Unmanned vehicle formation obstacle avoidance control method based on vector field histogram and artificial potential energy method

Abstract

The invention provides an unmanned vehicle formation obstacle avoidance control method based on a vector field histogram and an artificial potential energy method, and the method comprises the following steps: determining the operation information of an unmanned vehicle formation, transmitting the operation information to a pilot vehicle of the unmanned vehicle formation, and enabling the pilot vehicle of the unmanned vehicle formation to lead the unmanned vehicle formation to operate according to the operation information; pilot vehicles of the unmanned vehicle formation detect obstacles in real time during operation, when the pilot vehicles detect the obstacles, the pilot vehicles lead the unmanned vehicle formation to avoid the obstacles by adopting a vector histogram method, and the pilot vehicles lead the unmanned vehicle formation to operate by adopting an artificial potential energy field method so as to avoid collision among the unmanned vehicles; after the unmanned vehicle formation obstacle avoidance is completed, leading the unmanned vehicle formation to recover the queue by the pilot vehicle by adopting a distance constraint algorithm; wherein the pilot vehicle adopts a sliding mode control algorithm to update the running speed and the steering angular speed of the unmanned vehicle formation during the running of the unmanned vehicles, the obstacle avoidance process of the unmanned vehicle formation and the formation recovery process.

Description

Technical Field

[0001] The present invention relates to an unmanned vehicle obstacle avoidance control method, and in particular to an unmanned vehicle formation obstacle avoidance control method based on a vector field histogram and an artificial potential energy method. Background Art

[0002] Unmanned vehicles (such as AGV, the full name is automatic guided vehicle) are widely used in industrial environments, such as cargo handling in warehousing, road garbage cleaning, etc.

[0003] Unmanned vehicles usually operate in a convoy. The control center sends control information to the lead vehicle, and the following vehicles communicate with the lead vehicle. The following vehicles receive control instructions from the lead vehicle, which is used to explore the status of obstacles in front of the convoy and perform corresponding obstacle avoidance.

[0004] Although the lead vehicle can avoid obstacles in the prior art, it is easy to cause collisions between unmanned vehicles during the obstacle avoidance process, and the convoy cannot keep moving normally. This is because the unmanned vehicle formation is not always in a linear formation. For example, the linear formation is arranged in a vertical line (the vertical direction refers to the forward direction of the unmanned vehicle), while the actual formation is that the control center sends the formation command to the lead vehicle, and the lead vehicle sends the command to each following vehicle. The following vehicles may form a triangular queue or a diamond queue, etc., which will cause collisions between unmanned vehicles during the obstacle avoidance process. Moreover, it is difficult for the convoy to accurately operate according to the expected control information (i.e., the control information given by the control center) during operation, which causes the final operation trajectory of the unmanned vehicle formation to deviate.

[0005] Therefore, in order to solve the above technical problems, it is urgent to propose a new technical means. Summary of the invention

[0006] In view of this, the purpose of the present invention is to provide an unmanned vehicle formation obstacle avoidance control method based on vector field histogram and artificial potential energy method, which can accurately identify obstacles and perform obstacle avoidance operations during the operation of the unmanned vehicle formation, and can also maintain the original formation sequence after the obstacle avoidance is completed, thereby ensuring the orderliness of the unmanned vehicle formation, and can correct the posture error according to the real-time information of the vehicle during the operation, ensuring that the unmanned vehicle formation can reach the specified position according to the established operation trajectory, and ensuring that the task can be successfully completed.

[0007] The present invention provides an unmanned vehicle formation obstacle avoidance control method based on a vector field histogram and an artificial potential energy method, comprising the following steps:

[0008] S1. Determine the operation information of the unmanned vehicle formation, and send the operation information to the pilot vehicle of the unmanned vehicle formation, and the pilot vehicle of the unmanned vehicle formation leads the unmanned vehicle formation to operate according to the operation information; wherein the operation information includes the operation trajectory, the operation speed and the steering angular velocity;

[0009] S2. The leader vehicle of the unmanned vehicle formation detects obstacles in real time during operation. When the leader vehicle detects an obstacle, it uses the vector histogram method to lead the unmanned vehicle formation to avoid the obstacle, and the leader vehicle uses the artificial potential field method to lead the unmanned vehicle formation to avoid collisions between unmanned vehicles;

[0010] S3. After the unmanned vehicle formation has completed obstacle avoidance, the leading vehicle uses the distance constraint algorithm to lead the unmanned vehicle formation to restore the queue;

[0011] Among them: when the unmanned vehicle is in operation, when the unmanned vehicle formation is avoiding obstacles, and when the formation is restored, the leading vehicle uses a sliding film control algorithm to update the running speed and steering angular velocity of the unmanned vehicle formation.

[0012] Furthermore, when the unmanned vehicle is in operation, the unmanned vehicle formation is in the process of avoiding obstacles, and the unmanned vehicle formation is in the process of restoring the formation, the pilot vehicle uses the sliding film control algorithm to update the running speed and steering angular velocity of the unmanned vehicle formation, including:

[0013] Project the actual coordinates C(x,y) of the center of mass of the unmanned vehicle in the global coordinate system XOY to the local coordinate system X'CY' to obtain the error e on the X' axis x , Error on the Y' axis e y And the steering angle error e θ :

[0014]

[0015] Where: x is the horizontal coordinate of the center of mass of the unmanned vehicle in the global coordinate system XOY, x r is the expected horizontal coordinate of the unmanned vehicle, y is the vertical axis label of the center of mass of the unmanned vehicle in the global coordinate system XOY, and y r is the desired ordinate of the unmanned vehicle, θ is the actual steering angle of the unmanned vehicle, and θ r is the expected steering angle of the autonomous vehicle;

[0016] Based on formula (1), the pose error matrix is ​​constructed:

[0017] ω r is the expected angular acceleration of the unmanned vehicle; v r is the expected running speed of the unmanned vehicle;

[0018] Construct a sliding membrane controller, the sliding membrane surface is:

[0019]

[0020] Introduce a constant velocity approaching law to the synovial surface:

[0021]

[0022] Where: k1 and k2 are the approach speeds of the synovial switching functions of functions s1 and s2, respectively, and both k1 and k2 are greater than zero;

[0023] Based on equations (2)-(4), the control law of the unmanned vehicle tracking sliding film controller is constructed:

[0024]

[0025] in:

[0026] The running speed v' and steering angular velocity ω' obtained by the control law of the unmanned vehicle tracking sliding film controller control the unmanned vehicle formation operation, where ω r is the desired steering angular velocity.

[0027] Furthermore, the initial running speed v of the unmanned vehicle formation and the initial steering angular velocity ω of the unmanned vehicle formation are determined by the following method:

[0028]

[0029] Where: r is the radius of the driving wheel of the unmanned vehicle, ω L and ω R are the angular velocities of the equivalent left and right wheels of the unmanned vehicle, respectively; L is the distance between the left and right symmetrical driving wheels of the unmanned vehicle;

[0030] Furthermore, the pilot vehicle uses the artificial potential field method to lead the unmanned vehicle formation operation to avoid collisions between unmanned vehicles. Specifically, the following are included:

[0031] Define the range of the repulsive field of the unmanned vehicle r out ;

[0032] Calculate the distance between unmanned vehicle i and unmanned vehicle j:

[0033] ||d ij ||=||d i -d j ||; where: d i is the distance between the unmanned vehicle i and the origin in the global coordinate system, d j is the distance between the unmanned vehicle j and the origin in the global coordinate system;

[0034] Construct the repulsion function Λ(||d ij ||):

[0035]

[0036] Where: r in is the rigid body radius of the unmanned vehicle, k is the attenuation coefficient, a r Indicates the magnitude of the repulsive force, represents the unit vector from unmanned vehicle i to unmanned vehicle j, represents the repulsive force vector acting on unmanned vehicle i;

[0037] Defining the desired vector of the autonomous vehicle And calculate the resultant force vector of unmanned vehicle i

[0038] The unmanned vehicle i follows the resultant force vector sports.

[0039] Furthermore, the expected steering angular velocity ω of the unmanned vehicle is determined by the following method: r :

[0040] The resultant force vector of the unmanned vehicle i Decompose into the X and Y directions of the global coordinate system to form the X component F x and Y component F y ;

[0041] Calculate the desired direction angle θ of the unmanned vehicle total :

[0042] θ total =atan2(F y ,F x );

[0043] Calculate the desired steering angular velocity ω r :

[0044] ω r =k w *(θ total -θ current );

[0045] Where: k w represents the proportionality constant, θ current Indicates the current direction angle of the unmanned vehicle.

[0046] Furthermore, the expected steering angular velocity v of the unmanned vehicle is determined by the following method: r :

[0047] Where: k o represents the proportionality constant.

[0048] The beneficial effects of the present invention are as follows: it can accurately identify obstacles and perform obstacle avoidance operations during the operation of an unmanned vehicle formation, and after the obstacle avoidance is completed, it can also maintain the original formation sequence, thereby ensuring the orderliness of the unmanned vehicle formation, and during the operation process, it can correct the posture error according to the real-time information of the vehicle, ensuring that the unmanned vehicle formation can reach the designated position according to the established operation trajectory, and ensuring that the task can be successfully completed. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0050] Figure 1 It is a flow chart of the present invention.

[0051] Figure 2 Schematic diagram of the collision avoidance area of ​​the artificial potential field algorithm. DETAILED DESCRIPTION

[0052] The present invention is further described in detail below:

[0053] The present invention provides an unmanned vehicle formation obstacle avoidance control method based on a vector field histogram and an artificial potential energy method, comprising the following steps:

[0054] S1. Determine the operation information of the unmanned vehicle formation, and send the operation information to the pilot vehicle of the unmanned vehicle formation, and the pilot vehicle of the unmanned vehicle formation leads the unmanned vehicle formation to operate according to the operation information; wherein the operation information includes the operation trajectory, the operation speed and the steering angular velocity;

[0055] S2. The leader vehicle of the unmanned vehicle formation detects obstacles in real time during operation. When the leader vehicle detects an obstacle, it uses the vector histogram method to lead the unmanned vehicle formation to avoid the obstacle, and the leader vehicle uses the artificial potential field method to lead the unmanned vehicle formation to avoid collisions between unmanned vehicles;

[0056] S3. After the unmanned vehicle formation has completed obstacle avoidance, the leading vehicle uses the distance constraint algorithm to lead the unmanned vehicle formation to restore the queue;

[0057] Among them: the leading vehicle adopts the sliding film control algorithm to update the running speed and steering angular velocity of the unmanned vehicle formation when the unmanned vehicle is in operation, the unmanned vehicle formation is in the process of avoiding obstacles, and the process of restoring the formation; through the above method, obstacles can be accurately identified and obstacle avoidance operations can be performed during the operation of the unmanned vehicle formation, and the original formation sequence can be maintained after the obstacle avoidance is completed, thereby ensuring the orderliness of the unmanned vehicle formation, and during the operation, the posture error can be corrected according to the real-time information of the vehicle, ensuring that the unmanned vehicle formation can reach the designated position according to the established operation trajectory, and ensuring that the task can be successfully completed.

[0058] In this embodiment, the pilot vehicle uses the sliding film control algorithm to update the running speed and steering angular velocity of the unmanned vehicle formation during the operation of the unmanned vehicle, the obstacle avoidance process of the unmanned vehicle formation, and the restoration process of the formation, specifically including:

[0059] Project the actual coordinates C(x,y) of the center of mass of the unmanned vehicle in the global coordinate system XOY to the local coordinate system X'CY' to obtain the error e on the X' axis x , Error on the Y' axis e y And the steering angle error e θ :

[0060]

[0061] Where: x is the horizontal coordinate of the center of mass of the unmanned vehicle in the global coordinate system XOY, x r is the expected horizontal coordinate of the unmanned vehicle, y is the vertical axis label of the center of mass of the unmanned vehicle in the global coordinate system XOY, and y r is the desired ordinate of the unmanned vehicle, θ is the actual steering angle of the unmanned vehicle, and θ r is the expected steering angle of the autonomous vehicle;

[0062] Based on formula (1), the pose error matrix is ​​constructed:

[0063] ω r is the expected angular acceleration of the unmanned vehicle; v r is the expected running speed of the unmanned vehicle;

[0064] Construct a sliding membrane controller, the sliding membrane surface is:

[0065]

[0066] Introduce a constant velocity approaching law to the synovial surface:

[0067]

[0068] Where: k1 and k2 are the approach speeds of the synovial switching functions of functions s1 and s2, respectively, and both k1 and k2 are greater than zero;

[0069] Based on equations (2)-(4), the control law of the unmanned vehicle tracking sliding film controller is constructed:

[0070]

[0071] in:

[0072] The running speed v' and steering angular velocity ω' obtained by the control law of the unmanned vehicle tracking sliding film controller control the unmanned vehicle formation operation, where ω rBy using the above method, the running speed and the steering angular velocity of the unmanned vehicle fleet can be updated and corrected in real time, thereby ensuring the smoothness of the unmanned vehicle during operation and facilitating the completion of the task.

[0073] Among them, the initial running speed v of the unmanned vehicle formation and the initial steering angular velocity ω of the unmanned vehicle formation are determined by the following method:

[0074]

[0075] Where: r is the radius of the driving wheel of the unmanned vehicle, ω L and ω R are the angular velocities of the equivalent left wheel and the equivalent right wheel of the unmanned vehicle, and L is the distance between the left and right symmetrical driving wheels of the unmanned vehicle.

[0076] In this embodiment, the pilot vehicle uses the artificial potential field method to lead the unmanned vehicle formation to avoid collisions between unmanned vehicles, specifically including:

[0077] Define the range of the repulsive field of the unmanned vehicle r out ;

[0078] Calculate the distance between unmanned vehicle i and unmanned vehicle j:

[0079] ||d ij ||=||d i -d j ||; where: d i is the distance between the unmanned vehicle i and the origin in the global coordinate system, d j is the distance between the unmanned vehicle j and the origin in the global coordinate system;

[0080] Construct the repulsion function Λ(||d ij ||):

[0081]

[0082] Where: r in is the rigid body radius of the unmanned vehicle, k is the attenuation coefficient, a r Indicates the magnitude of the repulsive force, represents the unit vector from unmanned vehicle i to unmanned vehicle j, represents the repulsive force vector acting on unmanned vehicle i;

[0083] Defining the desired vector of the autonomous vehicle And calculate the resultant force vector of the unmanned vehicle i

[0084] The unmanned vehicle i follows the resultant force vector Movement. The direction of the resultant force vector determines the direction of the unmanned vehicle, and its magnitude determines the speed of the vehicle. Through the above process, it is possible to effectively prevent collisions between unmanned vehicles during the operation and obstacle avoidance of the unmanned vehicle formation.

[0085] In this embodiment, the desired steering angular velocity ω of the unmanned vehicle is determined by the following method: r :

[0086] The resultant force vector of the unmanned vehicle i Decompose into the X and Y directions of the global coordinate system to form the X component F x and Y component F y ;

[0087] Calculate the desired direction angle θ of the unmanned vehicle total :

[0088] θ total =atan2(F y ,F x );

[0089] Calculate the desired steering angular velocity ω r :

[0090] ω r =k w *(θ total -θ current );

[0091] Where: k w represents the proportionality constant, θ current Indicates the current direction angle of the unmanned vehicle.

[0092] Determine the desired steering angular velocity v of the unmanned vehicle by the following method: r :

[0093] Where: k o represents the proportionality constant.

[0094] The desired steering angle θ r It is determined by the following formula:

[0095] Where: (x cur ,y cur ) represents the current coordinates of the unmanned vehicle, (x tar ,y tar) is the expected position coordinate of the unmanned vehicle at the current moment. Through the above method, the expected steering angular velocity and expected running speed are determined by the artificial potential field method, and then the two parameters are combined with the sliding film surface control algorithm to determine the final running speed v' and steering angular velocity ω', thereby effectively ensuring the smoothness of the operation of the human-vehicle formation and the consistency with the expected trajectory. When an obstacle occurs to the pilot vehicle during operation, the vector histogram algorithm is activated. The vector histogram algorithm is an existing technology and is briefly described below:

[0096] 1) Burn a laser radar program into the AGV to scan the information of dynamic obstacles around the AGV. According to the radar resolution f, the 360-degree scanning area is divided into n small sectors (such as ), and number the sectors from 0 to 360 degrees (such as Q i (i <n,i∈N + ), i represents the number of the small sector). For the obstacle information detected by the lidar, calculate the Euclidean distance from the AGV to the obstacle and angle α

[0097] Then, according to the angle between the obstacle and the laser radar, the obstacle can be divided into corresponding small sectors (such as round means taking the nearest integer), so that all obstacles can be divided into corresponding sectors. For environments with dense obstacles, there may be multiple obstacles in the same sector. The closest obstacle is taken. As the obstacle distance of the sector. Then the obstacle distance of each sector can be expressed as D i (i<n,i∈N + ).

[0098] 2) Raster to vector

[0099] In the VFH algorithm, a two-dimensional grid map is established to represent the surrounding environment, so the working area can be divided into several two-dimensional grid spaces. Each grid has a probability value cv, which represents the credibility of the existence of obstacles in the grid space. The larger the cv value, the higher the credibility of the existence of obstacles in the grid space. The formula for the vectorization value of the grid map is:

[0100]

[0101] Among them, m i is the value of the grid vector, indicating the obstacle strength value of the i-th sector, m i The larger the value, the greater the danger of obstacles in the sector to the drone, that is, the closer the distance between the obstacle and the drone is. a is a constant and b is a control Weights for vectorized values.

[0102] 3) Feasibility sector analysis

[0103] By setting the threshold M, the credibility of safe passage of each sector is evaluated. The credibility of passage of each sector can be expressed as

[0104]

[0105] Among them, H i It represents the passability of the ith sector, 0 means it is safe and passable; 1 means it is dangerous and impassable.

[0106] 4) Obstacle avoidance direction selection

[0107] After analyzing the passability of all sectors, it is necessary to search for the best direction to avoid obstacles in the passable sectors. When performing the optimization calculation, first, the passable area should be distinguished into wide and narrow troughs. Assuming that the i-th sector and the j-th sector are both passable sectors, the passable sector between the i-th sector and the j-th sector is called a trough. If ji>s, the trough is called a wide trough, otherwise it is a narrow trough. Then, candidate directions are selected from the wide and narrow troughs. There are different standards for selecting wide and narrow troughs. The leftmost sector of the trough is denoted as k. l , the rightmost sector is denoted as k r , then the candidate direction selection criteria for the narrow trough are as follows:

[0108]

[0109] Among them, c s Represents candidate directions selected from narrow valleys.

[0110] The selection criteria for candidate directions of wide troughs are as follows:

[0111]

[0112] Through (5)(6), the candidate direction c is selected from the traversable sector s Finally, it is necessary to optimize these candidate directions through the cost function to select the obstacle avoidance direction. The cost function is shown as follows:

[0113] C(θ)=μ1|θ-θ goal |+μ2|θ-θ current |+μ3|θ-θ t-1 |

[0114] where |θ-θ goal | is the angle between the candidate direction and the direction of the AGV target; |θ-θ current| is the angle between the candidate direction and the current moving direction of the AGV; |θ-θ t-1 | is the angle between the candidate direction and the AGV's previous moving direction. μ i is the corresponding coefficient. The direction with the minimum cost value selected from C(θ) is the obstacle avoidance direction.

[0115] The distance constraint method adopts the existing method, which is briefly described below:

[0116] Since it is a leader-following mode, the formation needs to be restored. The goal of the formation is to keep the two AGV followers in a fixed relative position relative to the AGV leader to form a triangle. The position update law of the two AGV followers is designed as:

[0117] p i =p1+l i1 , i=2,3 (7)

[0118] Among them, p i is the coordinate of the AGV follower [x i ,y i ,0] T , p1 is the position coordinate of the AGV leader

[0119] [x1, y1, 0] T , l 21 =[-γ x d, -γ y d,0] T , l 31 =[-γ x d, γ y ,d,0] T .

[0120] So according to what we know above, the navigator only needs to move towards the target point at the speed and direction at which obstacle avoidance ends, and the follower is relative to the target coordinates (x target ,y target ), determined according to formula (7).

[0121] Then we adjust the speed of the follower to move towards the formation's target based on the error.

[0122] beneficial.

[0124] In order to restore the angular velocity of the formation, the angular velocity can be calculated based on the error direction and the current orientation of the agent. Calculate the formation target direction angle and compare it with the current orientation to get the angular velocity:

[0125] θ target =atan2(e y , e x ), w=kw (θ target -θ current ) (8)

[0126] Among them, k w is the angular velocity gain, e y , e x is the decomposition direction of the error vector.

[0127] In order to avoid too drastic changes in speed, the speed can be smoothed. Use the exponential moving average (EMA) to process the speed.

[0128] v smooth =αv current +(1-α)*v desired

[0129] w smooth =α*w current +(1-α)*w desired

[0130] Where α is a smoothing factor (usually between 0 and 1), v current , w current is the velocity and angular velocity after obstacle avoidance. desired , w desired It is the speed and angular velocity required to restore the formation, which is calculated by equation (8).

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. An obstacle avoidance control method for unmanned vehicle formation based on vector field histogram and artificial potential energy method, characterized by: The following steps are involved: S1. Determine the operation information of the unmanned vehicle formation, and send the operation information to the pilot vehicle of the unmanned vehicle formation, and the pilot vehicle of the unmanned vehicle formation leads the unmanned vehicle formation to operate according to the operation information; wherein the operation information includes the operation trajectory, the operation speed and the steering angular velocity; S2. The leader vehicle of the unmanned vehicle formation detects obstacles in real time during operation. When the leader vehicle detects an obstacle, it uses the vector histogram method to lead the unmanned vehicle formation to avoid the obstacle, and the leader vehicle uses the artificial potential field method to lead the unmanned vehicle formation to avoid collisions between unmanned vehicles; S3. After the unmanned vehicle formation has completed obstacle avoidance, the leading vehicle uses the distance constraint algorithm to lead the unmanned vehicle formation to restore the queue; Among them: when the unmanned vehicle is in operation, when the unmanned vehicle formation is avoiding obstacles, and when the formation is restored, the leading vehicle uses a sliding film control algorithm to update the running speed and steering angular velocity of the unmanned vehicle formation.

2. The unmanned vehicle formation obstacle avoidance control method based on vector field histogram and artificial potential energy method according to claim 1, characterized in that: When the unmanned vehicle is in operation, the unmanned vehicle formation is in the process of avoiding obstacles, and the process of restoring the formation, the pilot vehicle uses the sliding film control algorithm to update the running speed and steering angular velocity of the unmanned vehicle formation. The specific details include: Project the actual coordinates C(x,y) of the center of mass of the unmanned vehicle in the global coordinate system XOY to the local coordinate system X'CY' to obtain the error e on the X' axis x , Error on the Y' axis e y And the steering angle error e θ : Where: x is the horizontal coordinate of the center of mass of the unmanned vehicle in the global coordinate system XOY, x r is the expected horizontal coordinate of the unmanned vehicle, y is the vertical axis label of the center of mass of the unmanned vehicle in the global coordinate system XOY, and y r is the desired ordinate of the unmanned vehicle, θ is the actual steering angle of the unmanned vehicle, and θ r is the expected steering angle of the autonomous vehicle; Based on formula (1), the pose error matrix is ​​constructed: ω r is the expected angular acceleration of the unmanned vehicle; v r is the expected running speed of the unmanned vehicle; Construct a sliding membrane controller, the sliding membrane surface is: Introduce a constant velocity approaching law for the synovial surface: Where: k1 and k2 are the approach speeds of the synovial switching functions of functions s1 and s2, respectively, and both k1 and k2 are greater than zero; Based on equations (2)-(4), the control law of the unmanned vehicle tracking sliding film controller is constructed: in: The running speed v' and steering angular velocity ω' obtained by the control law of the unmanned vehicle tracking sliding film controller control the unmanned vehicle formation operation, where ω r is the desired steering angular velocity.

3. The unmanned vehicle formation obstacle avoidance control method based on vector field histogram and artificial potential energy method according to claim 1, characterized in that: The initial running speed v of the unmanned vehicle formation and the initial steering angular velocity ω of the unmanned vehicle formation are determined by the following method: Where: r is the radius of the driving wheel of the unmanned vehicle, ω L and ω R are the angular velocities of the equivalent left wheel and the equivalent right wheel of the unmanned vehicle, and L is the distance between the left and right symmetrical driving wheels of the unmanned vehicle.

4. The unmanned vehicle formation obstacle avoidance control method based on vector field histogram and artificial potential energy method according to claim 2, characterized in that: The pilot vehicle uses the artificial potential field method to lead the unmanned vehicle formation to avoid collisions between unmanned vehicles. The specific methods include: Define the range of the repulsive field of the unmanned vehicle r out ; Calculate the distance between unmanned vehicle i and unmanned vehicle j: ||d ij ||=||d i -d j ||; where: d i is the distance between the unmanned vehicle i and the origin in the global coordinate system, d j is the distance between the unmanned vehicle j and the origin in the global coordinate system; Construct the repulsion function Λ(||d ij ||): Where: r in is the rigid body radius of the unmanned vehicle, k is the attenuation coefficient, a r Indicates the magnitude of the repulsive force, represents the unit vector from unmanned vehicle i to unmanned vehicle j, represents the repulsive force vector acting on unmanned vehicle i; Defining the desired vector of the autonomous vehicle And calculate the resultant force vector of unmanned vehicle i The unmanned vehicle i follows the resultant force vector sports.

5. The unmanned vehicle formation obstacle avoidance control method based on vector field histogram and artificial potential energy method according to claim 4, characterized in that: Determine the desired steering angular velocity ω of the unmanned vehicle by the following method: r : The resultant force vector of the unmanned vehicle i Decompose into the X and Y directions of the global coordinate system to form the X component F x and Y component F y ; Calculate the desired direction angle θ of the unmanned vehicle total : i total =atan2(F y ,F x ); Calculate the desired steering angular velocity ω r : oh r =k w *(θ total -θ current ); Where: k w represents the proportionality constant, θ current Indicates the current direction angle of the unmanned vehicle.

6. The unmanned vehicle formation obstacle avoidance control method based on vector field histogram and artificial potential energy method according to claim 5, characterized in that: The expected steering angular velocity v of the unmanned vehicle is determined by the following method: r : Where: k o represents the proportionality constant.

Images