Intelligent vehicle path planning method based on dynamic window and improved artificial potential field method

By combining the dynamic window method and the improved artificial potential field method, the problem of inefficiency in the disconnection function in the existing technology is solved, and the efficiency and safety of intelligent vehicle path planning and obstacle avoidance in complex environments is achieved, which significantly improves the robustness and success rate.

CN118960771APending Publication Date: 2024-11-15四川吉利学院
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411063046.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-05
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The detachment function in the existing artificial potential field method is inefficient in complex environments and has limited success rate, making it difficult to effectively help vehicles get rid of local minimum values, resulting in poor results in intelligent vehicle path planning and obstacle avoidance.

Method used

Combining the dynamic window method and the improved artificial potential field method, through safety judgment and multiple obstacle avoidance strategies, the robustness and success rate of the disengagement function are enhanced, and the dynamic window method is used to calculate the synergy and adjust the vehicle status to ensure that the vehicle is in a complex environment. Arrive to the end point safely and effectively.

Benefits of technology

It improves the robustness and success rate of intelligent vehicle path planning and obstacle avoidance, enhances the safety assessment and response speed in complex environments, and significantly improves the effectiveness of driving safety and obstacle avoidance strategies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118960771A_ABST
    Figure CN118960771A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of intelligent vehicles, and provides an intelligent vehicle path planning method based on a dynamic window and an improved artificial potential field method, and the method comprises the following steps: 1, confirming a planning path, extracting related data after completing parameter initialization and environment preparation, and executing global planning, so as to determine an advancing track, and calculating the current resultant force direction of the vehicle, judging whether the vehicle is trapped in a local optimal solution, and if so, calling a separation function to help the vehicle get rid of the local optimal solution. By enhancing safety evaluation in a path planning strategy, an intelligent vehicle can make a more accurate obstacle avoidance decision in a complex environment, in the path planning strategy, a collision prediction model is a key component, the principles of a dynamic window method and an artificial potential field method are fused, a plurality of factors influencing the braking performance are comprehensively considered, and the path planning strategy is optimized. Such as road conditions, relative distance between the vehicles and respective brake acceleration, so that the safety distance between the vehicles can be accurately evaluated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent vehicles, and in particular to an intelligent vehicle path planning method based on a dynamic window and an improved artificial potential field method. Background Art

[0002] Artificial Potential Fields (APF) is an effective method for mobile vehicle path planning. The core of this method is to construct a potential field in the vehicle motion space, which consists of the attraction of the target point and the repulsion of obstacles, guiding the vehicle to move from the starting point to the target point along the direction of the potential energy gradient. The escape function is a key mechanism designed in the APF to enhance the robustness of the method, especially in complex environments. Its main function is to help the vehicle escape from the local minimum point of the potential field. When the vehicle may be trapped in a local optimal solution, that is, when the resultant force is too small to drive the vehicle to move effectively, the escape function helps the vehicle escape from this state by applying random perturbations. In the original APF, this mechanism is simplified to let the vehicle randomly move a small step in any direction at the next moment to break the local optimal deadlock. The resultant force calculation is a key step in the APF, which is based on two physical principles: the principle of virtual work and the principle of structural static equilibrium. Dynamic Window Approach (DWA) is a widely used method in local path planning and obstacle avoidance strategies. It takes into account the mechanical characteristics of the vehicle itself and uses the dynamic window mechanism to quickly determine the most suitable movement trajectory and speed configuration. The calculation window in the dynamic window method refers to the set of all possible speeds (linear and angular velocities) that can be achieved within a specific time period based on the dynamic constraints of the vehicle (such as maximum acceleration and speed limit). This window has dynamic characteristics and changes with the current speed and acceleration capability of the vehicle.

[0003] The main purpose of the escape function is to help the vehicle find a new path to continue moving forward when it encounters a local minimum. The escape function in the original algorithm simply guides the vehicle to move forward randomly to escape the local minimum, but this method is inefficient, has a limited success rate, and is not suitable for real vehicle driving environments.

[0004] In order to solve the above problems, the study combined the disengagement function with the dynamic window method to enhance its robustness and success rate in practical applications, so as to improve the effectiveness of the fusion algorithm in intelligent vehicle path planning and obstacle avoidance. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an intelligent vehicle path planning method based on a dynamic window and an improved artificial potential field method. The study combines the escape function with the dynamic window method to enhance its robustness and success rate in practical applications, so as to improve the effectiveness of the fusion algorithm in intelligent vehicle path planning and obstacle avoidance.

[0006] To achieve the above objectives, the present invention is implemented by the following technical solutions: an intelligent vehicle path planning method based on a dynamic window and an improved artificial potential field method comprises the following steps:

[0007] Step 1: Confirm the planned route. After completing parameter initialization and environment preparation, extract relevant data and perform global planning to establish the travel trajectory, calculate the direction of the current force on the vehicle, and determine whether it is trapped in the local optimal solution. If it is trapped in the local optimal solution, call the escape function to help the vehicle escape from the local optimal solution. Otherwise, retreat the speed, direction and position until the reasonable direction is calculated and judged before reaching the destination;

[0008] Step 2: Safety judgment: extract the vehicle's current status information and run the safety judgment function to determine the vehicle's current safety status and make a safety judgment. The safety level is divided into three levels: high, medium, and low. The system will adopt any of the following actions: continuous monitoring of obstacle positions, mild obstacle avoidance, or emergency obstacle avoidance strategies.

[0009] Step 3: Choosing the obstacle avoidance mode: According to the return result of the safety judgment function, select a suitable obstacle avoidance mode from the three different obstacle avoidance execution modes: high, medium, and low. Update the vehicle status accordingly, and then make a judgment to determine whether the system needs to return to the global plan to recalculate the path to ensure that the vehicle reaches the destination safely and efficiently.

[0010] Preferably, in step 1, the threshold value of the resultant force acting on the vehicle is set to M. When the resultant force acting on the vehicle is less than M for five consecutive times, it is determined to be trapped in a local minimum problem, and the dynamic window method is used to plan the next path.

[0011] Preferably, the dynamic window method is a calculation method for escaping from the function, and the number of iterations of the escaping function is marked as an upper limit N to record the number of iterations of the escaping function. When the number of iterations reaches the upper limit and the vehicle still fails to escape from the local minimum point, the algorithm will reduce the calculation step size and try multiple possibilities to help the vehicle escape from the local minimum problem.

[0012] Preferably, the safety judgment in step 2 includes the vehicle's dynamic performance, environmental perception capability, accuracy of the decision-making system and its ability to cope with unknown factors, and the braking distance safety factor and environmental density coefficient are used in combination to judge the vehicle's current safety, so as to enhance environmental adaptability and driving safety.

[0013] Preferably, the environmental density coefficient is used to judge the environmental density coefficient by constructing a special function to evaluate the density of the vehicle's surrounding environment and provide data for comprehensive judgment of the current safety status. The function adopts two distance measurement methods: the Euclidean distance used to quantify the straight-line distance between the vehicle and each obstacle, indicating the shortest distance between the vehicle and the obstacle, and the Manhattan distance considering the actual navigation path, representing the actual feasible path distance between the vehicle and the obstacle.

[0014] Preferably, the braking distance safety factor uses an exponential function model to describe the relationship between the safety distance after braking and the braking distance safety factor, which can simulate the rapid increase behavior of the safety factor when the braking distance increases, thereby providing a more accurate and real-time safety assessment for the driving system.

[0015] Preferably, when the safety in step three is high, the environment will be monitored to detect whether the position of obstacles in the environment has changed. When the position of obstacles changes, the global path planning process will be triggered to recalculate the driving route.

[0016] Preferably, when the safety is medium in step three, a mild obstacle avoidance strategy under the dynamic window method will be adopted. First, the linear speed will be adjusted, and the linear speed and linear acceleration calculation windows in the calculation window function will be called, and the evaluation window, vehicle speed and obstacle window will be called out. After selecting the optimal window in the evaluation window, the system status will be updated and the safety will be evaluated again. If the safety is high, the evaluation will be terminated, otherwise the linear speed adjustment and window evaluation will be performed again.

[0017] Preferably, when the safety is low in step three, an emergency obstacle avoidance strategy will be adopted. First, the linear speed and angular speed are adjusted, and the linear speed and angular speed related windows in the calculation window function are called, and the evaluation window, vehicle speed and obstacle window are called out. After selecting the optimal window in the evaluation window, the system status is updated and the safety is evaluated again. If the safety is high, the evaluation is terminated, otherwise the linear speed adjustment and window evaluation are performed again.

[0018] Preferably, the process of the system updating the vehicle status is continuous until the system's safety rating is improved, and as the safety rating changes, the system will adjust its obstacle avoidance strategy accordingly to ensure real-time response to various situations to maximize the safety of the vehicle.

[0019] The present invention provides an intelligent vehicle path planning method based on a dynamic window and an improved artificial potential field method. It has the following beneficial effects:

[0020] 1. The present invention helps intelligent vehicles make more accurate obstacle avoidance decisions in complex environments by enhancing the safety assessment in the path planning strategy. In the path planning strategy, the collision prediction model is a key component, which integrates the principles of the dynamic window method and the artificial potential field method, and comprehensively considers multiple factors that affect the braking performance, such as road conditions, the relative distance between vehicles and their respective braking accelerations, so as to accurately assess the safe distance between vehicles.

[0021] 2. The method of the present invention improves the reaction speed of the intelligent driving system and significantly improves driving safety. Combining these factors, this model provides strong data support for intelligent vehicles to effectively avoid obstacles and plan paths in complex environments.

[0022] 3. The process of updating the vehicle status by the method of the present invention is a continuous state until the safety rating of the system is improved, and as the safety rating changes, the system will adjust its obstacle avoidance strategy accordingly to ensure real-time response to various situations to maximize the safety of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is the algorithm flow chart of the present invention;

[0024] Figure 2 This is an obstacle avoidance flow chart when the safety level is high in the present invention;

[0025] Figure 3 This is a flowchart of obstacle avoidance when the safety level is medium of the present invention;

[0026] Figure 4 This is the obstacle avoidance flow chart when the safety level is low according to the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Example:

[0029] Please see attached Figure 1 - Attachment Figure 4 The embodiment of the present invention provides an intelligent vehicle path planning method based on a dynamic window and an improved artificial potential field method, comprising the following steps:

[0030] Step 1: Confirm the planned route. After completing parameter initialization and environmental preparation, extract relevant data and perform global planning to establish the travel trajectory, calculate the direction of the current resultant force on the vehicle, and determine whether it is trapped in the local optimal solution. If it is trapped in the local optimal solution, call the escape function to help the vehicle escape the local optimal solution. Otherwise, retreat the speed, direction and position until the reasonable direction is calculated and judged before reaching the destination; set the resultant force threshold of the car to M. When the resultant force value of the vehicle is less than M for five consecutive times, it is determined to be trapped in the local minimum problem, and switch to the dynamic window method to plan the next path. The formula for calculating the resultant force on the vehicle is as follows:

[0031] net_force=α*attr_force+β*rep_force

[0032] Among them, α and β represent the weight coefficients of attraction and repulsion respectively. attr_force is the attraction force on the car. This step involves the vector addition calculation of the resultant force.

[0033] The dynamic window method is a calculation method for escaping from the function, and the number of iterations of the escaping function is marked as the upper limit N to record the number of iterations of the escaping function. When the number of iterations reaches the upper limit and the vehicle still fails to escape from the local minimum point, the algorithm will reduce the calculation step size and try multiple possibilities to help the vehicle escape from the local minimum problem. The specific expression is as follows:

[0034]

[0035] The above formula transforms time dt into To continue trying to escape from the local minimum problem. This improvement makes the escape function more robust and successful in practical applications, which helps to improve the effectiveness of the fusion algorithm in intelligent vehicle path planning and obstacle avoidance;

[0036] Step 2: Safety judgment: extract the vehicle's current status information and run the safety judgment function to determine the vehicle's current safety status and make safety judgments. The system will adopt any of the following actions: continuous monitoring of obstacle positions, mitigation obstacle avoidance, or emergency obstacle avoidance strategies. Safety judgment includes the vehicle's dynamic performance, environmental perception, accuracy of the decision-making system, and ability to respond to unknown factors. The braking distance safety factor and environmental density factor are used to comprehensively judge the vehicle's current safety in order to enhance environmental adaptability and driving safety. The vehicle safety factor formula is as follows:

[0037] S=EDI*S d

[0038] In the above formula, EDI is the environmental density coefficient, S dis the relative distance safety factor after braking. The final safety factor is the product of the environmental density factor and the braking distance safety factor. In this formula, it is stipulated that when 0<S≤0.4, the current safety level of the vehicle is low, when 0.4<S≤0.7, the safety level is medium, and when 0.7<S≤1, the safety level is high;

[0039] The environmental density coefficient is determined by constructing a special function to evaluate the density of the vehicle's surroundings and provide data for a comprehensive judgment of the current safety status. The function uses two distance measurement methods: the Euclidean distance, which is used to quantify the straight-line distance between the vehicle and each obstacle, indicating the shortest distance between the vehicle and the obstacle, and the Manhattan distance, which takes into account the actual navigation path and represents the actual feasible path distance between the vehicle and the obstacle. The environmental density index (EDI) can be applied to vehicle path planning to help select paths with fewer obstacles. Similarly, EDI also shows high practical value in obstacle avoidance decisions and environmental assessments. Most importantly, by using EDI, the subsequent update and maintenance costs can be effectively reduced, thereby improving overall efficiency and performance. The calculation formula is as follows:

[0040]

[0041] In the above formula, ω e and ω m Represent the weights of Euclidean distance and Manhattan distance in the EDI formula. i is the Euclidean distance, which calculates the straight-line distance between points. i is the Manhattan distance, which calculates the distance of the actual feasible route. In this formula, the weights of Manhattan distance and Euclidean distance can be adjusted to adapt to different environmental characteristics, which has better adaptability;

[0042] The collision prediction model is part of the intelligent vehicle path planning strategy that integrates the dynamic window method and the artificial potential field method. Based on physical principles, the model uses the relative distance to the vehicle in front after braking as the basis for judging the distance safety of the vehicle and provides a basis for obstacle avoidance operations. It comprehensively considers factors such as braking distance under emergency braking, initial relative distance and road conditions to evaluate the probability of collision. Since the current setting is vehicle automatic driving control, the reaction time and reaction process in non-automatic driving are not set. The current formula for the distance required for vehicle braking is as follows:

[0043] distance_self=(v**2) / (2*abs(a_self))

[0044] In the above formula, distance_self is the distance required for the current vehicle to brake, v is the speed of the current vehicle, and a_self is the braking acceleration of the current vehicle. The formula for the distance required for the front vehicle to brake is as follows:

[0045] distance_front=(car_v**2) / (2*abs(a_front))

[0046] In the above formula, distance_front is the distance required for the front vehicle to brake, car_v is the speed of the front vehicle, and a_front is the braking acceleration of the front vehicle. The final relative distance formula after braking is as follows:

[0047] d * =ds self +s front

[0048] Among them, d * That is the relative distance after braking;

[0049] The braking distance safety factor uses an exponential function model to describe the relationship between the safety distance after braking and the braking distance safety factor. It can simulate the rapid increase of the safety factor when the braking distance increases, thereby providing a more accurate and real-time safety assessment for the driving system. The braking safety factor determination formula is as follows:

[0050]

[0051] Among them, S d is the brake safety factor, d * is the relative distance after braking. In this study, the braking distance range is set between 0 and 50 meters, and the braking safety factor range is between 0 and 10;

[0052] This model enhances the safety assessment in the path planning strategy and helps intelligent vehicles make more accurate obstacle avoidance decisions in complex environments. In the path planning strategy of this study, the collision prediction model is a key component. It combines the principles of the dynamic window method and the artificial potential field method, and comprehensively considers multiple factors that affect braking performance, such as road conditions, the relative distance between vehicles, and their respective braking accelerations, so that the safe distance between vehicles can be accurately assessed. This method not only improves the reaction speed of the intelligent driving system, but also significantly improves driving safety. Combining these factors, this model provides strong data support for effective obstacle avoidance and path planning of intelligent vehicles in complex environments;

[0053] Step 3: Choosing the obstacle avoidance mode: According to the return result of the safety judgment function, select a suitable obstacle avoidance mode from the three different obstacle avoidance execution modes: high, medium, and low. Update the vehicle status accordingly, and then make a judgment to determine whether the system needs to return to the global plan to recalculate the path to ensure that the vehicle reaches the destination safely and efficiently.

[0054] When safety is high, the environment will be monitored to detect whether the position of obstacles in the environment has changed. When the position of obstacles changes, the global path planning process will be triggered to recalculate the driving route. During this process, the system will start the artificial potential field method and use its powerful computing power to recalculate the driving route in time. In this way, the system not only ensures high safety, but also significantly improves the flexibility and adaptability of path planning. It can quickly adapt to changes in the environment to ensure that the vehicle can maintain the best driving state in all situations;

[0055] When the safety is medium, the mitigation obstacle avoidance strategy under the dynamic window method will be adopted. First, the linear speed is adjusted, and the linear speed and linear acceleration calculation windows in the calculation window function are called. The evaluation window, the vehicle speed and obstacle window are called out. After selecting the optimal window in the evaluation window, the system status is updated and the safety is evaluated again. If the safety is high, the evaluation is ended. Otherwise, the linear speed adjustment and window evaluation are re-performed. This strategy adopts the mitigation obstacle avoidance strategy under the improved dynamic window method. The core of this strategy is to focus on calculating the vehicle's speed (v) and acceleration (V_ACC) in order to accurately control the vehicle's motion state and ensure that obstacles can be effectively avoided when facing potential risks. Through the precise calculation of speed and acceleration, this strategy not only ensures safety, but also takes into account the smoothness and comfort of driving.

[0056] The modified calculation window expression is as follows:

[0057] vd=[robot.v-kinematic.V_ACC*dt,

[0058] robot.v+kinematic.V_ACC*dt,

[0059] robot.w,robot.w]

[0060] In the above calculation window, the study adopted a method that calculates the speed-related parameters and fixes the angular velocity-related parameters. This mild obstacle avoidance method effectively avoids complex direction calculations and adjustments, and implements obstacle avoidance only by adjusting the speed, thereby greatly reducing the demand for computing resources. At the same time, this method reduces the risks that may be caused by excessive obstacle avoidance actions and enhances overall safety and comfort. During the evaluation window stage, the system analyzes the vehicle's speed and the relative position of the obstacle to determine the optimal driving window. The system then updates the vehicle status accordingly. This process continues until the system's safety rating improves. If the safety level changes, the system will automatically switch to the corresponding obstacle avoidance strategy to ensure the safety of the vehicle;

[0061] When the safety is low, an emergency obstacle avoidance strategy will be adopted. First, the linear speed and angular speed will be adjusted, and the linear speed and angular speed related windows in the calculation window function will be called. The evaluation window, the vehicle speed and obstacle evaluation window will be called out. After selecting the optimal window in the evaluation window, the system status will be updated and the safety will be evaluated again. If the safety is high, the evaluation will be ended. Otherwise, the linear speed adjustment and window evaluation will be performed again. This strategy uses the calculation window to comprehensively analyze the vehicle's speed (v), acceleration (V_ACC), angular velocity (w) and angular acceleration (W_ACC) to evaluate the safety under different driving conditions.

[0062] In the evaluation window step, in addition to the original orientation evaluation, obstacle distance evaluation and speed evaluation, the study also added a collision risk evaluation function. This function calculates the safety of the current path in real time based on the relative distance, speed and acceleration between the vehicle and the obstacle. When the model assesses a potential collision risk, it will adjust the path to avoid the collision. The specific mathematical formula of the collision risk evaluation function involves multiple factors, including the vehicle's speed, acceleration, relative distance and time, allowing this model to effectively evaluate and respond to potential risks in emergency situations. Its collision risk evaluation function is shown below:

[0063]

[0064] In the above formula, R(t) is the collision risk index at time t. rel (t) is the relative speed between the smart car and the obstacle, that is, the vector difference between their speeds. d(t) is the relative distance between the smart car and the obstacle. rel (t) is the relative acceleration between the smart car and the obstacle, that is, the vector difference of their accelerations. k is a proportional constant used to adjust the impact of acceleration on risk assessment. This formula can predict the risk value of collision and provide data support for the evaluation window. The optimized evaluation window expression is as follows:

[0065]

[0066] In the above formula, μ is the weighted coefficient of the collision risk evaluation function, and the inverse of the collision risk evaluation function is used as the score of the collision risk. The study optimized the angle calculation and selection in the emergency obstacle avoidance function. This optimization enables the function to select the obstacle avoidance angle more reasonably and safely, and the planned path tends to choose the direction with a lower probability of accidents. Such improvements avoid larger accidents that may be caused by randomly selecting the avoidance direction;

[0067] The system updates the vehicle status continuously until the system's safety rating improves, and as the safety rating changes, the system adjusts its obstacle avoidance strategy accordingly to ensure real-time response to various situations to maximize vehicle safety.

[0068] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent vehicle path planning method based on dynamic window and improved artificial potential field method, characterized in that: The following steps are involved: Step 1: Confirm the planned route. After completing parameter initialization and environment preparation, extract relevant data and perform global planning to establish the travel trajectory, calculate the direction of the current force on the vehicle, and determine whether it is trapped in the local optimal solution. If it is trapped in the local optimal solution, call the escape function to help the vehicle escape from the local optimal solution. Otherwise, retreat the speed, direction and position until the reasonable direction is calculated and judged before reaching the destination; Step 2: Safety judgment: extract the vehicle's current status information and run the safety judgment function to determine the vehicle's current safety status and make a safety judgment. The safety level is divided into three levels: high, medium, and low. The system will adopt any of the following actions: continuous monitoring of obstacle positions, mild obstacle avoidance, or emergency obstacle avoidance strategies. Step 3: Choosing the obstacle avoidance mode: According to the return result of the safety judgment function, select a suitable obstacle avoidance mode from the three different obstacle avoidance execution modes: high, medium, and low. Update the vehicle status accordingly, and then make a judgment to determine whether the system needs to return to the global plan to recalculate the path to ensure that the vehicle reaches the destination safely and efficiently.

2. The intelligent vehicle path planning method based on dynamic window and improved artificial potential field method according to claim 1 is characterized in that: In the step 1, the threshold of the resultant force acting on the vehicle is set to M. When the resultant force acting on the vehicle is less than M for five consecutive times, it is determined to be trapped in a local minimum problem, and the dynamic window method is used to plan the next path.

3. The intelligent vehicle path planning method based on dynamic window and improved artificial potential field method according to claim 2 is characterized in that: The dynamic window method is a calculation method for escaping from the function, and the number of iterations of the escaping function is marked as the upper limit N to record the number of iterations of the escaping function. When the number of iterations reaches the upper limit and the vehicle still fails to escape from the local minimum point, the algorithm will reduce the calculation step size and try multiple possibilities to help the vehicle escape from the local minimum problem.

4. The intelligent vehicle path planning method based on dynamic window and improved artificial potential field method according to claim 1 is characterized in that: The safety judgment in step 2 includes the vehicle's dynamic performance, environmental perception capability, accuracy of the decision-making system and its ability to cope with unknown factors. The braking distance safety factor and environmental density factor are used in combination to judge the vehicle's current safety, so as to enhance environmental adaptability and driving safety.

5. The intelligent vehicle path planning method based on dynamic window and improved artificial potential field method according to claim 4 is characterized in that: The environmental density coefficient is determined by constructing a special function to evaluate the density of the vehicle's surrounding environment and provide data for comprehensive judgment of the current safety status. The function adopts two distance measurement methods: the Euclidean distance, which is used to quantify the straight-line distance between the vehicle and each obstacle and indicates the shortest distance between the vehicle and the obstacle; and the Manhattan distance, which takes into account the actual navigation path and represents the actual feasible path distance between the vehicle and the obstacle.

6. The intelligent vehicle path planning method based on dynamic window and improved artificial potential field method according to claim 4 is characterized in that: The braking distance safety factor uses an exponential function model to describe the relationship between the safety distance after braking and the braking distance safety factor, and can simulate the rapid increase behavior of the safety factor when the braking distance increases, thereby providing a more accurate and real-time safety assessment for the driving system.

7. The intelligent vehicle path planning method based on dynamic window and improved artificial potential field method according to claim 1 is characterized in that: When the safety is high in step three, the environment will be monitored to detect whether the position of obstacles in the environment has changed. When the position of obstacles changes, the global path planning process will be triggered to recalculate the driving route.

8. The intelligent vehicle path planning method based on dynamic window and improved artificial potential field method according to claim 1 is characterized in that: When the safety is medium in the step three, a mild obstacle avoidance strategy under the dynamic window method will be adopted. First, the linear speed will be adjusted, and the linear speed and linear acceleration calculation windows in the calculation window function will be called, and the evaluation window, vehicle speed and obstacle window will be called out. After selecting the optimal window in the evaluation window, the system status will be updated and the safety will be evaluated again. If the safety is high, the evaluation will be ended. Otherwise, the linear speed adjustment and window evaluation will be performed again.

9. The intelligent vehicle path planning method based on dynamic window and improved artificial potential field method according to claim 1, characterized in that: When the safety is low in the step three, an emergency obstacle avoidance strategy will be adopted. First, the linear speed and angular speed are adjusted, and the linear speed and angular speed related windows in the calculation window function are called, and the evaluation window, vehicle speed and obstacle window are called out. After selecting the optimal window in the evaluation window, the system status is updated and the safety is evaluated again. If the safety is high, the evaluation is terminated. Otherwise, the linear speed adjustment and window evaluation are performed again.

10. The intelligent vehicle path planning method based on dynamic window and improved artificial potential field method according to any one of claims 7 to 9, characterized in that: The system updates the vehicle status continuously until the system's safety rating improves, and as the safety rating changes, the system adjusts its obstacle avoidance strategy accordingly to ensure real-time response to various situations to maximize vehicle safety.