Omnidirectional collision risk prediction method for large bus

Abstract

The invention discloses a large bus omni-directional collision risk prediction method, which comprises the following steps: S1, based on vehicle configuration, speed and course angle elements, establishing a large bus body coordinate system and a global coordinate system, and constructing a directional bounding box of a large bus and other traffic participants around the large bus; s2, considering forms and motion states of surrounding vehicles, constructing a collision buffer area of the large bus and detecting a potential collision boundary of the large bus; and S3, based on the principles of geometry and vehicle kinematics, predicting the nearest distance between the collision boundaries of the large bus and the surrounding vehicles within a certain time range and the occurrence time of the nearest distance. According to the method, real-time prediction of surrounding potential collision risks of the large bus in a complex traffic flow environment is facilitated, and technical support is provided for guaranteeing safety of drivers and passengers and other traffic participants.

Description

Technical Field

[0001] The invention belongs to the technical field of vehicle collision risk prediction, and in particular relates to a method for predicting omnidirectional collision risk of a large passenger bus. Background Art

[0002] Road passenger and freight transport has the characteristics of "long driving time, complex operating environment, and serious consequences of accidents", which seriously affects the overall situation of road traffic safety in my country.

[0003] At present, my country has a large number of large passenger buses and a variety of usage scenarios. Once a collision accident occurs, it is easy to cause mass casualties and cause adverse effects. The existing on-board collision risk real-time prediction system is mainly used in the field of small and medium-sized vehicles, and the promotion and application depth in the field of road passenger vehicles are still very limited. At the same time, the existing collision prediction system mainly studies the risk of forward collision, but lacks attention to the prediction of lateral and rear collisions. The effective prediction time of the system is short, and the response to complex situations is slow, and it cannot provide safe and reliable omnidirectional collision risk prediction. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention provides a method for predicting the omnidirectional collision risk of large buses, which provides fast, accurate and real-time prediction of potential collision risks in all directions for large buses, accurately predicts the collision location and the affected part of the vehicle, and provides strong technical support for guiding drivers to avoid collisions and reduce the consequences of serious collisions.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for predicting the omnidirectional collision risk of a large passenger car, comprising the following steps:

[0006] S1. Based on the vehicle configuration, speed, and heading angle elements, the large bus body coordinate system and the global coordinate system are established, and then the directional bounding box of the large bus and surrounding vehicles is constructed;

[0007] S2, combining the shapes and motion states of the vehicles around the large bus, constructing a collision buffer area for the large bus and detecting the collision boundaries between the surrounding vehicles and the large bus;

[0008] S3. Based on the principles of geometry and vehicle kinematics, predict the risk indicators of large buses and surrounding vehicles within a preset time range.

[0009] Furthermore, the aforementioned establishment of the large bus body coordinate system and the global coordinate system represents the vertex coordinates of the oriented bounding box, as follows:

[0010] Rear wheel inner vertex e nbi , whose coordinates (x nbi ,y nbi )for:

[0011]

[0012] Rear wheel outer vertex e nbo , whose coordinates (x nbo ,y nbo )for:

[0013]

[0014] Front wheel inner vertex e nfi , whose coordinates (x nfi ,y nfi )for:

[0015]

[0016] Front wheel outer vertex e nfo , whose coordinates (x nfo ,y nfo )for:

[0017]

[0018] In the formula, θ n is the vehicle heading angle, (x nbc ,y nbc ) is the center point of the rear axle of the vehicle, L n ,W n is the length and width of the vehicle's oriented bounding box.

[0019] Furthermore, in the aforementioned step S1, the directional bounding box includes the length, width, real-time speed and heading angle elements of the bus, and the minimum circumscribed rectangle E of the vehicle is obtained with the front and rear wheels and the inner and outer sides of the bus as the boundaries. n , as follows:

[0020] E n ={e nbi ,e nbo ,e nfi ,e nfo};

[0021] Where n is the vehicle number, b and f are the front and rear wheels of the vehicle, i and o are the inside and outside of the vehicle, respectively. nbi is the inner vertex of the rear wheel, e nbo It is the outer vertex of the rear wheel of the bus, e nfi It is the vertex inside the front wheel of the bus, e nfo It is the outer vertex of the front wheel of the bus.

[0022] Furthermore, in the aforementioned step S1, the oriented bounding box vertex position vector coordinate r e It is expressed in vehicle body coordinate system and global coordinate system as follows:

[0023] re =r nbc +R n ·L size ;

[0024] In the formula, r nbc is the position vector of the center point of the rear axle of the vehicle, L size is the size matrix of the vehicle's oriented bounding box, R n is the vehicle rotation matrix.

[0025] Furthermore, in the aforementioned step S1, the oriented bounding box boundary is l nk , the relationship between the boundary and the vertex is as follows:

[0026]

[0027] l ni is the inner collision boundary of the bus, l nf is the collision boundary of the rear wheel of the bus, l no is the outer collision boundary of the bus, l nb It is the collision boundary of the front wheels of the bus.

[0028] Furthermore, the aforementioned step S2 includes the following sub-steps:

[0029] S2.1. Obtain the minimum convex hull Vex that encloses the oriented bounding boxes of the two vehicles mn , Vex mn Satisfy the following formula:

[0030] E m ,E n ∈Vex mn ;

[0031] In the formula, E m ,E n are the vertices of the oriented bounding boxes of the surrounding vehicles m and bus n respectively;

[0032] S2.2. Perform an XOR operation on the minimum convex hull and the vehicle oriented bounding box to obtain the collision buffer area between the two vehicles, as shown in the following formula:

[0033] buffer = xor(xor(Vex mn ,E m ),E n ),

[0034] The potential collision boundary between the surrounding vehicles and the bus satisfies the following formula:

[0035]

[0036] In the formula, l mi is the inner collision boundary of the bus, lmf is the collision boundary of the rear wheel of the bus, l mo is the outer collision boundary of the bus, l mb It is the collision boundary of the front wheel of the bus;

[0037] S2.3 traverses and solves the change of the shortest distance between the collision boundary of the bus and the surrounding vehicles within a preset time range to determine the collision risk between the bus and the surrounding vehicles.

[0038] Furthermore, in the aforementioned step S3, the risk indicators include: the shortest distance between the collision boundary of the bus and surrounding vehicles, the time of occurrence, and the shortest reaction time, which specifically includes the following sub-steps:

[0039] S3.1. Upper limit of predicted time for buses t upper Can be dynamically calibrated as:

[0040]

[0041] In the formula, the driver’s reaction time is Δt react , are the limit deceleration and safety margin of the bus respectively, vn is the speed vector of the bus;

[0042] S3.2, bus n and surrounding vehicle m begin to intersect at time t0. The surrounding vehicle m is considered as a relatively stationary vehicle and bus n is considered as a relatively moving vehicle. Then the collision boundary l no Keep moving towards l mi Get closer and calculate the time range of the collision prediction at time t0, as follows:

[0043] t∈[t0,t0+t upper ],

[0044] S3.3. The shortest distance d between collision boundaries within the time range of collision prediction min The change in the collision prediction time range d min The minimum value d cpa The time t at which the minimum occurs cpa ;

[0045] S3.4. Calculate the relative velocity vector v between the surrounding vehicles and the bus o , as follows:

[0046] v o =v n -v m ,

[0047] In the formula, v m ,v n are the velocity vectors of the surrounding vehicles and the bus respectively;

[0048] S3.5. Obtain the collision boundary l of the surrounding vehicle m mi The parametric equation is as follows:

[0049]

[0050] In the formula, x mi ,y mi Represents the collision boundary l mi The vertex coordinates of any point, β represents the control parameter of the parametric equation, and the value range of β is [0,1]; x mbi is the horizontal coordinate of the inner vertex of the rear wheel of the surrounding vehicle, y mbi is the ordinate of the vertex on the inner side of the rear wheel of the surrounding vehicle, x mfi is the horizontal coordinate of the vertex on the inner side of the front wheel of the surrounding vehicle, y mfi is the vertical coordinate of the inner vertex of the front wheel of the surrounding vehicles; S3.6, after Δt time, obtain the collision boundary l of bus n no The parametric equation is:

[0051]

[0052] In the formula, x no ,y no Represents the collision boundary l no The coordinates of any point on x ,μ y represents the unit vector in the X-axis and Y-axis directions, α represents the control parameter of the parametric equation, and the value range of α is [0,1]; x nbo is the horizontal coordinate of the vertex on the outside of the rear wheel of the bus, y nbo is the ordinate of the vertex on the outside of the rear wheel of the bus, x nfo is the horizontal coordinate of the vertex on the outside of the front wheel of the bus, y nfo It is the ordinate of the vertex on the outside of the front wheel of the bus.

[0053] S3.7. Calculate the collision boundary l at time t0+Δt mi ,l no The distance d between any two points p and q on pq (t0+Δt), as follows:

[0054]

[0055] S3.8. Order beg When the minimum value is obtained, it is the shortest distance between collision boundaries.

[0056] S3.9. Calculate the minimum value of the closest distance between collision boundaries within the time range of collision prediction, as well as the corresponding occurrence time and the shortest reaction time.

[0057] Furthermore, the aforementioned step S3.8 is specifically: The partial derivative about (α, β) is 0, and we get a set of equations. Solving the equations we get The equation is:

[0058]

[0059] If satisfied This indicates that points p and q are on the collision boundary; if Then let α=1orβ=1, and convert the shortest distance between line segments into the shortest distance from the vertex to the line segment, then we have:

[0060]

[0061] The shortest distance d between collision boundaries min (t0+Δt) is expressed as:

[0062]

[0063] Furthermore, the aforementioned step S3.9 is specifically as follows: in [t0, t0+t upper ] within the range of d min The minimum value of (t0+Δt), the corresponding time and the shortest reaction time are recorded as d cpa ,t cpa ,Δt min , then:

[0064]

[0065] t cpa =t0+Δt min .

[0066] When there are several surrounding vehicles, determine the distance between the bus and the surrounding vehicles. min The minimum value of can be used to obtain the vehicle that poses the greatest collision threat to the bus at the current moment, and the shortest reaction time of multiple vehicles, as shown in the following formula:

[0067] Δt gt =min(Δt min (m,n)),m=1,2,3,…;

[0068] In the formula, Δt gt Indicates the shortest reaction time of multiple workshops, Δt min (m,n) represents the shortest reaction time between surrounding vehicle m and bus n.

[0069] Furthermore, the aforementioned depiction of the posture changes of the vehicle during movement by using a directional bounding box can detect and screen out potential collision boundaries during the vehicle intersection process.

[0070] Furthermore, the aforementioned conversion of the vehicle collision prediction problem into the prediction of the shortest distance between potential collision boundaries can accurately calculate the shortest distance between two vehicles within a certain time range, the time when the shortest distance occurs, and the potential damaged part.

[0071] Compared with the prior art, the beneficial technical effects of the present invention using the above technical solution are as follows:

[0072] 1. Comprehensively consider the interaction between physical information such as vehicle shape, speed, heading angle and other traffic participants, characterize the vehicle motion state through the directional bounding box model, establish a collision buffer area to detect the potential collision boundary between vehicles, and implement dynamic calibration of the collision prediction time range based on the vehicle state and the driver's reaction ability. Without the need for complex means such as large models, only the principles of geometry and vehicle kinematics are needed to predict risk indicators such as the shortest distance between vehicle collision boundaries and the shortest reaction time.

[0073] 2. The present invention helps the driver to quickly predict potential collision risks in all directions, which is of great significance to ensuring the driving safety of the driver and other traffic participants. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is the overall implementation flow chart of the present invention.

[0075] Figure 2 It is a vehicle oriented bounding box model diagram of the present invention.

[0076] Figure 3 It is a schematic diagram of vehicle collision boundary detection of the present invention.

[0077] Figure 4 It is a schematic diagram of the vehicle collision prediction principle of the present invention.

[0078] Figure 5 It is a schematic diagram of omnidirectional collision prediction of the present invention.

[0079] Figure 6 This is a simulation experiment diagram of the omnidirectional collision prediction of the present invention. DETAILED DESCRIPTION

[0080] In order to better understand the technical content of the present invention, specific embodiments are given and described as follows in conjunction with the accompanying drawings.

[0081] Various aspects of the invention are described herein with reference to the accompanying drawings, in which many illustrative embodiments are shown. The embodiments of the invention are not limited to those described in the accompanying drawings. It should be understood that the invention is implemented by any of the various concepts and embodiments described above, as well as the concepts and embodiments described in detail below, because the concepts and embodiments disclosed in the invention are not limited to any implementation. In addition, some aspects disclosed in the invention may be used alone or in any appropriate combination with other aspects disclosed in the invention.

[0082] like Figure 1 As shown, the present invention provides a method for predicting the omnidirectional collision risk of a large passenger car, comprising the following steps:

[0083] S1. Based on the vehicle configuration, speed, and heading angle elements, the large bus body coordinate system and the global coordinate system are established, and then the directional bounding box of the large bus and surrounding vehicles is constructed;

[0084] S2, combining the shapes and motion states of the vehicles around the large bus, constructing a collision buffer area for the large bus and detecting the collision boundaries between the surrounding vehicles and the large bus;

[0085] S3. Based on the principles of geometry and vehicle kinematics, predict the risk indicators of large buses and surrounding vehicles within a preset time range.

[0086] As a preferred embodiment of the present invention, refer to Figure 2 , considering the vehicle length, width, real-time speed and heading angle elements, construct a directional bounding box surrounding the vehicle. Taking the front and rear wheels and the inner and outer sides of the vehicle as the boundaries, obtain the vehicle's minimum circumscribed rectangle E n :

[0087] E n ={e nbi ,e nbo ,e nfi ,e nfo};

[0088] Where n is the vehicle number, b and f are the front and rear sides of the vehicle, i and o are the inside and outside of the vehicle, respectively; e nbi is the inner vertex of the rear wheel, e nbo It is the outer vertex of the rear wheel of the bus, e nfi It is the vertex inside the front wheel of the bus, e nfo It is the outer vertex of the front wheel of the bus.

[0089] The vehicle body coordinate system and the global coordinate system are used to represent the vertex coordinates of the oriented bounding box, as follows:

[0090] Rear wheel inner vertex e nbi , whose coordinates (x nbi ,y nbi )for:

[0091]

[0092] Rear wheel outer vertex e nbo , whose coordinates (x nbo ,y nbo )for:

[0093]

[0094] Front wheel inner vertex e nfi , whose coordinates (x nfi ,y nfi )for:

[0095]

[0096] Front wheel outer vertex e nfo , whose coordinates (x nfo ,y nfo )for:

[0097]

[0098] In the formula, θ n is the vehicle heading angle, (x nbc ,y nbc ) is the center point of the rear axle of the vehicle, L n ,W n is the length and width of the vehicle's oriented bounding box. The same method can be used to obtain the vertex information of the bounding box boundary of the surrounding vehicles of the bus;

[0099] The boundary of the oriented bounding box is represented by l nk Represented by, then the relationship between the boundary and the vertex is

[0100]

[0101] In the formula, l ni is the inner collision boundary of the bus, l nf is the collision boundary of the rear wheel of the bus, l no is the outer collision boundary of the bus, l nb It is the collision boundary of the front wheels of the bus.

[0102] As a preferred embodiment of the present invention, when exploring the collision boundary between the surrounding vehicle m and the bus n, first find the minimum convex hull Vex that can enclose the circumscribed rectangle of the two vehicles. mn , then Vex mn Should meet:

[0103] E m ,E n ∈Vex mn ;

[0104] Perform an XOR operation on the minimum convex hull and the vehicle bounding box to obtain the collision buffer between the two vehicles:

[0105] buffer = xor(xor(Vex mn ,E m ),E n );

[0106] The larger the buffer area, the more space there is for both drivers to make dynamic adjustments before the vehicles collide.

[0107] The potential collision boundary between the bus and surrounding vehicles should meet the following requirements:

[0108]

[0109] Therefore, in Figure 3 In the example, the collision boundary between the surrounding vehicle m and the bus n is l mi ,l mf ,l no ,l nb , it can be predicted that in this case, there is a certain risk of collision between the side and rear of the bus and the front and side of the car. Then, the change of the shortest distance between the collision boundaries of the two vehicles within a certain time range is traversed to determine whether there is a risk of collision.

[0110] Combination Figure 4 As shown, in step S3, the risk indicators include: the shortest distance between the collision boundary of the bus and the surrounding vehicles, the time of occurrence, and the shortest reaction time, which specifically includes the following sub-steps:

[0111] S3.1. Upper limit of predicted time for buses t upper Can be dynamically calibrated as:

[0112]

[0113] In the formula, the driver’s reaction time is Δt react , are the limit deceleration and safety margin of the bus, v n is the velocity vector of the bus;

[0114] S3.2, bus n and surrounding vehicle m begin to intersect at time t0. The surrounding vehicle m is considered as a relatively stationary vehicle and bus n is considered as a relatively moving vehicle. Then the collision boundary l no Keep moving towards l mi Get closer and calculate the time range of the collision prediction at time t0, as follows:

[0115] t∈[t0,t0+t upper ],

[0116] S3.3. The shortest distance d between collision boundaries within the time range of collision prediction min The change in the collision prediction time range d min The minimum value d cpa The time t at which the minimum occurs cpa ;

[0117] S3.4. Calculate the relative velocity vector v between the surrounding vehicles and the bus o , as follows:

[0118] v o =v n -v m ,

[0119] In the formula, v m ,v n are the velocity vectors of the surrounding vehicles and the bus respectively;

[0120] S3.5. Obtain the collision boundary l of the surrounding vehicle m mi The parametric equation is as follows:

[0121]

[0122] In the formula, x mi ,y mi Represents the collision boundary l mi The vertex coordinates of any point, β represents the control parameter of the parametric equation, and the value range of β is [0,1]; x mbi is the horizontal coordinate of the inner vertex of the rear wheel of the surrounding vehicle, y mbi is the ordinate of the vertex on the inner side of the rear wheel of the surrounding vehicle, x mfi is the horizontal coordinate of the vertex on the inner side of the front wheel of the surrounding vehicle, y mfi is the vertical coordinate of the inner vertex of the front wheel of the surrounding vehicles; S3.6, after Δt time, obtain the collision boundary l of bus n no The parametric equation is:

[0123]

[0124] In the formula, x no ,y no Represents the collision boundary l no The coordinates of any point on x ,μ y represents the unit vector in the X-axis and Y-axis directions, α represents the control parameter of the parametric equation, and the value range of α is [0,1]; x nbo is the horizontal coordinate of the vertex on the outside of the rear wheel of the bus, y nbo is the ordinate of the vertex on the outside of the rear wheel of the bus, x nfois the horizontal coordinate of the vertex on the outside of the front wheel of the bus, y nfo It is the ordinate of the vertex on the outside of the front wheel of the bus.

[0125] S3.7. Calculate the collision boundary l at time t0+Δt mi ,l no The distance d between any two points p and q on pq (t0+Δt), as follows:

[0126]

[0127] S3.8. Order beg When the minimum value is obtained, it is the shortest distance between the collision boundaries. Specifically: The partial derivative about (α, β) is 0, and we get a set of equations. Solving the equations we get The equation is:

[0128]

[0129] If satisfied This indicates that points p and q are on the collision boundary; if Then let α=1orβ=1, and convert the shortest distance between line segments into the shortest distance from the vertex to the line segment, then we have:

[0130]

[0131] The shortest distance d between collision boundaries min (t0+Δt) is expressed as:

[0132]

[0133] S3.9. Calculate the minimum value of the closest distance between collision boundaries within the time range of collision prediction, as well as the corresponding occurrence time and the shortest reaction time. Specifically, in [t0, t0+t upper ] within the range of d min The minimum value of (t0+Δt), the corresponding time and the shortest reaction time are recorded as d cpa ,t cpa ,Δt min , then:

[0134]

[0135] t cpa =t0+Δt min .

[0136] like Figure 5 As shown in the figure, when there are several surrounding vehicles, it is determined that the bus is Δt away from the surrounding vehicles. minThe minimum value of can be used to obtain the vehicle that poses the greatest collision threat to the bus at the current moment, and the shortest reaction time of multiple vehicles, as shown in the following formula:

[0137] Δt gt =min(Δt min (m,n)),m=1,2,3,...;

[0138] In the formula, Δt gt Indicates the shortest reaction time of multiple workshops, Δt min (m,n) represents the shortest reaction time between surrounding vehicle m and bus n.

[0139] like Figure 6 As shown in the figure, the dynamic changes of collision risks in different directions during the driving process of a large bus are shown. The red area represents the collision buffer space of the vehicle, and the green area is the vehicle operation response space. When the collision buffer space is reduced, the vehicle collision risk will increase.

[0140] In summary, the method of the present invention comprehensively considers physical information such as vehicle shape, speed, heading angle, etc., establishes an interactive relationship with other traffic participants, characterizes the vehicle's spatial state through a directional bounding box model, establishes a collision buffer area between vehicles to detect and perceive the potential collision boundary between vehicles, and implements dynamic calibration of the collision prediction time range based on the vehicle state and the driver's reaction ability. Finally, the method uses the principles of geometry and vehicle kinematics to predict the closest distance and the shortest reaction time between vehicle collision boundaries, and determines the vehicle with the greatest risk of collision with the vehicle under complex traffic flows, thereby providing strong technical support for ensuring the driver's driving safety.

[0141] Although the present invention has been described above with preferred embodiments, it is not intended to limit the present invention. A person skilled in the art of the present invention may make various modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the definition of the claims.

Claims

1. A method for predicting the omnidirectional collision risk of a large passenger vehicle, characterized in that: The following steps are involved: S1. Based on the vehicle configuration, speed, and heading angle elements, a large bus body coordinate system and a global coordinate system are established, and then a directional bounding box of the large bus and surrounding vehicles is constructed; S2, combining the shapes and motion states of the vehicles around the large bus, constructing a collision buffer area for the large bus and detecting the collision boundaries between the surrounding vehicles and the large bus; S3. Based on the principles of geometry and vehicle kinematics, predict the risk indicators of large buses and surrounding vehicles within a preset time range.

2. A large bus omnidirectional collision risk prediction method according to claim 2, characterized in that: Establish the large bus body coordinate system and the global coordinate system to represent the vertex coordinates of the oriented bounding box, as follows: Rear wheel inner vertex e nbi , whose coordinates (x nbi ,y nbi )for: Rear wheel outer vertex e nbo , whose coordinates (x nbo ,y nbo )for: Front wheel inner vertex e nfi , whose coordinates (x nfi ,y nfi )for: Front wheel outer vertex e nfo , whose coordinates (x nfo ,y nfo )for: In the formula, θ n is the vehicle heading angle, (x nbc ,y nbc ) is the center point of the rear axle of the vehicle, L n ,W n is the length and width of the vehicle's oriented bounding box.

3. A method for predicting omnidirectional collision risk of a large passenger vehicle according to claim 1, characterized in that: In step S1, the directional bounding box includes the length, width, real-time speed and heading angle elements of the bus, and the minimum circumscribed rectangle E of the vehicle is obtained with the front and rear wheels and the inner and outer sides of the bus as the boundaries. n , as follows: AND n ={and nbi ,And nbo ,And nfi ,And nfo }; Where n is the vehicle number, b and f are the front and rear wheels of the vehicle, i and o are the inside and outside of the vehicle, respectively. nbi is the inner vertex of the rear wheel, e nbo It is the outer vertex of the rear wheel of the bus, e nfi It is the inner vertex of the front wheel of the bus, e nfo It is the outer vertex of the front wheel of the bus.

4. A method for predicting omnidirectional collision risk of a large passenger vehicle according to claim 1, characterized in that: In step S1, the oriented bounding box vertex position vector coordinates r e The vehicle body coordinate system and the global coordinate system are expressed as follows: r e =r nbc +R n ·L size ; In the formula, r nbc is the position vector of the center point of the rear axle of the vehicle, L size is the size matrix of the vehicle's oriented bounding box, R n is the vehicle rotation matrix.

5. A method for predicting omnidirectional collision risk of a large passenger vehicle according to claim 3, characterized in that: In step S1, the oriented bounding box boundary is l nk , the relationship between the boundary and the vertex is as follows: l ni is the inner collision boundary of the bus, l nf is the collision boundary of the rear wheel of the bus, l no is the outer collision boundary of the bus, l nb It is the collision boundary of the front wheels of the bus.

6. A method for predicting omnidirectional collision risk of a large passenger vehicle according to claim 4, characterized in that: Step S2 includes the following sub-steps: S2.

1. Obtain the minimum convex hull Vex that encloses the oriented bounding boxes of the two vehicles mn , Vex mn Satisfy the following formula: AND m ,AND n ∈Vex mn ; In the formula, E m ,E n are the vertices of the oriented bounding boxes of the surrounding vehicles m and bus n respectively; S2.

2. Perform an XOR operation on the minimum convex hull and the vehicle oriented bounding box to obtain the collision buffer area between the two vehicles, as shown in the following formula: buffer=xor(xor(Vex mn ,E m ),E n ), The potential collision boundary between the surrounding vehicles and the bus satisfies the following formula: In the formula, l mi is the inner collision boundary of the bus, l mf is the collision boundary of the rear wheel of the bus, l mo is the outer collision boundary of the bus, l mb It is the collision boundary of the front wheel of the bus; S2.3 traverses and solves the change of the shortest distance between the collision boundary of the bus and the surrounding vehicles within a preset time range to determine the collision risk between the bus and the surrounding vehicles.

7. A method for predicting omnidirectional collision risk of a large passenger vehicle according to claim 2, characterized in that: In step S3, the risk indicators include: the shortest distance between the collision boundary of the bus and surrounding vehicles, the time of occurrence, and the shortest reaction time, which specifically includes the following sub-steps: S3.

1. Upper limit of predicted time for buses t upper Can be dynamically calibrated as: In the formula, the driver’s reaction time is Δt react , Δt s are the limit deceleration and safety margin of the bus, v n is the velocity vector of the bus; S3.2, bus n and surrounding vehicle m begin to intersect at time t0. The surrounding vehicle m is considered as a relatively stationary vehicle and bus n is considered as a relatively moving vehicle. Then the collision boundary l no Keep moving towards l mi Get closer and calculate the time range of the collision prediction at time t0, as follows: t∈[t0,t0+t upper ], S3.

3. The shortest distance d between collision boundaries within the time range of collision prediction min The change in the collision prediction time range d min The minimum value d cpa The time t at which the minimum occurs cpa ; S3.

4. Calculate the relative velocity vector v between the surrounding vehicles and the bus o , as follows: v o =v n -v m , In the formula, v m ,v n are the velocity vectors of the surrounding vehicles and the bus respectively; S3.

5. Obtain the collision boundary l of the surrounding vehicle m mi The parametric equation is as follows: In the formula, x mi ,y mi Represents the collision boundary l mi The vertex coordinates of any point, β represents the control parameter of the parametric equation, and the value range of β is [0,1]; x mbi is the horizontal coordinate of the inner vertex of the rear wheel of the surrounding vehicle, y mbi is the vertical coordinate of the vertex on the inner side of the rear wheel of the surrounding vehicle, x mfi is the horizontal coordinate of the vertex on the inner side of the front wheel of the surrounding vehicle, y mfi is the ordinate of the vertex on the inner side of the front wheel of the surrounding vehicle; S3.

6. After Δt time, obtain the collision boundary l of bus n no The parametric equation is: In the formula, x no ,y no Represents the collision boundary l no The coordinates of any point on x ,μ y represents the unit vector in the X-axis and Y-axis directions, α represents the control parameter of the parametric equation, and the value range of α is [0,1]; x nbo is the horizontal coordinate of the vertex on the outside of the rear wheel of the bus, y nbo is the ordinate of the vertex on the outside of the rear wheel of the bus, x nfo is the horizontal coordinate of the vertex on the outside of the front wheel of the bus, y nfo It is the ordinate of the vertex on the outside of the front wheel of the bus. S3.

7. Calculate the collision boundary l at time t0+Δt mi ,l no The distance between any two points p and q on the graph is dpq(t0+Δt), as follows: S3.

8. Order beg When the minimum value is obtained, it is the shortest distance between collision boundaries. S3.

9. Calculate the minimum value of the closest distance between collision boundaries within the time range of collision prediction, as well as the corresponding occurrence time and the shortest reaction time.

8. A method for predicting omnidirectional collision risk of a large passenger vehicle according to claim 7, characterized in that: Step S3.8 is as follows: The partial derivative about (α, β) is 0, and we get a set of equations. Solving the equations we get The equation is: If satisfied This indicates that points p and q are on the collision boundary; if Then let α=1orβ=1, and convert the shortest distance between line segments into the shortest distance from the vertex to the line segment, then we have: The shortest distance d between collision boundaries min (t0+Δt) is expressed as:

9. A method for predicting omnidirectional collision risk of a large passenger vehicle according to claim 8, characterized in that: Step S3.9 is as follows: in [t0, t0+t upper ] within the range of d min The minimum value of (t0+Δt), the corresponding time and the shortest reaction time are recorded as d cpa ,t cpa ,Δt min , then: t cpa =t0+Δt min 。 When there are several surrounding vehicles, determine the distance between the bus and the surrounding vehicles. min The minimum value of can be used to obtain the vehicle that poses the greatest collision threat to the bus at the current moment, and the shortest reaction time of multiple vehicles, as shown in the following formula: Δt gt =min(Δt min (m,n)),m=1,2,3,...; In the formula, Δt gt Indicates the shortest reaction time of multiple workshops, Δt min (m,n) represents the shortest reaction time between surrounding vehicle m and bus n.

Images