Path planning method based on improved artificial potential field method

Abstract

The invention discloses a path planning method based on an improved artificial potential field method, and the method comprises the steps: generating a corresponding planning path according to a driving mode selected by a driver; according to the driving mode, a gravitational potential field, a repulsive potential field, a road boundary potential field and a lane line potential field in the corresponding mode are obtained in real time; the gravitational force, the repulsive force, the boundary force and the lane force borne by the main vehicle are calculated in real time, then the resultant force is calculated through a gradient descent method, and the driving direction of the vehicle along the planned path is adjusted in real time according to the resultant force. By introducing the concept of the dynamic risk field, the complex dynamic traffic environment can be flexibly dealt with, and collision with static or moving obstacles is effectively avoided, so that the driving safety and adaptability are improved; different driving modes are designed, the transverse and longitudinal weight factors and the gear parameters are adjusted according to the requirements of a driver, the path planning scheme meeting the personalized requirements is generated, and the comfort and diversity of driving experience are enhanced.

Description

Technical Field

[0001] The invention relates to a path planning method based on an improved artificial potential field method, and belongs to the technical field of vehicle automatic driving. Background Art

[0002] In autonomous driving and intelligent transportation systems, path planning is a crucial link, which is directly related to vehicle safety, driving efficiency and passenger comfort. Traditional path planning algorithms include Dijkstra algorithm based on graph theory, A* algorithm and RRT algorithm based on sampling, but these methods have certain limitations when dealing with complex dynamic environments and multiple constraints. In order to solve these problems, the artificial potential field method Artificial Potential Field (APF) is introduced. The artificial potential field method is widely used in robot path planning and obstacle avoidance tasks.

[0003] The artificial potential field method simulates a virtual force field, which makes the robot feel the attraction of the target point and the repulsion of obstacles during navigation, thus guiding it to move along a safe and efficient path. However, the traditional artificial potential field method has some shortcomings: it is prone to local optimal solutions and oscillation when dealing with dynamic obstacles and complex environments. In addition, the preferences and driving habits of different drivers for path selection have not been fully considered in traditional methods. Summary of the invention

[0004] Purpose of the invention: In view of the deficiencies in the prior art, the present invention provides a path planning method based on an improved artificial potential field method. The present invention generates a corresponding path by considering the driving mode selected by the driver, guides the vehicle along the path according to the change of the potential field, and effectively avoids collision with moving obstacles, thereby improving driving safety and adaptability.

[0005] Technical solution: A path planning method based on an improved artificial potential field method, comprising the following steps:

[0006] S1. According to the driving mode selected by the driver, the corresponding planned path is generated by simulation, and the vehicle drives according to the planned path;

[0007] S2, according to the driving mode selected in S1, combined with the current vehicle position, real-time acquisition of the gravitational potential field, repulsive potential field, road boundary potential field, and lane line potential field in the corresponding mode;

[0008] S3. Calculate the gravitational force, repulsive force, boundary force and lane force of the main vehicle in real time according to the gravitational potential field, repulsive potential field, road boundary potential field and lane line potential field respectively, and then use the gradient descent method to calculate the resultant force. According to the resultant force, adjust the direction of the vehicle along the planned path in real time, and judge in real time whether the vehicle has reached the end of the planned path. If not, return to S2 and continuously adjust the direction of the vehicle along the planned path. If it has arrived, end the path planning.

[0009] Preferably, the driving modes in S1 include a sport driving mode, an economic driving mode, and a comfort driving mode; the sport driving mode includes a sport first gear, a sport second gear, and a sport third gear; the economic driving mode includes an economy first gear, an economy second gear, and an economy third gear; the comfort driving mode includes a comfort first gear, a comfort second gear, and a comfort third gear; and the corresponding planned path is generated by simulation according to the gear position in the driving mode selected by the driver.

[0010] Preferably, the S2 specifically includes:

[0011] S201: Initialize artificial potential field method parameters;

[0012] S202: Obtaining gravitational potential field, repulsive potential field, road boundary potential field, and lane line potential field;

[0013] Preferably, the S201 is specifically:

[0014] Set the horizontal coordinate of the starting position of the main vehicle to X self , the vertical coordinate is Y self , the length of the main vehicle L self , Width W self , Main vehicle speed V self , target position, speed, length and width;

[0015] The horizontal coordinate of the obstacle vehicle position is X obs(i) , the vertical coordinate is Y obs(i) , the length of the obstacle vehicle L obs(i) , width W obs(i) , obstacle vehicle speed V obs(i) , i=1,2,...n;

[0016] Field strength coefficient A, parameter α used to adjust the impact of obstacles, speed difference proportional coefficient k v , the standard deviation of the horizontal expansion σ x , the standard deviation of the longitudinal expansion σ y , nonlinear factor β, nonlinear correction coefficient θ, speed difference weight factor γ, position difference weight factor δ.

[0017] Preferably, the step of obtaining the gravitational potential field in S202 is as follows:

[0018]

[0019] Among them, η att is the intensity coefficient of the gravitational potential field, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x goal is the horizontal coordinate of the target point, y goal is the vertical coordinate of the target point.

[0020] Preferably, the repulsive potential field is obtained in S202 specifically as follows:

[0021] The repulsive potential field includes static risk field and dynamic risk field;

[0022] Define the static risk field U sta for:

[0023]

[0024] Where A is the field strength coefficient, ω x is the lateral weight factor, ω y is the longitudinal weight factor, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x obs is the horizontal coordinate of the obstacle, y obs is the ordinate of the obstacle, σ x is the standard deviation of the horizontal expansion, σ y is the standard deviation of the longitudinal expansion, β is the nonlinear factor, and η is the correction parameter.

[0025] Define the dynamic risk field U dyn for:

[0026]

[0027] Among them, x obs (i) is the horizontal coordinate of the ith obstacle, y obs (i) is the ordinate of the ith obstacle, α is the parameter used to adjust the impact of the obstacle, and R v is the relative speed sign factor, which indicates the relationship between the obstacle speed and the vehicle speed: γ is the speed difference weight factor, which is used to adjust the impact of speed difference on the risk field. obs (i) is the length of the i-th obstacle, v obs (i) is the speed of the ith obstacle, v self is the speed of the main vehicle, σ v is the standard deviation of the velocity difference, σ v The formula is as follows:

[0028] σ v =k v |v obs(i)-v self |

[0029] Among them, k v It is the speed difference proportional coefficient. This parameter controls the influence of speed difference on standard deviation.

[0030] The nonlinear correction function is defined as:

[0031]

[0032] Among them, θ is the nonlinear correction coefficient, which controls the strength of the nonlinear correction part, that is, the sensitivity to the speed difference, and δ is the position difference weight factor, which controls the influence of the position difference on the correction function.

[0033] Preferably, the road boundary potential field is obtained in S202 as follows:

[0034] Define the road boundary potential field U edge :

[0035]

[0036] Among them, U max_edge is the maximum value of the boundary potential field, k edge is the attenuation coefficient of the boundary potential field, d edge1 is the distance from the current point to the upper boundary, d edge2 The distance from the current point to the lower boundary.

[0037] Preferably, the lane line potential field is obtained in S202 specifically as follows:

[0038] Define the lane line potential field U lane for:

[0039]

[0040] Among them, U max_lane is the maximum value of the lane line potential field, k lane is the attenuation coefficient of the lane line potential field, d lane is the distance from the current point to the lane line.

[0041] The road boundary potential field is used to describe the constraint of the road boundary on vehicle driving, ensuring that the vehicle will not go beyond the driving range of the road when planning the route and will stay within the legal area of ​​the road. The definition of the road boundary potential field is based on two core parameters: the maximum value of the boundary potential field and the attenuation coefficient. The maximum potential field value represents the strongest constraint of the boundary on the vehicle. When the vehicle approaches the road boundary, the potential field gradually increases. The attenuation coefficient controls the speed at which the potential field changes with distance. The closer the vehicle is to the road boundary, the greater the boundary potential field it is subjected to, driving the vehicle away from the road boundary and back to the safe driving area. In this way, the boundary potential field effectively prevents the vehicle from deviating from the road and ensures driving safety.

[0042] The lane line potential field is used to guide the vehicle to maintain a suitable driving position within the lane, avoid deviating from the lane or crossing the lane line, and ensure that the vehicle drives safely within the lane.

[0043] The lane line potential field is also defined by the maximum potential field value and attenuation coefficient of the lane line. The maximum potential field value of the lane line determines the magnitude of the repulsive force of the lane line on the vehicle. When the vehicle deviates from the center line of the lane, the potential field gradually increases, causing the vehicle to drive toward the center line of the lane with a lower potential field value. The attenuation coefficient determines how the potential field changes with the distance between the vehicle and the lane line. Specifically, the closer the vehicle is to the lane line, the greater the potential field value of the lane line, pushing the vehicle back to the center of the lane, thereby maintaining the driving stability of the vehicle in the lane.

[0044] The static risk field is used to describe the potential risk between the host vehicle and the surrounding stationary vehicles, and is only related to the length and width of the vehicle. The static risk field determines the potential collision risk by calculating the relative position between the host vehicle and the stationary obstacle. When the host vehicle approaches the stationary obstacle, the value of the static risk field will increase rapidly, indicating that the risk is increasing, forcing the host vehicle to avoid it. This risk assessment mechanism enables autonomous vehicles to effectively identify and avoid stationary obstacles in complex urban environments, ensuring driving safety.

[0045] The dynamic risk field is used to describe the risk between the main vehicle and the moving vehicles, considering not only the relative position between the main vehicle and the surrounding vehicles, but also their relative speed. As the speed of the main vehicle and the surrounding vehicles changes, the dynamic risk field adjusts the risk intensity in real time to reflect the potential collision risk. For the definition of the dynamic risk field, unlike the static risk field, the dynamic risk field not only considers the positional relationship between the main vehicle and the obstacle, but also focuses on the impact of the relative speed difference and position difference on the risk. The potential field combines various parameters to generate a dynamically adjusted risk field to reflect the risk level between the main vehicle and the moving obstacle in real time. When the relative speed of the main vehicle and the moving obstacle is greater or the distance is closer, the value of the dynamic risk field will increase, indicating a higher risk of collision, thereby prompting the main vehicle to take evasive action. This mechanism enables autonomous driving vehicles to respond to the threat of moving obstacles more flexibly and intelligently in high-speed or complex traffic environments, improving driving safety and efficiency.

[0046] The gravitational potential field is a virtual position used to guide the main vehicle towards the target point, and its function is to ensure that the vehicle can eventually reach the target position smoothly.

[0047] Preferably, S3 is specifically:

[0048] For the gravitational potential field U defined by S2 att , the partial derivative with respect to x is:

[0049]

[0050] The partial derivative of y is:

[0051]

[0052] Among them, η att is the intensity coefficient of the gravitational potential field, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x goal is the horizontal coordinate of the target point, y goal is the ordinate of the target point. The expression is:

[0053]

[0054] For the boundary potential field U defined by S2 edge , the partial derivative with respect to x is:

[0055]

[0056] The partial derivative of y is:

[0057]

[0058] Among them, U max_edge is the maximum value of the boundary potential field, k edge is the attenuation coefficient of the boundary potential field, d edge1 is the distance from the current point to the upper boundary, d edge2 is the distance from the current point to the lower boundary, i.e. the boundary force The expression is:

[0059]

[0060] For the lane line potential field U defined by S2 lane , the partial derivative with respect to x is:

[0061]

[0062] The partial derivative of y is:

[0063]

[0064] Among them, U max_lane is the maximum value of the lane line potential field, k lane is the attenuation coefficient of the lane line potential field, d lane is the distance from the current point to the lane line, i.e., lane force The expression is:

[0065]

[0066] For the static risk potential field U defined by S2 sta ,set up

[0067]

[0068] The partial derivative with respect to x is:

[0069]

[0070] The partial derivative of y is:

[0071]

[0072] Where A is the field strength coefficient, ω x is the lateral weight factor, ω y is the longitudinal weight factor, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x obs is the horizontal coordinate of the obstacle, y obs is the ordinate of the obstacle, σ x is the standard deviation of the horizontal expansion, σ y is the standard deviation of the vertical expansion, β is the nonlinear factor, and η is the correction parameter, that is, the static risk factor The expression is:

[0073]

[0074] For the dynamic risk field U defined by S2 dyn ,set up

[0075]

[0076]

[0077] The partial derivative with respect to x is:

[0078]

[0079] The partial derivative of y is:

[0080]

[0081] in,

[0082]

[0083]

[0084] Among them, x obs (i) is the horizontal coordinate of the ith obstacle, y Obs (i) is the ordinate of the ith obstacle, α is the parameter used to adjust the impact of the obstacle, and R v is the relative speed sign factor, which indicates the relationship between the obstacle speed and the vehicle speed. γ is the speed difference weight factor, which is used to adjust the impact of the speed difference on the risk field. obs (i) is the length of the i-th obstacle, v obs (i) is the speed of the ith obstacle, v self is the speed of the main vehicle, σ v is the standard deviation of the speed difference, and its formula is as follows:

[0085] σ v =k v |v obs (i)-v self |

[0086] Among them, k v is the speed difference proportional coefficient, which indicates the influence of speed difference on standard deviation, i.e. dynamic risk The expression is:

[0087]

[0088] The resultant force on the main vehicle The gravitational force on the main vehicle Boundary Force Lane Force and repulsion and Add together, the expression is:

[0089]

[0090] Beneficial effects: By introducing the concept of dynamic risk field, the present invention can flexibly respond to complex dynamic traffic environments and effectively avoid collisions with moving obstacles, thereby improving driving safety and adaptability; by designing different driving modes and adjusting the lateral and longitudinal weight factors and gear parameters according to driver needs, a path planning scheme that meets personalized needs is generated, thereby enhancing the comfort and diversity of the driving experience; by using the comprehensive application of boundary potential field, lane line potential field, gravitational potential field, static and dynamic risk fields, it can ensure that the vehicle travels within the legal area of ​​the road while avoiding static or moving obstacles, thereby improving the safety and reliability of path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0092] Figure 1 is a flow chart of the method of the present invention;

[0093] Figure 2 : is a road boundary potential field diagram of the present invention; wherein, (a) is a top view of the road boundary potential field; (b) is a front view of the road boundary potential field;

[0094] Figure 3 It is the improved total driving risk field map of the present invention;

[0095] Figure 4 1 is a simulation result diagram of the driving risk field under the economic driving mode of the present invention; wherein (a) is the economic driving mode - economic first gear; (b) is the economic driving mode - economic second gear; (c) is the economic driving mode - economic third gear;

[0096] Figure 5 It is a speed simulation diagram of three modes under the economic driving mode of the present invention;

[0097] Figure 6 It is the acceleration simulation diagram of three modes under the economic driving mode of the present invention;

[0098] Figure 71 is a simulation result diagram of the driving risk field under the comfortable driving mode of the present invention; wherein (a) is the comfortable driving mode - comfortable first gear; (b) is the comfortable driving mode - comfortable second gear; (c) is the comfortable driving mode - comfortable third gear;

[0099] Figure 8 It is a speed simulation diagram of three modes under the comfort driving mode of the present invention;

[0100] Fig. 9 It is the acceleration simulation diagram of three modes under the comfort driving mode of the present invention;

[0101] Fig.10 1 is a simulation result diagram of the driving risk field under the sports driving mode of the present invention; wherein (a) is the sports driving mode - sports first gear; (b) is the sports driving mode - sports second gear; (c) is the sports driving mode - sports third gear;

[0102] Fig.11 It is a speed simulation diagram of three modes under the sports driving mode of the present invention. DETAILED DESCRIPTION

[0103] 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.

[0104] In the description of the present invention, it is necessary to understand that the terms "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0105] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0106] like Figure 1 As shown, a path planning method based on an improved artificial potential field method comprises the following steps:

[0107] S1. According to the driving mode selected by the driver, the corresponding planned path is generated by simulation, and the vehicle drives according to the planned path;

[0108] The driving modes in S1 include a sport driving mode, an economic driving mode, and a comfort driving mode; the sport driving mode includes a sport first gear, a sport second gear, and a sport third gear; the economic driving mode includes an economic first gear, an economic second gear, and an economic third gear; the comfort driving mode includes a comfort first gear, a comfort second gear, and a comfort third gear. According to the gear position in the driving mode selected by the driver, the corresponding planned path is simulated and generated.

[0109] Parameters corresponding to different driving modes and gears ω x and ω y And the gear parameter η is shown in Table 1.

[0110] Table 1 Parameter settings of three driving modes

[0111]

[0112] Compared with the comfort driving mode, the economy driving mode x and ω y The value is lower because the smaller ω x and ω y It can enable the vehicle to make more agile avoidance actions when approaching an obstacle vehicle, thereby adjusting the path in a shorter time and distance, reducing detours, optimizing driving distance and time, and helping to save fuel or energy. x and ω y The potential field can guide the vehicle more accurately in a local area, thereby achieving local path optimization, avoiding long acceleration and deceleration processes, and maintaining the energy efficiency of the vehicle. x and ω y The higher the value, the larger the ω x and ω y The potential field works in a wider range, allowing the vehicle to sense obstacles or targets at a longer distance, thereby planning a smoother and more stable path. This can avoid frequent sharp turns and acceleration and deceleration, and improve ride comfort. x and ω y The value is smaller, which enhances the dynamic performance of the vehicle, allowing it to maintain high controllability and stability at high speeds and meet the vehicle's high requirements for speed and acceleration.

[0113] S2, according to the driving mode selected in S1, combined with the current vehicle position, real-time acquisition of the gravitational potential field, repulsive potential field, road boundary potential field, and lane line potential field in the corresponding mode;

[0114] S201: Initialize artificial potential field method parameters;

[0115] The research scenario of this embodiment is an urban scenario. In order to prevent the vehicle from deviating from the safe driving area and ensure that the vehicle travels within the predetermined road range, a road boundary potential field is established, such as Figure 2 The urban scene is a one-way two-lane road, each lane is 3.5m wide and the road length is 85m.

[0116] Vehicle operating environment Figure 3 As shown, the main vehicle is a blue box, and the position of the main vehicle is X 0 (10,1.75), target position X g (75, 1.75), the length of the main vehicle L self =4.8, width W self =2, the obstacle vehicle is a red box, and the positions of the obstacle vehicles are X 1 (10, -1.75), X 2 (30,1.75) and X 3 (50, -1.75), the lengths of the obstacle vehicles are L obs(1) =4.5, L obs(2) =5.5 and L obs(3) =8.2, the width of the obstacle vehicle is W obs(1) =1.8,W obs(2) =2 and W obs(3) =2.5, the speeds of the obstacle vehicles are V obs(1) =1, V obs(2) =3.8 and V obs(3) =2.

[0117] In the urban scene, the field strength coefficient A = 15, the parameter used to adjust the impact of obstacles α = 0.5, the speed difference proportional coefficient k v =0.1, standard deviation of lateral expansion σ x =2, standard deviation of longitudinal expansion σ y = 2, nonlinear factor β = 1.5, nonlinear correction coefficient θ = 1.5, velocity difference weight factor γ = 1, position difference weight factor δ = 1, maximum value of boundary potential field U max_edge =15, the attenuation coefficient k of the boundary potential field edge = 1, the maximum value of the lane line potential field U max_lane =6, the attenuation coefficient k of the lane line potential fieldlane =0.7.

[0118] S202: Obtaining gravitational potential field, repulsive potential field, road boundary potential field, and lane line potential field;

[0119] Define the gravitational potential field U att for:

[0120]

[0121] Among them, η att is the intensity coefficient of the gravitational potential field, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x goal is the horizontal coordinate of the target point, y goal is the vertical coordinate of the target point.

[0122] The repulsive potential field includes static risk field and dynamic risk field;

[0123] Define the static risk field U sta for:

[0124]

[0125] Where A is the field strength coefficient, ω x is the lateral weight factor, ω y is the longitudinal weight factor, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x obs is the horizontal coordinate of the obstacle, y obs is the ordinate of the obstacle, σ x is the standard deviation of the horizontal expansion, σ y is the standard deviation of the longitudinal expansion, β is the nonlinear factor, and η is the correction parameter.

[0126] Define the dynamic risk field U dyn for:

[0127]

[0128] Among them, x obs (i) is the horizontal coordinate of the ith obstacle, y obs (i) is the ordinate of the ith obstacle, α is the parameter used to adjust the impact of the obstacle, and R v is the relative speed sign factor, which indicates the relationship between the obstacle speed and the vehicle speed: γ is the speed difference weight factor, which is used to adjust the impact of speed difference on the risk field. obs (i) is the length of the i-th obstacle, v obs (i) is the speed of the ith obstacle, v self is the speed of the main vehicle, σ v is the standard deviation of the velocity difference, σv The formula is as follows:

[0129] σ v =k v |v obs (i)-v self |

[0130] Among them, k v It is the speed difference proportional coefficient. This parameter controls the influence of speed difference on standard deviation.

[0131] The nonlinear correction function is defined as:

[0132]

[0133] Among them, θ is the nonlinear correction coefficient, which controls the strength of the nonlinear correction part, that is, the sensitivity to the speed difference, and δ is the position difference weight factor, which controls the influence of the position difference on the correction function.

[0134] Define the road boundary potential field U edge :

[0135]

[0136] Among them, U max_edge is the maximum value of the boundary potential field, k edge is the attenuation coefficient of the boundary potential field, d edge1 is the distance from the current point to the upper boundary, d edge2 The distance from the current point to the lower boundary.

[0137] Define the lane line potential field U lane for:

[0138]

[0139] Among them, U max_lane is the maximum value of the lane line potential field, k lane is the attenuation coefficient of the lane line potential field, d lane is the distance from the current point to the lane line.

[0140] The road boundary potential field is used to describe the constraints of the road boundary on vehicle driving, ensuring that the vehicle does not go beyond the driving range of the road when planning the path and stays within the legal area of ​​the road.

[0141] The lane line potential field is used to guide the vehicle to maintain a suitable driving position within the lane, avoid deviating from the lane or crossing the lane line, and ensure that the vehicle drives safely within the lane.

[0142] The static risk field is used to describe the potential risk between the host vehicle and the surrounding stationary vehicles and is only related to the length and width of the vehicle.

[0143] The dynamic risk field is used to describe the risk between the host vehicle and the moving vehicles, taking into account not only the relative position between the host vehicle and the surrounding vehicles, but also their relative speed. As the speed of the host vehicle and the surrounding vehicles changes, the dynamic risk field adjusts the risk intensity in real time to reflect the potential collision risk.

[0144] The gravitational potential field is a virtual position used to guide the main vehicle towards the target point, and its function is to ensure that the vehicle can eventually reach the target position smoothly.

[0145] S3. Calculate the gravitational force, repulsive force, boundary force, and lane force of the main vehicle in real time based on the gravitational potential field, repulsive potential field, road boundary potential field, and lane line potential field, and then use the gradient descent method to calculate the resultant force. The basic idea is to move along the negative gradient direction of the calculation formula of the gravitational potential field, repulsive potential field, road boundary potential field, and lane line potential field function, because the negative gradient direction is the direction in which the function value decreases fastest. For a given function f(x), its gradient, that is, the partial derivative vector, represents the rate of change of the function in each direction. The gradient is the partial derivative of the potential field with respect to the position (x, y), and the negative direction of the gradient corresponds to the direction in which the object is subjected to force in the potential field.

[0146] The direction of the vehicle along the planned path is adjusted in real time according to the combined force, and it is determined in real time whether the vehicle has reached the end of the planned path. If not, it returns to S2 and continuously adjusts the direction of the vehicle along the planned path. If it has arrived, the path planning is terminated.

[0147] For the gravitational potential field U defined by S2 att , the partial derivative with respect to x is:

[0148]

[0149] The partial derivative of y is:

[0150]

[0151] Among them, η att is the intensity coefficient of the gravitational potential field, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x goal is the horizontal coordinate of the target point, y goal is the ordinate of the target point. The expression is:

[0152]

[0153] For the boundary potential field U defined by S2 edge , the partial derivative with respect to x is:

[0154]

[0155] The partial derivative of y is:

[0156]

[0157] Among them, U max_edge is the maximum value of the boundary potential field, k edge is the attenuation coefficient of the boundary potential field, d edge1 is the distance from the current point to the upper boundary, d edge2 is the distance from the current point to the lower boundary, i.e. the boundary force The expression is:

[0158]

[0159] For the lane line potential field U defined by S2 lane , the partial derivative with respect to x is:

[0160]

[0161] The partial derivative of y is:

[0162]

[0163] Among them, U max_lane is the maximum value of the lane line potential field, k lane is the attenuation coefficient of the lane line potential field, d lane is the distance from the current point to the lane line, i.e., lane force The expression is:

[0164]

[0165] For the static risk potential field U defined by S2 sta ,set up

[0166]

[0167] The partial derivative with respect to x is:

[0168]

[0169] The partial derivative of y is:

[0170]

[0171] Where A is the field strength coefficient, ω x is the lateral weight factor, ω y is the longitudinal weight factor, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x obs is the horizontal coordinate of the obstacle, y obs is the ordinate of the obstacle, σ x is the standard deviation of the horizontal expansion, σy is the standard deviation of the vertical expansion, β is the nonlinear factor, and η is the correction parameter, that is, the static risk factor The expression is:

[0172]

[0173] For the dynamic risk field U defined by S2 dyn ,set up

[0174]

[0175]

[0176] The partial derivative with respect to x is:

[0177]

[0178] The partial derivative of y is:

[0179]

[0180] in,

[0181]

[0182]

[0183] Among them, x obs (i) is the horizontal coordinate of the ith obstacle, y obs (i) is the ordinate of the ith obstacle, α is the parameter used to adjust the impact of the obstacle, and R v is the relative speed sign factor, which indicates the relationship between the obstacle speed and the vehicle speed. γ is the speed difference weight factor, which is used to adjust the impact of the speed difference on the risk field. obs (i) is the length of the i-th obstacle, v obs (i) is the speed of the ith obstacle, v self is the speed of the main vehicle, σ v is the standard deviation of the speed difference, and its formula is as follows:

[0184] σ v =k v |v obs (i)-v self |

[0185] Among them, k v is the speed difference proportional coefficient, which indicates the influence of speed difference on standard deviation, i.e. dynamic risk The expression is:

[0186]

[0187] The resultant force on the main vehicle The gravitational force on the main vehicle Boundary Force Lane Force and repulsion and Add together, the expression is:

[0188]

[0189] This embodiment selects the economical driving mode, and the simulation results of the driving risk field are as follows: Figure 4 As shown in the figure, it can be seen that the distance between the vehicle and the front vehicle is the largest in the economic third gear mode, followed by the economic second gear mode, and the distance between the vehicle and the front vehicle is the smallest in the economic first gear mode, and no lane change is selected in the three modes. Figure 5 It can be seen that the initial speed of the economic first gear mode is the fastest, followed by the economic second gear mode, and the initial speed of the economic third gear is the slowest, which is also in line with the actual situation. In the three modes, when the main vehicle approaches the first obstacle vehicle, because there is another obstacle vehicle in the other lane and its speed is slower, the overall speed of the three modes tends to be flat. Figure 6 It can be seen that the change in acceleration is the smallest in the third economic gear mode, and the change in acceleration is the largest in the first economic gear mode, which is also in line with the actual situation.

[0190] The simulation results of the driving risk field under the comfortable driving mode are as follows: Figure 7 As shown. It can be seen that the distance to the front car is the largest in the comfort third gear mode, followed by the comfort second gear mode, and the distance to the front car is the smallest in the comfort first gear mode. However, all three modes maintain a large distance from the front car, driving at a lower potential field position, and none of them chooses to change lanes. Figure 8 It can be seen that the overall speed of the comfort third gear mode is lower, the comfort second gear mode is second, and the comfort first gear mode is faster. Since the present invention adopts the gradient descent method, when the main vehicle approaches the target position, the gravity will gradually decrease. Therefore, when the main vehicle approaches the target point, the speeds of the three modes will gradually converge. Fig. 9 It can be seen that the acceleration changes in all modes in the comfort driving mode are small, basically maintained between 0-1. Compared with the economy driving mode, the acceleration changes are lower, which is in line with the characteristics of the comfort driving mode.

[0191] The simulation results of the driving risk field under the sports driving mode are as follows: Fig.10 As shown in Figure 2, it can be seen that the sports third gear mode changes lanes earlier than the other two modes, followed by the sports second gear mode, and the sports first gear mode changes lanes the latest. This is because the obstacle vehicle exhibits a lower potential field. Fig.11It can be seen that at the beginning of the main vehicle's driving, the speed is fast due to the small repulsive potential field around it. When the main vehicle gradually approaches the first obstacle vehicle in front, the repulsive potential field gradually increases, and the speed of the main vehicle gradually decreases. Due to the effect of the gravitational potential field, the main vehicle will avoid the first obstacle vehicle and choose to change lanes, and then start to accelerate. During the lane change process, it encounters the second obstacle vehicle, the repulsive potential field increases again, and the speed of the main vehicle decreases again, but due to the effect of the gravitational potential field, the main vehicle gradually accelerates and chooses to change lanes again, driving towards the target point. Due to the use of the gradient descent method, when the main vehicle gradually approaches the target point, the gravitational force gradually decreases, and the speed of the main vehicle also decreases until it reaches the target point.

[0192] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0193] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A path planning method based on an improved artificial potential field method, characterized in that: The following steps are involved: S1. According to the driving mode selected by the driver, the corresponding planned path is generated by simulation, and the vehicle drives according to the planned path; S2, according to the driving mode selected in S1, combined with the current vehicle position, real-time acquisition of the gravitational potential field, repulsive potential field, road boundary potential field, and lane line potential field in the corresponding mode; S3. Calculate the gravitational force, repulsive force, boundary force and lane force of the main vehicle in real time according to the gravitational potential field, repulsive potential field, road boundary potential field and lane line potential field respectively, and then use the gradient descent method to calculate the resultant force. According to the resultant force, adjust the direction of the vehicle along the planned path in real time, and judge in real time whether the vehicle has reached the end of the planned path. If not, return to S2 and continuously adjust the direction of the vehicle along the planned path. If it has arrived, end the path planning.

2. The path planning method based on the improved artificial potential field method according to claim 1, characterized in that: The driving modes in S1 include a sport driving mode, an economic driving mode, and a comfort driving mode; the sport driving mode includes a sport first gear, a sport second gear, and a sport third gear; the economic driving mode includes an economic first gear, an economic second gear, and an economic third gear; the comfort driving mode includes a comfort first gear, a comfort second gear, and a comfort third gear. According to the gear position in the driving mode selected by the driver, the corresponding planned path is simulated and generated.

3. The path planning method based on the improved artificial potential field method according to claim 2, characterized in that: The S2 specifically includes: S201: Initialize artificial potential field method parameters; S202: Obtaining the gravitational potential field, the repulsive potential field, the road boundary potential field, and the lane line potential field.

4. The path planning method based on the improved artificial potential field method according to claim 3, characterized in that: The S201 is specifically: Set the horizontal coordinate of the starting position of the main vehicle to X self , the vertical coordinate is Y self , the length of the main vehicle L self , Width W self , Main vehicle speed V self , target position, speed, length and width; The horizontal coordinate of the obstacle vehicle position is X obs(i) , the vertical axis is Y obs(i) , the length of the obstacle vehicle L obs(i) , width W obs(i) , obstacle vehicle speed V obs(i) , i=1,2,…n; Field strength coefficient A, parameter α used to adjust the impact of obstacles, speed difference proportional coefficient k v , the standard deviation of the horizontal expansion σ x , the standard deviation of the longitudinal expansion σ y , nonlinear factor β, nonlinear correction coefficient θ, speed difference weight factor γ, position difference weight factor δ.

5. The path planning method based on the improved artificial potential field method according to claim 4, characterized in that: The specific steps of obtaining the gravitational potential field in S202 are: Among them, η att is the intensity coefficient of the gravitational potential field, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x goal is the horizontal coordinate of the target point, y goal is the vertical coordinate of the target point.

6. The path planning method based on the improved artificial potential field method according to claim 5, characterized in that: The repulsive potential field is obtained in S202 specifically as follows: The repulsive potential field includes static risk field and dynamic risk field; Define the static risk field U sta for: Where A is the field strength coefficient, ω x is the lateral weight factor, ω y is the longitudinal weight factor, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x obs is the horizontal coordinate of the obstacle, y obs is the ordinate of the obstacle, σ x is the standard deviation of the horizontal expansion, σ y is the standard deviation of longitudinal expansion, β is the nonlinear factor, and η is the correction parameter; Defining the dynamic risk field U dyn for: Among them, x obs (i) is the horizontal coordinate of the ith obstacle, y obs (i) is the ordinate of the ith obstacle, α is the parameter used to adjust the impact of the obstacle, and R v is the relative speed sign factor, which indicates the relationship between the obstacle speed and the vehicle speed: γ is the speed difference weight factor, which is used to adjust the impact of speed difference on the risk field. obs (i) is the length of the i-th obstacle, v obs (i) is the speed of the ith obstacle, v self is the speed of the main vehicle, σ v is the standard deviation of the velocity difference, σ v The formula is as follows: σ v =k v |v obs (i)-v self | Among them, k v is the speed difference proportional coefficient, which controls the influence of speed difference on standard deviation; The nonlinear correction function is defined as: Among them, θ is the nonlinear correction coefficient, which controls the strength of the nonlinear correction part, that is, the sensitivity to the speed difference, and δ is the position difference weight factor, which controls the influence of the position difference on the correction function.

7. The path planning method based on the improved artificial potential field method according to claim 6, characterized in that: The road boundary potential field is obtained in S202 specifically as follows: Define the road boundary potential field U edge : Among them, U max_edge is the maximum value of the boundary potential field, k edge is the attenuation coefficient of the boundary potential field, d edge1 is the distance from the current point to the upper boundary, d edge2 The distance from the current point to the lower boundary.

8. The path planning method based on the improved artificial potential field method according to claim 7, characterized in that: The lane line potential field is obtained in S202 specifically as follows: Define the lane line potential field U lane for: Among them, U max_lane is the maximum value of the lane line potential field, k lane is the attenuation coefficient of the lane line potential field, d lane is the distance from the current point to the lane line.

9. The path planning method based on the improved artificial potential field method according to claim 8, characterized in that: The S3 is specifically: For the gravitational potential field U defined by S2 att , the partial derivative of x is: The partial derivative of y is: Among them, η att is the intensity coefficient of the gravitational potential field, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x goal is the horizontal coordinate of the target point, y goal is the ordinate of the target point; i.e. gravity The expression is: For the boundary potential field U defined by S2 edge , the partial derivative with respect to x is: The partial derivative of y is: Among them, U max_edg is the maximum value of the boundary potential field, k edge is the attenuation coefficient of the boundary potential field, d edge1 is the distance from the current point to the upper boundary, d edge2 is the distance from the current point to the lower boundary, i.e. the boundary force The expression is: For the lane line potential field U defined by S2 lane , the partial derivative with respect to x is: The partial derivative of y is: Among them, U max_lane is the maximum value of the lane line potential field, k lane is the attenuation coefficient of the lane line potential field, d lane is the distance from the current point to the lane line, i.e., lane force The expression is: For the static risk potential field U defined by S2 sta ,set up The partial derivative with respect to x is: The partial derivative of y is: Where A is the field strength coefficient, ω x is the lateral weight factor, ω y is the longitudinal weight factor, x is the current horizontal coordinate of the main vehicle, y is the current vertical coordinate of the main vehicle, and x obs is the horizontal coordinate of the obstacle, y obs is the ordinate of the obstacle, σ x is the standard deviation of the horizontal expansion, σ y is the standard deviation of the vertical expansion, β is the nonlinear factor, and η is the correction parameter, that is, the static risk factor The expression is: For the dynamic risk field U defined by S2 dyn ,set up The partial derivative with respect to x is: The partial derivative of y is: in, Among them, x obs (i) is the horizontal coordinate of the ith obstacle, y obs (i) is the ordinate of the ith obstacle, α is the parameter used to adjust the impact of the obstacle, and R v is the relative speed sign factor, which indicates the relationship between the obstacle speed and the vehicle speed. γ is the speed difference weight factor, which is used to adjust the impact of the speed difference on the risk field. obs (i) is the length of the i-th obstacle, v obs (i) is the speed of the ith obstacle, v self is the speed of the main vehicle, σ v is the standard deviation of the speed difference, and its formula is as follows: σ v =k v |v obs (i)-v self | Among them, k v is the speed difference proportional coefficient, which indicates the influence of speed difference on standard deviation, i.e. dynamic risk The expression is: The resultant force on the main vehicle The gravitational force on the main vehicle Boundary Force Lane Force and repulsion and Add together, the expression is: 。

Images