Vehicle safety warning system and method integrating trajectory prediction and side obstacle monitoring

Abstract

The present invention relates to a vehicle safety warning system and method integrating trajectory prediction and side obstacle monitoring. The system is installed on a vehicle and comprises a main controller and a laser radar recognition subsystem, a safety warning subsystem and an emergency braking subsystem respectively connected to the main controller. The laser radar recognition subsystem comprises a trajectory prediction module, a side obstacle monitoring module and a data processing module, which are used to obtain the predicted rear wheel driving trajectory and side obstacle coordinates and obtain the current danger level through comparison. The main controller controls the graded opening of the safety warning subsystem and the emergency braking subsystem according to the current danger level. Compared with the prior art, the present invention has the advantages of improving the steering safety of large vehicles.

Description

Technical Field

[0001] The present invention relates to the field of motor vehicle safety assistance technology, and in particular to a vehicle safety warning system and method integrating trajectory prediction and side obstacle monitoring. Background Art

[0002] When a vehicle turns, the trajectories of the front and rear wheels do not overlap, and the front inner wheel may pass through while the rear inner wheel cannot pass through normally, resulting in collision with other objects.

[0003] The difference between the front inner wheel and the rear inner wheel is called the inner wheel difference area. This area is in the driver's blind spot, and pedestrians or non-motor vehicle drivers often cannot realize that they are in it. Especially for large trucks, buses and other commercial vehicles, the inner wheel difference blind spot is relatively large. Even with the help of the rearview mirror on the vehicle, the driver of a large vehicle cannot fully take into account the surroundings of the vehicle during driving. Therefore, safety accidents related to the inner wheel difference of large vehicles are more likely to occur. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a vehicle safety warning system and method integrating trajectory prediction and side obstacle monitoring.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A vehicle safety warning system that integrates trajectory prediction and side obstacle monitoring. The system is installed on a vehicle and includes a main controller and a laser radar recognition subsystem, a safety warning subsystem and an emergency braking subsystem respectively connected to the main controller. The laser radar recognition subsystem includes a trajectory prediction module, a side obstacle monitoring module and a data processing module, which are used to obtain the predicted rear wheel driving trajectory and side obstacle coordinates and obtain the current danger level through comparison. The main controller controls the safety warning subsystem and the emergency braking subsystem to be opened in stages according to the current danger level.

[0007] Furthermore, the side obstacle monitoring module includes a laser radar sensor installed on the top of the vehicle, and the laser radar sensor is used to collect point cloud data of side obstacles. The data processing module obtains the geometric cluster center of the side obstacles by processing the point cloud data, and is used to compare it with the predicted vehicle rear wheel driving trajectory and determine the current danger level.

[0008] Furthermore, the trajectory prediction module includes a wheel speed sensor and an acceleration sensor, two of the wheel speed sensors and two of the acceleration sensors are respectively provided, and both the wheel speed sensors and the acceleration sensors are installed on the two front wheels of the vehicle, and are respectively used to collect the front wheel speed and front wheel acceleration of the vehicle;

[0009] The data processing module obtains the front wheel speed and front wheel acceleration, calculates and predicts the rear wheel trajectory, and uses the predicted rear wheel driving trajectory to compare with the geometric cluster center of the side obstacle and determine the current danger level.

[0010] Furthermore, the current risk level includes no collision risk, collision risk, and imminent collision;

[0011] When the geometric cluster center of the side obstacle is outside the rear wheel driving track of the vehicle, the current danger level is judged as no collision risk;

[0012] When the geometric cluster center of the side obstacle is inside the rear wheel driving track of the vehicle, the current danger level is judged as a collision risk;

[0013] When the distance between the geometric cluster center of the side obstacle and the vehicle body is less than the set threshold, the current danger level is judged as an imminent collision.

[0014] Furthermore, the safety warning subsystem includes a projection lamp group, a display and a speaker respectively connected to the main controller, the projection lamp group is installed on both sides of the top of the vehicle, the display and the speaker are arranged in the vehicle cab and connected to the main controller, and the emergency brake subsystem is a vehicle wire brake module;

[0015] When the vehicle is in a turning state, the main controller obtains the side obstacle coordinates collected in real time by the laser radar recognition subsystem and displays them visually through the display; at the same time, it receives the vehicle driving trajectory predicted by the laser radar recognition subsystem, calculates the inner wheel difference area graphic and sends it to the projection light group to project the warning area warning line to the ground. When the vehicle is driving in a straight line or turning to the other side, the projection light group is turned off.

[0016] Furthermore, when the current danger level is no collision risk, the projection light group and the display are turned on to play an auxiliary reminder role;

[0017] When the current danger level is that there is a risk of collision, the speaker is turned on to serve as a warning;

[0018] When the current danger level is that a collision is about to occur, the vehicle wire-controlled brake module performs forced braking on the vehicle.

[0019] A method for identifying danger levels of fusion trajectory prediction and side obstacle monitoring for the vehicle safety warning system, comprising:

[0020] Trajectory prediction step: Based on the collected front wheel speed information and acceleration information, the driving trajectory of the vehicle's rear wheels is calculated and predicted in real time;

[0021] Side obstacle detection step: According to the side obstacle point cloud data collected by the LiDAR sensor, the geometric cluster center coordinates of the side obstacles are calculated in real time;

[0022] And the danger level judgment step: compare the driving trajectory of the vehicle's rear wheels with the geometric cluster center coordinates of the side obstacles, and judge the current danger level as no collision risk, collision risk or imminent collision based on the positional relationship between the two.

[0023] Furthermore, the side obstacle detection step specifically includes:

[0024] 11) Obtain point cloud data of side obstacles;

[0025] 12) Convert the point cloud data in polar coordinates to obtain point cloud data in rectangular coordinates;

[0026] 13) The DBSCAN algorithm is used to calculate the cluster center coordinates of the side obstacles and filter the noise point cloud data.

[0027] Furthermore, the step 13) specifically includes:

[0028] 131) Selecting an unprocessed point from the point cloud data;

[0029] 132) Determine whether the selected point is a core point, if so, execute step 133), otherwise execute step 134);

[0030] 133) Find all objects that are density-reachable from the point to form a cluster;

[0031] 134) Identify the point as a noise point and filter it;

[0032] 135) Repeat steps 131)-134) until all points in the point cloud data center are processed.

[0033] Furthermore, in the step of determining the danger level, when the geometric cluster center of the side obstacle is outside the rear wheel driving track of the vehicle, the current danger level is determined to be no collision risk;

[0034] When the geometric cluster center of the side obstacle is inside the rear wheel driving track of the vehicle, the current danger level is judged as a collision risk;

[0035] When the distance between the geometric cluster center of the side obstacle and the vehicle body is less than the set threshold, the current danger level is judged as an imminent collision;

[0036] When judging whether the geometric cluster center of the side obstacle is inside or outside the vehicle's rear wheel driving track, the left front wheel of the vehicle is used as the center of the coordinate system and the following formula is used to judge that when the following formula is satisfied, the geometric cluster center of the side obstacle is judged to be inside the vehicle's rear wheel driving track:

[0037]

[0038]

[0039] Among them, (x c ,y c ) is the geometric cluster center coordinate of the lateral obstacle, R c is the geometric clustering radius of the side obstacles, (x rb ,y rb ) is the coordinate of the center of the right rear wheel steering circle of the vehicle, R rb is the radius of the right rear wheel turning circle of the vehicle, (x lf ,y lf ) is the coordinate of the center of the left front wheel steering circle of the vehicle, R lf is the radius of the turning circle of the left front wheel of the vehicle.

[0040] Compared with the prior art, the present invention has the following advantages:

[0041] The present invention determines the dangerous situation of the pedestrian's position by calculating the positional relationship between the cluster center formed by the side obstacles and the dangerous area formed when the vehicle turns, so as to remind the driver and reduce traffic accidents. It can integrate the trajectory prediction method and the side obstacle monitoring method into one, and effectively improve the driving safety of large commercial vehicles when turning right. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the working process of the system of the present invention;

[0043] Figure 2 This is a schematic diagram of the system hardware installation;

[0044] Figure 3 This is a schematic diagram of the installation of the laser radar recognition subsystem of the present invention;

[0045] Figure 4 Schematic diagram of the established rectangular coordinate system.

[0046] Among them, 1. Main controller, 2. LiDAR recognition system, 3. Projection light group, 4. Display, 5. Speaker, 6. Wheel speed sensor, 7. Acceleration sensor, 8. Wire-controlled brake module. DETAILED DESCRIPTION

[0047] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than 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 work should belong to the scope of protection of the present invention.

[0048] Example

[0049] like Figure 1 and Figure 2 As shown, the present invention discloses a vehicle safety warning system integrating trajectory prediction and side obstacle monitoring. The system is installed on a vehicle and includes a main controller 1 and a laser radar recognition subsystem, a safety warning subsystem and an emergency braking subsystem respectively connected to the main controller 1. The laser radar recognition subsystem includes a trajectory prediction module, a side obstacle monitoring module and a data processing module, which are used to obtain the predicted rear wheel driving trajectory and side obstacle coordinates and obtain the current danger level through comparison. The main controller 1 controls the safety warning subsystem and the emergency braking subsystem to be opened in stages according to the current danger level.

[0050] The side obstacle monitoring module includes a lidar sensor installed on the top of the vehicle. The lidar sensor is used to collect point cloud data of side obstacles. The data processing module obtains the geometric cluster center of the side obstacles by processing the point cloud data, and uses it to compare with the predicted vehicle rear wheel driving trajectory and determine the current danger level.

[0051] The trajectory prediction module includes a wheel speed sensor 6 and an acceleration sensor 7. Two wheel speed sensors 6 and two acceleration sensors 7 are respectively provided. Both the wheel speed sensor 6 and the acceleration sensor 7 are installed on the two front wheels of the vehicle, and are respectively used to collect the front wheel speed and the front wheel acceleration of the vehicle; the data processing module obtains the front wheel speed and the front wheel acceleration, calculates and predicts the rear wheel trajectory, and uses the predicted rear wheel driving trajectory to compare with the geometric cluster center of the side obstacle and judge the current danger level.

[0052] The safety warning subsystem includes a projection lamp group 3, a display 4 and a speaker 5 respectively connected to the main controller 1. The projection lamp group 3 is installed on both sides of the top of the vehicle. The display 4 and the speaker 5 are arranged in the vehicle cab and connected to the main controller 1. The emergency brake subsystem is a vehicle wire brake module 8;

[0053] When the vehicle is in a turning state, the main controller 1 obtains the side obstacle coordinates collected in real time by the laser radar recognition subsystem and displays them visually through the display 4; at the same time, it receives the vehicle driving trajectory predicted by the laser radar recognition subsystem, calculates the inner wheel difference area graphic and sends it to the projection light group 3 for projecting the warning area warning line to the ground. When the vehicle travels in a straight line or turns to the other side, the projection light group 3 is turned off.

[0054] Current danger levels include no collision risk, collision risk, and imminent collision;

[0055] When the geometric cluster center of the side obstacle is outside the driving track of the rear wheel of the vehicle, it is judged that the current danger level is no collision risk. At this time, the projection light group 3 and the display 4 are turned on to play an auxiliary reminder role;

[0056] When the geometric cluster center of the side obstacle is inside the driving track of the rear wheel of the vehicle, it is judged that the current danger level is a collision risk, and the speaker 5 is turned on to play a warning role;

[0057] When the distance between the geometric cluster center of the side obstacle and the vehicle body is less than the set threshold, it is determined that the current danger level is that a collision is about to occur. At this time, the vehicle wire-controlled brake module 8 performs forced braking on the vehicle.

[0058] The present invention also provides a method for identifying danger levels by integrating trajectory prediction and side obstacle monitoring, comprising:

[0059] Trajectory prediction step: Based on the collected front wheel speed information and acceleration information, the driving trajectory of the vehicle's rear wheels is calculated and predicted in real time;

[0060] Side obstacle detection steps: Based on the side obstacle point cloud data collected by the LiDAR sensor, the geometric cluster center coordinates of the side obstacles are calculated in real time, including:

[0061] 11) Obtain point cloud data of side obstacles;

[0062] 12) Convert the point cloud data in polar coordinates to obtain point cloud data in rectangular coordinates;

[0063] 13) Use the DBSCAN algorithm to calculate the cluster center coordinates of the side obstacles and filter the noise point cloud data;

[0064] 131) Selecting an unprocessed point from the point cloud data;

[0065] 132) Determine whether the selected point is a core point, if so, execute step 133), otherwise execute step 134);

[0066] 133) Find all objects that are density-reachable from the point to form a cluster;

[0067] 134) Identify the point as a noise point and filter it;

[0068] 135) Repeat steps 131)-134) until all points in the point cloud data center are processed.

[0069] Danger level judgment step: compare the driving track of the vehicle's rear wheels with the coordinates of the geometric cluster center of the side obstacle, and judge the current danger level as no collision risk, collision risk, or imminent collision based on the positional relationship between the two. In this step, when the geometric cluster center of the side obstacle is outside the driving track of the vehicle's rear wheels, the current danger level is judged to be no collision risk;

[0070] When the geometric cluster center of the side obstacle is inside the rear wheel driving track of the vehicle, the current danger level is judged as a collision risk;

[0071] When the distance between the geometric cluster center of the side obstacle and the vehicle body is less than the set threshold, the current danger level is judged as an imminent collision;

[0072] When judging whether the geometric cluster center of the side obstacle is inside or outside the vehicle's rear wheel driving track, the left front wheel of the vehicle is used as the center of the coordinate system and the following formula is used to judge that when the following formula is satisfied, the geometric cluster center of the side obstacle is judged to be inside the vehicle's rear wheel driving track:

[0073]

[0074]

[0075] Among them, (x c ,y c ) is the geometric cluster center coordinate of the lateral obstacle, R c is the geometric clustering radius of the side obstacles, (x rb ,y rb ) is the coordinate of the center of the right rear wheel steering circle of the vehicle, R rb is the radius of the right rear wheel turning circle of the vehicle, (x lf ,y lf ) is the coordinate of the center of the left front wheel steering circle of the vehicle, R lf is the radius of the turning circle of the left front wheel of the vehicle.

[0076] like Figure 1 and Figure 2 As shown, the enhanced commercial vehicle two-way safety assistance system provided by the present invention takes active warning measures for pedestrians and drivers by integrating the risk assessment method of trajectory prediction and side obstacle monitoring, and implements vehicle emergency braking when pedestrians are in a high-risk area. Specifically, the main controller calculates the geometric relationship between the vehicle trajectory and the side obstacles, and implements automatic emergency braking of the vehicle when pedestrians are in a high-risk area.

[0077] By connecting to the laser radar recognition subsystem 2 installed on the vehicle, the vehicle's driving trajectory is calculated and predicted in real time, and the vehicle's trajectory is compared with the side obstacle coordinates calculated by the laser radar recognition subsystem 2, so as to calculate the danger level, and the side obstacle coordinates and the danger level are fed back to the main controller 1 installed on the vehicle. The main controller feeds back the coordinates to the display module of the driver warning subsystem, and controls the actions of the safety warning subsystem and the emergency braking subsystem respectively according to the level of danger. The safety warning subsystem includes a driver warning subsystem and a pedestrian warning subsystem.

[0078] The laser radar recognition subsystem 2 uses the collected point cloud data to select and process the point cloud data, calculate the geometric center of the side obstacle cluster in real time, and calculate the current driving trajectory of the vehicle in real time and predict the vehicle trajectory by calculating the data collected by the wheel speed sensor module and the acceleration sensor module in the data acquisition subsystem. By judging the distance between the vehicle's driving trajectory and the cluster geometric center, the current danger level is judged, and the danger level and the real-time coordinates of the side obstacles are transmitted to the main controller 1 in real time. The main controller 1 feeds back the relevant danger level information to the pedestrian warning subsystem and the driver warning subsystem. The pedestrian warning subsystem includes a projection light group module, and the driver warning subsystem includes a display module and a speaker module. When the vehicle turns right, the danger level of the side of the vehicle will be given by the main controller 1, and the pedestrian warning subsystem, the driver warning subsystem and the emergency braking subsystem will make corresponding warning operations according to the danger level. When it is at a high risk danger level, the emergency braking subsystem is turned on to achieve active braking of the vehicle.

[0079] The pedestrian warning subsystem includes a projection light group 3, which is installed on both sides of the roof of the vehicle and is connected to the main controller 1. It is used to receive the inner wheel difference area warning graphic transmitted by the main controller and project the warning area warning line onto the ground; when the vehicle is driving in a straight line or turning to the other side, the projection light group 3 is turned off.

[0080] The driver warning module includes a speaker 5 and a display 4 connected to the main controller 1, wherein the display 4 will display the vehicle trajectory and cluster coordinate information in real time when the vehicle turns right to assist the driver in driving. When the danger level is that there is a risk of collision or a collision is about to occur, the display 4 will additionally serve as a visual warning to the driver. When the main controller determines that there is a risk of collision, the speaker 5 will issue a sound warning to remind the driver to avoid it.

[0081] There are two wheel speed sensors 6 and two acceleration sensors 7, which are respectively installed on the two front wheels of the vehicle and connected to the laser radar recognition subsystem 2. The laser radar recognition subsystem 2 calculates and predicts the rear wheel trajectory through the front wheel speed and front wheel acceleration information, and uses it to compare with the geometric center of the side obstacle cluster and determine the danger level.

[0082] When calculating the geometric center of the side obstacle cluster in real time, the point cloud data is first captured in real time. After the point cloud data scanned by the lidar sensor is captured, the point cloud data packet is filtered to filter out the coordinate points with poor quality. After that, the point cloud data can be further processed. After obtaining high-quality point cloud data, the point cloud data is converted from polar coordinates to Cartesian coordinates. The specific formula is:

[0083]

[0084]

[0085] Where PI represents the constant π, which can be estimated to be 3.1415926. x represents the x-coordinate of the point on the Cartesian coordinate axis. y represents the y-coordinate of the point on the Cartesian coordinate axis. tmpAngle represents the polar angle of the point in the polar coordinate system. tmpDistance represents the distance of the point in the polar coordinate system.

[0086] In the process of converting the original polar angle data to Cartesian coordinate data, additional processing is also done in consideration of the actual application environment. Since this LiDAR recognition system is only used to detect pedestrians or other potential risk objects that large commercial vehicles may encounter during right turns, the LiDAR does not need to care about the left side of the polar angle coordinate system during processing, that is, when the polar angle θ satisfies:

[0087] 90°≤θ∧θ≤270°

[0088] There is no need to convert the point into Cartesian coordinates, but you can choose to directly discard the points that meet the conditions and not consider them.

[0089] This embodiment uses the DBSCAN algorithm to calculate the cluster center coordinates and filter out the noise point cloud data. The DBSCAN algorithm used in the laser radar recognition system can be summarized as follows:

[0090] (1) Extract an unprocessed point from the database;

[0091] (2) If the extracted point is a core point, find all objects that are density-reachable from the point to form a cluster;

[0092] (3) Otherwise, the extracted point is an edge point (non-core object), and the point is marked as a noise point, and the loop is exited to find the next point;

[0093] (4) Repeat steps (1) to (3) until all points are processed.

[0094] Among them, the core point is the point where the number of points within a given radius exceeds the threshold, and the given radius and the number of thresholds can be set according to the usage scenarios and accuracy requirements of different vehicle models.

[0095] The formula for calculating the rear wheel trajectory is:

[0096]

[0097]

[0098] Among them, v fl is the speed of the front left wheel, v fr is the speed of the right front wheel, L is the vehicle wheelbase, B is the vehicle track, and R is the turning radius of the inner rear wheel during steering.

[0099] After calculating the vehicle's running trajectory, compare the front and rear wheel running trajectories with the cluster center coordinates to calculate the danger level status. Install the laser radar recognition subsystem at the designated location of the designated vehicle, such as Figure 3 As shown, the transformed rectangular coordinate system is as follows Figure 4 As shown, the left front wheel of the car is taken as the center of the coordinate system. According to the geometric relationship, it is easy to conclude that the side obstacle target is in a dangerous state only needs to meet the following conditions:

[0100]

[0101]

[0102] Among them, the cluster target center (x c ,y c ), radius R c , the center of the right rear wheel steering circle of the car (x rb ,y rb ), radius R rb , the center of the left front wheel steering circle of the car (x lf ,y lf ), radius R lf .

[0103] The danger level can be divided into three levels: no collision risk, collision risk and imminent collision. When the clustered target is outside the vehicle's running track, the laser radar recognition system determines that there is no collision risk; when the clustered target is inside the vehicle's running track, the laser radar recognition system determines that there is a collision risk; when the clustered target is too close to the vehicle body, the laser radar recognition system determines that a collision is imminent.

[0104] The laser radar recognition subsystem 2 transmits the danger level information and real-time coordinate information to the main controller 1, and the main controller 1 coordinates with other subsystems. The laser radar recognition subsystem 2 is connected to the wheel speed sensor 6, the acceleration sensor 7 and the main controller 1, and uses the wheel speed sensor 6 and the acceleration sensor 7 to calculate and predict the vehicle's driving trajectory, and transmits the danger level information and real-time coordinate information to the main controller 1. When the vehicle turns, no matter what danger level it is in, the main controller visualizes the real-time coordinates transmitted by the laser radar recognition system and outputs them to the display 4 to achieve driving assistance or driving warning for the driver. At the same time, the main controller 1 uses the projection light group 3 to remind pedestrians of vehicle turns. When the danger level is that there is a risk of collision, the main controller 1 controls the speaker 5 to give a voice warning to the driver. When the danger level is that a collision is about to occur, the main controller 1 controls the wire-controlled brake module 8 to perform emergency braking on the vehicle.

[0105] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person familiar with the technical field can easily think of various equivalent modifications or substitutions within the technical scope disclosed by the present invention, and these modifications or substitutions should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for identifying the danger level of a vehicle safety warning system that integrates trajectory prediction and side obstacle monitoring, the system being installed on a vehicle, the system comprising a main controller (1) and a laser radar recognition subsystem, a safety warning subsystem and an emergency braking subsystem respectively connected to the main controller (1), the laser radar recognition subsystem comprising a trajectory prediction module, a side obstacle monitoring module and a data processing module, for obtaining the predicted rear wheel driving trajectory and the side obstacle coordinates and obtaining the current danger level by comparison, the main controller (1) controlling the safety warning subsystem and the emergency braking subsystem to be opened in stages according to the current danger level; The hazard level identification method comprises: Trajectory prediction step: Based on the collected front wheel speed information and acceleration information, the driving trajectory of the vehicle's rear wheels is calculated and predicted in real time; Side obstacle detection step: According to the side obstacle point cloud data collected by the LiDAR sensor, the geometric cluster center coordinates of the side obstacles are calculated in real time; and the step of judging the danger level: comparing the driving trajectory of the rear wheels of the vehicle with the geometric cluster center coordinates of the side obstacles, and judging the current danger level as no collision risk, collision risk or imminent collision according to the positional relationship between the two; The side obstacle detection step specifically includes: 11) Obtain point cloud data of side obstacles; 12) Convert the point cloud data in polar coordinates to obtain point cloud data in rectangular coordinates; 13) Use the DBSCAN algorithm to calculate the cluster center coordinates of the side obstacles and filter the noise point cloud data; The step 13) specifically includes: 131) Selecting an unprocessed point from the point cloud data; 132) Determine whether the selected point is a core point, if so, execute step 133), otherwise execute step 134); 133) Find all objects that are density-reachable from the point to form a cluster; 134) Identify the point as a noise point and filter it; 135) Repeat steps 131)-134) until all points in the point cloud data center are processed; In the step of determining the danger level, when the geometric cluster center of the side obstacle is outside the rear wheel driving track of the vehicle, the current danger level is determined to be no collision risk; When the geometric cluster center of the side obstacle is inside the rear wheel driving track of the vehicle, the current danger level is judged as a collision risk; When the distance between the geometric cluster center of the side obstacle and the vehicle body is less than the set threshold, the current danger level is judged as an imminent collision; When judging whether the geometric cluster center of the side obstacle is inside or outside the vehicle's rear wheel driving track, the left front wheel of the vehicle is used as the center of the coordinate system and the following formula is used to judge that when the following formula is satisfied, the geometric cluster center of the side obstacle is judged to be inside the vehicle's rear wheel driving track: Among them, (x c ,y c ) is the geometric cluster center coordinate of the lateral obstacle, R c is the geometric clustering radius of the side obstacles, (x rb ,y rb ) is the coordinate of the center of the right rear wheel steering circle of the vehicle, R rb is the radius of the right rear wheel turning circle of the vehicle, (x lf ,y lf ) is the coordinate of the center of the left front wheel steering circle of the vehicle, R lf is the radius of the turning circle of the left front wheel of the vehicle.

2. The method according to claim 1, characterized in that The side obstacle monitoring module includes a laser radar sensor installed on the top of the vehicle. The laser radar sensor is used to collect point cloud data of side obstacles. The data processing module obtains the geometric cluster center of the side obstacles by processing the point cloud data, and uses it to compare with the predicted vehicle rear wheel driving trajectory and determine the current danger level.

3. The method according to claim 2, characterized in that The trajectory prediction module comprises a wheel speed sensor (6) and an acceleration sensor (7), two of the wheel speed sensors (6) and two of the acceleration sensors (7) are provided, and both of the wheel speed sensors (6) and the acceleration sensors (7) are installed on two front wheels of the vehicle, and are used to collect the front wheel speed and front wheel acceleration of the vehicle respectively; The data processing module obtains the front wheel speed and front wheel acceleration, calculates and predicts the rear wheel trajectory, and uses the predicted rear wheel driving trajectory to compare with the geometric cluster center of the side obstacle and determine the current danger level.

4. The method according to claim 2, characterized in that: The current danger level includes no collision risk, collision risk, and imminent collision; When the geometric cluster center of the side obstacle is outside the rear wheel driving track of the vehicle, the current danger level is judged as no collision risk; When the geometric cluster center of the side obstacle is inside the rear wheel driving track of the vehicle, the current danger level is judged as a collision risk; When the distance between the geometric cluster center of the side obstacle and the vehicle body is less than the set threshold, the current danger level is judged as an imminent collision.

5. The method according to claim 4, characterized in that The safety warning subsystem comprises a projection light group (3), a display (4) and a speaker (5) respectively connected to the main controller (1); the projection light group (3) is installed on both sides of the top of the vehicle; the display (4) and the speaker (5) are arranged in the vehicle cab and connected to the main controller (1); the emergency brake subsystem is a vehicle wire-controlled brake module (8); When the vehicle is in a turning state, the main controller (1) obtains the coordinates of the side obstacles collected in real time by the laser radar recognition subsystem and displays them visually through the display (4); at the same time, it receives the vehicle's driving trajectory predicted by the laser radar recognition subsystem, calculates the inner wheel difference area graphic and sends it to the projection light group (3) for projecting the warning area warning line on the ground. When the vehicle is driving in a straight line or turning to the other side, the projection light group (3) is turned off.

6. The method according to claim 5, characterized in that When the current danger level is no collision risk, the projection light group (3) and the display (4) are turned on to play an auxiliary reminder role; When the current danger level indicates that there is a risk of collision, the loudspeaker (5) is turned on to serve as a warning; When the current danger level indicates that a collision is about to occur, the vehicle wire-controlled brake module (8) performs forced braking on the vehicle.

Images