nearby vehicles, the distance between the vehicle on the right and the vehicle, and the width of the road at the location of the vehicle.

7. The vehicle-mounted danger warning method according to claim 6, characterized in that: The method for determining whether to execute the automatic driving instruction includes: If the road condition evaluation index is greater than or equal to the preset road condition evaluation index threshold, the automatic driving instruction is not executed and the vehicle braking is controlled; If the road condition assessment index is less than the preset road condition assessment index threshold, the automatic driving instruction is executed.

8. The vehicle-mounted danger warning method according to claim 7, characterized in that: The control assembly (20) for controlling the disconnection or connection of the steering wheel and the steering mechanism comprises a first electromagnetic ring (201), a first friction disc (202), a second friction disc (203), a second electromagnetic ring (204), a prism (205) and a prism (206). The first electromagnetic ring (201) and the first friction disc (202) are both fixedly sleeved on the bottom of the second transmission shaft (40), and the second friction disc (203) and the second electromagnetic ring (204) are both fixedly sleeved on the top of the first transmission shaft (30). The first friction disc (202) and the second friction disc (203) are arranged opposite to and adjacent to each other, a prism (205) is provided at the bottom of the first transmission shaft (30), the top of the prism (206) is movably arranged in the prism (205), and the bottom of the prism (206) is fixedly connected to the power input end of the steering mechanism of the front wheels of the vehicle; the top of the second transmission shaft (40) is fixedly connected to the steering wheel, and when the control component (20) is not disconnected, the prism (206) is used to transmit the steering wheel rotation force to control the steering of the front wheels of the vehicle.

9. A vehicle-mounted hazard warning system, characterized in that: include: A setting module is used to set a monitoring area for the side of the vehicle with a harassment risk within the driving area of ​​the vehicle, divide the monitoring area into n sub-areas along the length of the vehicle body, where n is a positive integer greater than 1, and number each sub-area in sequence; The acquisition module is used to continuously acquire video in each sub-area to determine whether there is a rod-shaped object; and determine whether to sequentially generate a deceleration instruction or an automatic driving instruction based on the number of rod-shaped objects entering the sub-area; the method of sequentially generating the deceleration instruction or the automatic driving instruction includes: the sub-area number increases sequentially from the side close to the steering wheel, and when the rod-shaped object sequentially enters the preset rth sub-area from the side away from the steering wheel, , generating a deceleration command, and when the rod passes through the rth sub-area, generating an automatic driving command; The collection module is used to collect the vehicle's traffic condition information in real time. An analysis module is used to evaluate the vehicle driving environment based on traffic condition information and generate road condition evaluation indicators; The control module is used to control the vehicle to decelerate when a deceleration instruction is generated. If an automatic driving instruction is generated, it is used to determine whether to execute the automatic driving instruction based on the road condition assessment index at that time. If the automatic driving instruction is executed, the connection between the steering wheel and the steering mechanism is disconnected, and the vehicle is switched to the automatic driving mode. If the automatic driving instruction is not executed, the connection state between the steering wheel and the steering mechanism remains unchanged, and the vehicle is controlled to perform braking operations on the basis of decelerating the vehicle.

Images