A navigation path optimization method considering real-time accident risks

By establishing a real-time dynamic path optimization model, the problem that the navigation system cannot accurately describe the time-varying characteristics of factors on accident risk and congestion index is solved, and more accurate road accident prediction and navigation path optimization are achieved, and road traffic efficiency is improved.

CN117537836BActive Publication Date: 2024-11-19HEFEI UNIV OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311495928.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-11-10
Publication Date
2024-11-19
Estimated Expiration
2043-11-10

AI Technical Summary

Technical Problem

The existing navigation systems cannot accurately describe the comprehensive benefits of time-varying characteristics of factors on accident risk and congestion index, resulting in inaccurate optimization of real-time navigation paths.

Method used

Establish a real-time dynamic path optimization model, define dangerous driving behavior types by building road network scenarios, calculate the quantitative index of accident risk and dynamic traffic weights on the road, and optimize the vehicle's driving path to reduce accident risks.

Benefits of technology

It significantly improves the accuracy of road accident prediction, improves the rationality of navigation path optimization and road traffic efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117537836B_ABST
    Figure CN117537836B_ABST
Patent Text Reader

Abstract

The present invention discloses a navigation path optimization method considering real-time accident risks, including: 1. establishing a basic scenario for model application; 2. counting the number of vehicles corresponding to various types of risk behaviors within a cycle through a GPS receiver; 3. constructing a quantitative model for road accident risks within this cycle; 4. constructing a dynamic path optimization model; 5. entering the next cycle and repeating the above steps. The present invention can update the road accident risk index and congestion index in real time through the corresponding data collected in real time, thereby updating the traffic weight of each road section and updating the path optimization model. The navigation system can reduce the driving distance of the congested road section according to the updated model, ensure the safety of vehicle driving, and optimize the vehicle driving path.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of intelligent networked driving applications, and specifically is a navigation path optimization method that takes real-time accident risks into consideration. Background Art

[0002] With the popularization of transportation and people's increasing requirements for travel routes, traffic accidents also occur frequently. People are paying more and more attention to real-time path risks, while the existing navigation system mainly relies on driver experience, often only considering basic factors such as the shortest time or shortest distance, and ignoring the risk factors on the path. At present, some research and exploration have been conducted on the optimization of navigation paths with real-time path risks. Some studies have proposed risk-based navigation path optimization algorithms based on historical traffic data and real-time traffic information, which can more accurately predict path risks. However, the algorithm has high complexity and poor real-time performance, and it is difficult to find a balance between algorithm complexity and real-time performance.

[0003] In summary, the traditional navigation path optimization model cannot accurately describe the overall evaluation index of the time-varying characteristics of factors on accident risk and congestion index under the comprehensive benefits, resulting in inaccurate real-time navigation path optimization based on this evaluation. Summary of the invention

[0004] The present invention aims to solve the deficiencies of the above-mentioned prior art and proposes a navigation path optimization method that takes into account real-time accident risks, in order to optimize the vehicle driving path by establishing a real-time dynamic path optimization model, thereby improving vehicle traffic efficiency and vehicle driving safety.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme:

[0006] The method for optimizing a navigation path considering real-time accident risks of the present invention is characterized in that it is applied to a mixed driving environment of roads including connected vehicles and human drivers, and is performed according to the following steps;

[0007] Step 1: Create a basic scene;

[0008] Step 1.1: Construct a road network G = (V, L), where V represents the set of intersection nodes and V = {V1, V2, ..., V i ,…,V j ,…,V n},V i represents the i-th intersection node, V j represents the jth intersection node, n represents the total number of intersection nodes, L represents the set of roads between intersections, and L = {l ij |i,j=1,2,…,n;i≠j},l ij Indicates V iWith V j If V i With V j If there is a road between them, let l ij =1; if V i With V j If there is no road between ij =0;

[0009] Step 1.2, take any vertex of the circumscribed rectangle of the road network G = (V, L) as the origin, take a right-angled side passing through the origin as the X-axis, and take another right-angled side passing through the origin as the Y-axis, so as to construct an XOY coordinate system;

[0010] Step 2: define the types of dangerous driving behaviors, including risky speed, sudden lane change, three levels of sudden speed change, dangerous following, and mixed risks, and number them in sequence; any type of dangerous driving behavior is recorded as the nth risk type; mixed risk refers to dangerous driving behaviors with two or more of the following: risky speed, sudden lane change, sudden speed change, and dangerous following;

[0011] Let the road l at time t be ij The number of vehicles of the nth risk type is

[0012] Step 2.1: Calculate the road l at time t according to formula (1): ij The standard deviation of all vehicle speeds σ ij (t);

[0013]

[0014] In formula (1), represents the road l at time t ij The speed of the kth car, represents the road l at time t ij The average speed of all vehicles on the ij (t) represents the road l at time t ij The number of all vehicles on the

[0015] Step 2.2: Calculate the road l at time t according to formula (2): ij Number of vehicles in the risky speed category

[0016]

[0017] Step 2.3: Calculate the road l at time t using formula (3): ij Number of vehicles on the sudden lane change type

[0018]

[0019] In formula (3), represents the road l at time t ij The lane change steering angle of the kth vehicle, θ γ represents the risk turning angle threshold;

[0020] Step 2.4: Calculate the road l at time t using formula (4) ij The time required for the kth vehicle to collide with the vehicle in front of it when they are in a following state

[0021]

[0022] In formula (4), represents the road l at time t ij The front position of the k-th vehicle, represents the road l at time t ij The front position of the kth vehicle, E′ k represents the road l at time t ij The length of the preceding vehicle of the kth vehicle, represents the road l at time t ij The speed of the car in front of the kth car;

[0023] Step 2.5: Calculate the road l at time t using formula (5) ij Number of vehicles with dangerous car-following characteristics

[0024]

[0025] In formula (5), T d is the dangerous collision time threshold;

[0026] Step 3: Construct the road l at time t ij Accident risk quantification model;

[0027] Step 3.1: Calculate the road l at time t according to formula (6): ij The traffic density ρ ij (t);

[0028]

[0029] In formula (5), d ij Indicates road l ij Length;

[0030] Step 3.2: Calculate the road l at time t according to formula (7): ij Risk quantification index F ij (t);

[0031]

[0032] In formula (7), β n represents the weight coefficient of the nth risk type, q ij Indicates road l ij The road quality coefficient;

[0033] Step 4: Build a dynamic path optimization model:

[0034] Step 4.1: Determine the road l at time t by using formula (8) ij The congestion coefficient r ij (t);

[0035]

[0036] Step 4.2: Calculate the road l at time t using formula (9): ij The decision weight coefficient ω ij (t);

[0037]

[0038] Step 4.3: Calculate the road l at time t using formula (10): ij The dynamic traffic weight W ij (t);

[0039] W ij (t) = (1 + F ij (t))ω ij (t) (10)

[0040] Step 4.4: Construct the objective function of the dynamic path optimization model through formula (11):

[0041]

[0042] Formula (11) represents the minimization of the accident risk of the navigation vehicle passing through all road sections, where x ij Indicates whether the navigation vehicle passes through road l ij , when x ij =1, indicating that the navigation vehicle passes through road l ij , when x ij =0, indicating that the navigation vehicle does not pass through the road l ij ;

[0043] Step 4.5: Construct the constraints of the dynamic path optimization model through equations (12) to (16):

[0044]

[0045] In formula (12) to formula (16), x sjIndicates whether the navigation vehicle has passed the intersection V from the starting point s To any adjacent j-th intersection V j The road, x jt Indicates whether the navigation vehicle passes through any j-th intersection node V j To its final intersection V t The road, x ji Indicates whether the navigation vehicle passes through any i-th intersection node V i To any adjacent j-th intersection node V j section of the road;

[0046] Step 5: Use the solver to solve the dynamic path optimization model of the navigation vehicle so that the navigation vehicle can move from the starting intersection V s To the final intersection V t The sum of the dynamic traffic weights of all roads passed is minimized, so as to obtain the optimal global path and feed it back to the navigation vehicle to achieve optimization of the navigation path.

[0047] An electronic device of the present invention includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the method, and the processor is configured to execute the program stored in the memory.

[0048] The present invention provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program executes the steps of the method when executed by a processor.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. Compared with the traditional accident frequency prediction model, the method of the present invention incorporates time-varying data such as traffic conditions within the prediction period into the model, fully considering the impact of the time-varying characteristics of risk factors on accidents, thereby significantly improving the accuracy of road accident prediction;

[0051] 2. The present invention integrates multiple risk factors with the congestion index benefit to calculate the dynamic path traffic weight of the road network, which facilitates the navigation system to achieve global path optimization and improves the rationality of path optimization;

[0052] 3. The method of the present invention improves the efficiency of the navigation system in optimizing the vehicle path in real time by establishing a real-time dynamic path optimization model, thereby improving the road traffic efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is the overall flow chart of the present invention;

[0054] Figure 2 It is a detailed flow chart of the present invention;

[0055] Figure 3 A vehicle turning diagram of the present invention;

[0056] Figure 4 Schematic diagram of TTC(t) calculation of the present invention. DETAILED DESCRIPTION

[0057] In this embodiment, a navigation path optimization method considering real-time accident risks is applied in a networked environment. Figure 1 As shown, the navigation path optimization method is performed in the following steps:

[0058] Step 1: Create a basic scene;

[0059] Step 1.1: Construct a road network G = (V, L), where V represents the set of intersection nodes and V = {V1, V2, ..., V i ,…,V j ,…,V n},V i represents the i-th intersection node, V j represents the jth intersection node, n represents the total number of intersection nodes, L represents the set of roads between intersections, and L = {l ij |i,j=1,2,…,n;i≠j},l ij Indicates V i With V j If V i With V j If there is a road between them, let l ij =1; if V i With V j If there is no road between ij =0;

[0060] Step 1.2, take any vertex of the circumscribed rectangle of the road network G = (V, L) as the origin, take a right-angled side passing through the origin as the X-axis, and take another right-angled side passing through the origin as the Y-axis, so as to construct an XOY coordinate system;

[0061] Step 2: define the types of dangerous driving behaviors, including risky speed, sudden lane change, three levels of sudden speed change, dangerous following, and mixed risks, and number them in sequence; any type of dangerous driving behavior is recorded as the nth risk type; mixed risk refers to dangerous driving behaviors with two or more of the following: risky speed, sudden lane change, sudden speed change, and dangerous following;

[0062] Let the road l at time t be ij The number of vehicles of the nth risk type is

[0063] Step 2.1: Calculate the road l at time t according to formula (1): ij The standard deviation of all vehicle speeds σ ij (t);

[0064]

[0065] In the formula, represents the road l at time t ij The speed of the kth car, represents the road l at time t ij The average speed of all vehicles on the ij (t) represents the road l at time t ij The number of all vehicles on the

[0066] Step 2.2: Calculate the road l at time t according to formula (2): ij Number of vehicles in the risky speed category

[0067]

[0068] Step 2.3: Use the steering angle to determine whether the vehicle is a sudden lane change vehicle. Figure 3 As shown, the vehicle lane change trajectory is divided into three nodes, and the road l at time t is calculated by formula (3). ij Number of vehicles on the sudden lane change type

[0069]

[0070] In the formula, represents the road l at time t ij The lane change steering angle of the kth vehicle, θ γ represents the risk turning angle threshold;

[0071] Step 2.4: According to Figure 4 As shown, it is assumed that the two vehicles are on the same straight line. The road l at time t is calculated using formula (4). ij The time required for the kth vehicle to collide with the vehicle in front of it when they are in a following state

[0072]

[0073] In formula (4), represents the road l at time t ij The front position of the k-th vehicle, represents the road l at time t ij The front position of the kth vehicle, E′ k represents the road l at time t ij The length of the preceding vehicle of the kth vehicle, represents the road l at time t ij The speed of the car in front of the kth car;

[0074] Step 2.5: Calculate the road l at time t using formula (5) ij Number of vehicles with dangerous car-following characteristics

[0075]

[0076] Where, T d is the dangerous collision time threshold;

[0077] Step 3: According to Figure 2 As shown, construct the road l at time t ij Accident risk quantification model;

[0078] Step 3.1: Calculate the road l at time t according to formula (6): ij The traffic density ρ ij (t);

[0079]

[0080] In formula (5), d ij Indicates road l ij Length;

[0081] Step 3.2: Calculate the road l at time t according to formula (7): ij Risk quantification index F ij (t);

[0082]

[0083] In the formula, β n represents the weight coefficient of the nth risk type, q ij Indicates road l ij The road quality coefficient;

[0084] Step 4: Build a dynamic path optimization model:

[0085] Step 4.1: The GPS receiver determines whether the road is congested based on the density of all vehicles on the road, and determines the road l at time t using formula (8): ij The congestion coefficient r ij (t);

[0086]

[0087] Step 4.2: Calculate the road l at time t using formula (9): ij The decision weight coefficient ω ij (t);

[0088]

[0089] Step 4.3: The road l at time t obtained according to the previous steps ij The decision weight coefficient ω on ij (t), and similarly obtain the dynamic traffic weights of other road sections, and then obtain the traffic weight matrix, and calculate the road l at time t through formula (10) ij The dynamic traffic weight W ij (t);

[0090] W ij (t) = (1 + F ij (t))ω ij (t) (10)

[0091] Step 4.4: Construct the objective function of the dynamic path optimization model through formula (11):

[0092]

[0093] Formula (11) shows that the accident risk of the navigation vehicle passing through all road sections is minimized. ij =1, it means the navigation vehicle passes the road l ij , when x ij =0, indicating that the navigation vehicle does not pass the road l ij ;

[0094] Step 4.5: Construct the constraints of the dynamic path optimization model through equations (12) to (16):

[0095]

[0096] In formula (12) to formula (16), x sj Indicates whether the navigation vehicle has passed the intersection V from the starting point s To any adjacent intersection V j The road section, x jt Indicates whether the navigation vehicle passes through any intersection V j To its final intersection V t The road section, x ji Indicates whether the navigation vehicle passes through any intersection V i To any adjacent intersection V j section of the road;

[0097] Step 5: Use the solver to solve the dynamic path optimization model of the navigation vehicle, so that the navigation vehicle moves from V s To V t The sum of the dynamic weights of all the road sections passed is minimized, and the optimal global path is fed back to the vehicle, thereby achieving navigation path optimization.

[0098] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0099] In this embodiment, a computer-readable storage medium stores a computer program on the computer-readable storage medium, and the computer program executes the steps of the above method when executed by a processor.

Claims

1. A navigation path optimization method considering real-time accident risks, characterized in that: It is applied to mixed driving environments including connected vehicles and human-driven vehicles, and is carried out in the following steps; Step 1: Create a basic scene; Step 1.1: Construct a road network G = (V, L), where V represents the set of intersection nodes and V = {V1, V2, ..., V i ,…,V j ,…,V n },V i represents the i-th intersection node, V j represents the jth intersection node, n represents the total number of intersection nodes, L represents the set of roads between intersections, and L = {l ij |i,j=1,2,…,n;i≠j},l ij Indicates V i With V j If V i With V j If there is a road between them, let l ij =1; if V i With V j If there is no road between ij =0; Step 1.2, take any vertex of the circumscribed rectangle of the road network G = (V, L) as the origin, take a right-angled side passing through the origin as the X-axis, and take another right-angled side passing through the origin as the Y-axis, so as to construct an XOY coordinate system; Step 2: define the types of dangerous driving behaviors, including risky speed, sudden lane change, three levels of sudden speed change, dangerous following, and mixed risks, and number them in sequence; any type of dangerous driving behavior is recorded as the nth risk type; mixed risk refers to dangerous driving behaviors with two or more of the following: risky speed, sudden lane change, sudden speed change, and dangerous following; Let the road l at time t be ij The number of vehicles of the nth risk type is Step 2.1: Calculate the road l at time t according to formula (1): ij The standard deviation of all vehicle speeds σ ij (t); In formula (1), represents the road l at time t ij The speed of the kth car, represents the road l at time t ij The average speed of all vehicles on the ij (t) represents the road l at time t ij The number of all vehicles on the Step 2.2: Calculate the road l at time t according to formula (2): ij Number of vehicles in the risky speed category Step 2.3: Calculate the road l at time t using formula (3): ij Number of vehicles on the sudden lane change type In formula (3), represents the road l at time t ij The lane change steering angle of the kth vehicle, θ γ represents the risk turning angle threshold; Step 2.4: Calculate the road l at time t using formula (4) ij The time required for the kth vehicle to collide with the vehicle in front of it when they are in a following state In formula (4), represents the road l at time t ij The front position of the k-th vehicle, represents the road l at time t ij The front position of the kth vehicle, E′ k represents the road l at time t ij The length of the preceding vehicle of the kth vehicle, represents the road l at time t ij The speed of the car in front of the kth car; Step 2.5: Calculate the road l at time t using formula (5) ij Number of vehicles with dangerous car-following characteristics In formula (5), T d is the dangerous collision time threshold; Step 3: Construct the road l at time t ij Accident risk quantification model; Step 3.1: Calculate the road l at time t according to formula (6): ij The traffic density ρ ij (t); In formula (5), d ij Indicates road l ij Length; Step 3.2: Calculate the road l at time t according to formula (7): ij Risk quantification index F ij (t); In formula (7), β n represents the weight coefficient of the nth risk type, q ij Indicates road l ij The road quality coefficient; Step 4: Build a dynamic path optimization model: Step 4.1: Determine the road l at time t by using formula (8) ij The congestion coefficient r ij (t); Step 4.2: Calculate the road l at time t using formula (9): ij The decision weight coefficient ω ij (t); Step 4.3: Calculate the road l at time t using formula (10): ij The dynamic traffic weight W ij (t); W ij (t)=(1+F ij (t))ω ij (t) (10) Step 4.4: Construct the objective function of the dynamic path optimization model through formula (11): Formula (11) represents the minimization of the accident risk of the navigation vehicle passing through all road sections, where x ij Indicates whether the navigation vehicle passes through road l ij , when x ij =1, indicating that the navigation vehicle passes through road l ij , when x ij =0, indicating that the navigation vehicle does not pass through road l ij ; Step 4.5: Construct the constraints of the dynamic path optimization model through equations (12) to (16): In formula (12) to formula (16), x sj Indicates whether the navigation vehicle has passed the intersection V from the starting point s To any adjacent j-th intersection V j The road, x jt Indicates whether the navigation vehicle passes through any j-th intersection node V j To its final intersection V t The road, x ji Indicates whether the navigation vehicle passes through any i-th intersection node V i To any adjacent j-th intersection node V j section of the road; Step 5: Use the solver to solve the dynamic path optimization model of the navigation vehicle so that the navigation vehicle can move from the starting intersection V s To the final intersection V t The sum of the dynamic traffic weights of all roads passed is minimized, so as to obtain the optimal global path and feed it back to the navigation vehicle to achieve optimization of the navigation path.

2. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store a program that supports a processor to execute the method according to claim 1, and the processor is configured to execute the program stored in the memory.

3. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program executes the steps of the method according to claim 1 when executed by a processor.