A vehicle platooning control method based on markers

Abstract

The application discloses a kind of based on identifier's human-vehicle platoon driving control method, belong to platoon driving control field, this method includes determining the leader and follower of queue, and according to the formation of platoon determines the formation reference object of follower;To unmanned transport vehicle, the identifier of formation reference object is identified in real time, center point offset and human-vehicle distance information are obtained based on visual method;And center point offset and human-vehicle distance information are respectively input into virtual spring model and fuzzy controller, obtain the desired linear velocity and desired angular velocity of unmanned transport vehicle;The desired linear velocity and desired angular velocity of unmanned transport vehicle are input into the kinematic model of unmanned transport vehicle, and the desired rotational speed of the wheel of unmanned transport vehicle is obtained;Wheel rotation is controlled by motor controller.The application solves the problem that the computing resources are limited when the processor of unmanned transport vehicle deploys target recognition model, the unmanned transport vehicle is prone to jitter when following the target pedestrian movement, and the transportation efficiency of a single unmanned transport vehicle is low.

Description

Technical Field

[0001] This invention belongs to the field of platooning control, and particularly relates to a method for controlling platooning of people and vehicles based on markers. Background Technology

[0002] Currently, unmanned vehicle platooning methods can be categorized into lidar-based, active device-based, and vision-based methods. However, lidar-based methods are costly and face challenges in target recognition and locking; active device-based methods require users to wear protective gear, limiting the detection range. Vision-based methods are less expensive and offer accurate target tracking, making them the current mainstream solution. However, even the most advanced vision-based methods still have some drawbacks.

[0003] Unmanned transport vehicles (UAVs) are based on autonomous mobile platforms with the addition of platooning and following capabilities. Human-vehicle platooning, as a new trend in the development of UAVs, offers advantages such as good maneuverability, high carrying capacity, and reduced manual labor intensity. Current research on human-vehicle platooning for UAVs primarily utilizes intelligent algorithms to identify human feature points, thereby determining the target to follow and guiding the UAV towards that target.

[0004] Currently, some scholars have designed a pedestrian following system based on depth camera vision, which obtains the distance d between the target pedestrian and the following vehicle, as well as the deviation angle between the virtual spring and the following vehicle. They have proposed the KCF algorithm and designed a motion control strategy for the following model.

[0005] This algorithm can effectively track target pedestrians in real time, obtain the image position of the target pedestrians, calculate the specific position of the target pedestrians relative to the robot based on the installation position of the depth camera on the robot and the depth flow information of the depth camera, and then achieve anti-interference and anti-loss pedestrian tracking through the tracking motion control strategy of virtual spring model.

[0006] The disadvantages are:

[0007] ① The unmanned transport vehicle uses an embedded edge data processing platform, which has limited computing resources and is costly to deploy large target recognition models;

[0008] ②The angular velocity and linear velocity of the unmanned transport vehicle are coupled together, and the random fluctuations caused by the gait of the target pedestrian will cause the unmanned transport vehicle to shake when it follows the movement.

[0009] ③ The transportation efficiency of a single unmanned transport vehicle is relatively low. Summary of the Invention

[0010] To address the aforementioned shortcomings in existing technologies, this invention provides a vehicle-human platooning driving control method based on markers, which solves the problems of limited computing resources when deploying target recognition models in unmanned transport vehicle processors, easy shaking when unmanned transport vehicles follow target pedestrians, and low transport efficiency of individual unmanned transport vehicles.

[0011] To achieve the aforementioned objectives, the present invention employs the following technical solution: a method for controlling platooning of vehicles and pedestrians based on markers, comprising:

[0012] Establish the formation and affix markers to the pedestrians and vehicles participating in the formation; determine the navigator and followers of the formation, and determine the formation reference objects for the followers based on the formation; the markers are double concentric circles;

[0013] For unmanned transport vehicles acting as followers, the lightweight YOLOv5s target recognition algorithm is used to identify the markers of the formation reference objects in real time, and the center point offset M is obtained based on a visual method. o and distance information from the formation reference object l d ; and offset the center point by M o and distance information from the formation reference object l d The desired linear velocity and desired angular velocity of the unmanned transport vehicle are obtained by inputting the virtual spring model and the fuzzy controller respectively. The desired linear velocity and desired angular velocity of the unmanned transport vehicle are then input into the kinematic model of the unmanned transport vehicle to obtain the desired rotational speed of the unmanned transport vehicle wheels. Based on the desired rotational speed of the unmanned transport vehicle wheels, the rotation of the wheels is controlled by the motor controller.

[0014] Furthermore, the expression for the desired linear velocity of the unmanned transport vehicle is:

[0015]

[0016] Among them, v t+1 v is the linear velocity at time t+1. t Let F be the linear velocity at time t; s be the virtual spring tension; F total The net force acting on the unmanned transport vehicle; m is the mass of the unmanned transport vehicle; d is the differential symbol; t is time; l d For distance information relative to the formation reference object; s1 For safe distance; l o k is the spring relaxation distance. f α is the weighted dynamic damping coefficient; k0 is the elastic coefficient of the virtual spring; α and β are both weighting coefficients.

[0017] Furthermore, the resultant force F acting on the unmanned transport vehicle total The expression is:

[0018]

[0019] F rh =k f |l rh -l s1

[0020]

[0021] F pf =k f |l pf -l s1

[0022]

[0023] Among them, F pri The net force on the first follower; F fol The net force experienced by ordinary followers; F rh F represents the spring traction force; i represents the obstacle number; n represents the total number of obstacles; pd For when l pd ≤l d1 The elastic repulsive force generated when entering the area of ​​influence of an obstacle; pd The distance between the priority follower and the obstacle when the follower is within the obstacle's influence range; d1 The range of influence of the obstacle; rh F represents the distance between the priority follower and the formation reference object. pf F is the spring-like traction force between ordinary followers and the formation reference object; fd For when l fd ≤l d1 The elastic repulsive force generated when entering the area of ​​influence of an obstacle; fd is the distance between a normal follower and the obstacle when the follower is within the obstacle's influence range; j is the number of the adjacent normal follower; m is the total number of adjacent normal followers; F ff For when l ff ≤l d2 The elastic repulsive force generated when entering the range of an adjacent ordinary follower; pf The distance between a regular follower and the formation reference object; l ff The distance between a regular follower and its neighboring regular followers when they are within the influence range; l d2 The influence range of adjacent ordinary followers; the priority follower is the unmanned transport vehicle whose formation reference object is a pedestrian; the ordinary follower is the unmanned transport vehicle whose formation reference object is a vehicle.

[0024] Furthermore, the fuzzy controller is a two-dimensional fuzzy controller with dual inputs and a single output, and the input of the controller is the center point offset M.o and distance information from the formation reference object l d The output is the desired angular velocity w. c .

[0025] Furthermore, the center point offset M in the fuzzy controller o The corresponding fuzzy set of variables is set as follows: based on the range of pixel values ​​in the horizontal coordinate [0.m] m The partition is divided into 7 fuzzy sets, including XS fuzzy set [0, m1), VS fuzzy set [m1, m2), S fuzzy set [m2, m3), M fuzzy set [m3, m4), L fuzzy set [m4, m5), VL fuzzy set [m5, m6) and XL fuzzy set [m6, m1, m2, m3, m4), L fuzzy set [m4, m5), VL fuzzy set [m5, m6), and XL fuzzy set [m6, m1, m2, m3, m4, m5, m6, m2, m3, m4, m5, m6, m4, m5, m6, m4, m5, m6, m6, m4, m5, m6, m6, m6, m7, m8, m9, m1, m1, m1, m2, m3, m1, m2 m m1, m2, m3, m4, m5, and m6 are all center point offsets M. o The corresponding variable fuzzy set partitioning threshold;

[0026] The distance information l between the fuzzy controller and the formation reference object d The corresponding fuzzy set of variables is set as follows: based on the distance between people and vehicles within the range [0, l m The five fuzzy sets are divided into five equal parts, including the VS fuzzy set [0, l1), S fuzzy set [l1, l2), M fuzzy set [l2, l3), L fuzzy set [l3, l4), and VL fuzzy set [l4, l1], which are sorted by their corresponding numerical values ​​from smallest to largest. m l1, l2, l3, and l4 are all distance information relative to the formation reference object. d The corresponding variable fuzzy set partitioning threshold;

[0027] The desired angular velocity w in the fuzzy controller c The corresponding fuzzy set of variables is set as follows: based on the range of angular velocity values ​​[-w m ,w m The 9 fuzzy sets are divided equally, including the NL fuzzy set [-w] whose corresponding values ​​are sorted from smallest to largest. m ,w1), NB fuzzy set [w1,w2), NM fuzzy set [w2,w3), NS fuzzy set [w3,0], ZO fuzzy set [0], PS fuzzy set (0,w4], PM fuzzy set (w4,w5], PB fuzzy set (w5,w6] and PL fuzzy set (w6,w4,w5], m w1, w2, w3, w4, w5, and w6 are all desired angular velocities w c The corresponding threshold for fuzzy set partitioning of variables.

[0028] Furthermore, the center point offset M oThe corresponding variable fuzzy sets are XS fuzzy set [0, m1), S fuzzy set [m2, m3), M fuzzy set [m3, m4), L fuzzy set [m4, m5) and XL fuzzy set [m6, m7]. m and the expected angular velocity w c The corresponding variable fuzzy set NL fuzzy set [-w m The NS fuzzy set [w3,0], the ZO fuzzy set after interval expansion [0-δ, 0+δ], the PS fuzzy set (0,w4], and the PL fuzzy set (w6,w1), are all fuzzy sets. m The membership function of ] is a bell-shaped membership function; δ is the magnified difference; the center point offset M o The corresponding variable fuzzy sets VS fuzzy set [m1, m2) and VL fuzzy set [m5, m6), and the distance information l with the formation reference object. d The corresponding variable fuzzy sets are VS fuzzy set [0, l1), S fuzzy set [l1, l2), M fuzzy set [l2, l3), L fuzzy set [l3, l4) and VL fuzzy set [l4, l1]. m ] and the desired angular velocity w c The membership functions of the corresponding fuzzy sets NB (w1, w2), NM (w2, w3), PM (w4, w5), and PB (w5, w6) are triangular membership functions.

[0029] Furthermore, the fuzzy control rule of the fuzzy controller is as follows:

[0030]

[0031] Among them, H(w) c () is based on the center point offset M o and distance information from the formation reference object l d The expected angular velocity w obtained from the corresponding fuzzy set of variables c The corresponding fuzzy set of variables, the distance information l between the behavior in the matrix and the formation reference object. d The corresponding variable fuzzy sets are, in order, VL fuzzy set, L fuzzy set, M fuzzy set, S fuzzy set, and VS fuzzy set, with the columns in the matrix representing the center point offset M. o The corresponding variable fuzzy sets are, in order, XS fuzzy set, VS fuzzy set, S fuzzy set, M fuzzy set, L fuzzy set, VL fuzzy set, and XL fuzzy set.

[0032] Furthermore, the fuzzy controller determines the center point offset M based on the Mandani fuzzy inference method. o Distance information relative to the formation reference object d and expected angular velocity w c The ternary fuzzy relation R.

[0033] Furthermore, the center point offset M o and distance information from the formation reference object l d The desired angular velocity of the unmanned transport vehicle is obtained by inputting the fuzzy controller, specifically:

[0034] Based on the center point offset M o and distance information from the formation reference object l d Using a fuzzy controller, the membership value C1 obtained through fuzzy inference is obtained; and the centroid method is used to defuzzify the membership value C1 to obtain the angular velocity of the unmanned transport vehicle.

[0035] Determine whether the unmanned transport vehicle is in a non-sensitive area. If so, set the expected angular velocity of the unmanned transport vehicle to 0; otherwise, set the expected angular velocity of the unmanned transport vehicle to the angular velocity of the unmanned transport vehicle.

[0036] Furthermore, the expression for the insensitive region is:

[0037]

[0038] Among them, I ss This is an insensitive region; i1, i2, and i3 are all constants; l d This represents the distance information to the formation reference object; e is a natural constant.

[0039] The beneficial effects of this invention are as follows: This invention decouples the linear velocity and angular velocity of unmanned transport vehicles and controls them separately. By improving the virtual spring model and designing the linear velocity controller with a weighted dynamic damping coefficient, and by setting an adaptive "insensitive region" and designing the angular velocity controller with fuzzy control, the problem of vibration that easily occurs in unmanned transport vehicles is effectively solved. The human-vehicle formation is extended to human-vehicle-vehicle formation, which can be flexibly configured according to different needs, thereby improving the transportation efficiency of unmanned transport vehicles. Attached Figure Description

[0040] Figure 1 This is a flowchart of the method of the present invention.

[0041] Figure 2 This is a schematic diagram of a virtual spring model in an embodiment of the present invention.

[0042] Figure 3 This is a schematic diagram illustrating the relationship between the weighted dynamic damping coefficient and elastic force in an embodiment of the present invention.

[0043] Figure 4 This is a schematic diagram of the input and output membership functions of the fuzzy controller in an embodiment of the present invention.

[0044] Figure 5 This is a schematic diagram of the fuzzy controller results in an embodiment of the present invention.

[0045] Figure 6 This is a schematic diagram of the marker in an embodiment of the present invention.

[0046] Figure 7 This is a schematic diagram of the virtual spring topology model in an embodiment of the present invention. Detailed Implementation

[0047] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0048] like Figure 1 As shown, in one embodiment of the present invention, a vehicle-pedestrian platooning control method based on identifiers includes:

[0049] Establish the formation and affix markers to the pedestrians and vehicles participating in the formation; determine the navigator and followers of the formation, and determine the formation reference objects for the followers based on the formation; the markers are double concentric circles;

[0050] For unmanned transport vehicles acting as followers, the lightweight YOLOv5s target recognition algorithm is used to identify the markers of the formation reference objects in real time, and the center point offset M is obtained based on a visual method. o and distance information from the formation reference object l d ; and offset the center point by M o and distance information from the formation reference object l d The desired linear velocity and desired angular velocity of the unmanned transport vehicle are obtained by inputting the virtual spring model and the fuzzy controller respectively. The desired linear velocity and desired angular velocity of the unmanned transport vehicle are then input into the kinematic model of the unmanned transport vehicle to obtain the desired rotational speed of the unmanned transport vehicle wheels. Based on the desired rotational speed of the unmanned transport vehicle wheels, the rotation of the wheels is controlled by the motor controller.

[0051] In this embodiment, the network structure upon which the lightweight YOLOv5s target recognition algorithm relies is specifically: the backbone network of the YOLOv5s network is replaced with a MobileNetV3 network. For example... Figure 6 The diagram shown is a schematic of two concentric circles.

[0052] like Figure 2 The diagram shows the mechanical properties of the virtual spring model containing the priority follower under compression, relaxation, and tension states.

[0053] The expression for the desired linear velocity of the unmanned transport vehicle is:

[0054]

[0055] Among them, v t+1 v is the linear velocity at time t+1. t Let F be the linear velocity at time t; s be the virtual spring tension; F total The net force acting on the unmanned transport vehicle; m is the mass of the unmanned transport vehicle; d is the differential symbol; t is time; l d For distance information relative to the formation reference object; s1 For safe distance; l o k is the spring relaxation distance. f α is the weighted dynamic damping coefficient; k0 is the elastic coefficient of the virtual spring; α and β are both weighting coefficients.

[0056] In this embodiment, Figure 3 This diagram illustrates the relationship between distance and elastic force when adjusting the weights in the formula for calculating the weighted dynamic damping coefficient.

[0057] The resultant force F acting on the unmanned transport vehicle total The expression is:

[0058]

[0059] F rh =k f |l rh -l s1

[0060]

[0061]

[0062] F pf =k f |l pf -l s1

[0063]

[0064] Among them, F pri The net force on the first follower; F fol The net force experienced by ordinary followers; F rh F represents the spring traction force; i represents the obstacle number; n represents the total number of obstacles; pd For when l pd ≤l d1 The elastic repulsive force generated when entering the area of ​​influence of an obstacle; pd The distance between the priority follower and the obstacle when the follower is within the obstacle's influence range; d1The range of influence of the obstacle; rh F represents the distance between the priority follower and the formation reference object. pf F is the spring-like traction force between ordinary followers and the formation reference object; fd For when l fd ≤l d1 The elastic repulsive force generated when entering the area of ​​influence of an obstacle; fd is the distance between a normal follower and the obstacle when the follower is within the obstacle's influence range; j is the number of the adjacent normal follower; m is the total number of adjacent normal followers; F ff For when l ff ≤l d2 The elastic repulsive force generated when entering the range of an adjacent ordinary follower; pf The distance between a regular follower and the formation reference object; l ff The distance between a regular follower and its neighboring regular followers when they are within the influence range; l d2 The influence range of adjacent ordinary followers; the priority follower is the unmanned transport vehicle whose formation reference object is a pedestrian; the ordinary follower is the unmanned transport vehicle whose formation reference object is a vehicle.

[0065] like Figure 7 The diagram shown is a schematic diagram of the virtual spring topology of the human-vehicle-vehicle system.

[0066] The fuzzy controller is a two-dimensional fuzzy controller with dual inputs and a single output. The input of the controller is the center point offset M. o and distance information from the formation reference object l d The output is the desired angular velocity w. c .

[0067] The center point offset M in the fuzzy controller o The corresponding fuzzy set of variables is set as follows: based on the range of pixel values ​​in the horizontal coordinate [0.m] m The partition is divided into 7 fuzzy sets, including XS fuzzy set [0, m1), VS fuzzy set [m1, m2), S fuzzy set [m2, m3), M fuzzy set [m3, m4), L fuzzy set [m4, m5), VL fuzzy set [m5, m6) and XL fuzzy set [m6, m1, m2, m3, m4), L fuzzy set [m4, m5), VL fuzzy set [m5, m6), and XL fuzzy set [m6, m1, m2, m3, m4, m5, m6, m2, m3, m4, m5, m6, m4, m5, m6, m4, m5, m6, m6, m4, m5, m6, m6, m6, m7, m8, m9, m1, m1, m1, m2, m3, m1, m2 m m1, m2, m3, m4, m5, and m6 are all center point offsets M. o The corresponding variable fuzzy set partitioning threshold;

[0068] The distance information l between the fuzzy controller and the formation reference object d The corresponding fuzzy set of variables is set as follows: based on the distance between people and vehicles within the range [0, l mThe five fuzzy sets are divided into five equal parts, including the VS fuzzy set [0, l1), S fuzzy set [l1, l2), M fuzzy set [l2, l3), L fuzzy set [l3, l4), and VL fuzzy set [l4, l1], which are sorted by their corresponding numerical values ​​from smallest to largest. m l1, l2, l3, and l4 are all distance information relative to the formation reference object. d The corresponding variable fuzzy set partitioning threshold;

[0069] The desired angular velocity w in the fuzzy controller c The corresponding fuzzy set of variables is set as follows: based on the range of angular velocity values ​​[-w m ,w m The 9 fuzzy sets are divided equally, including the NL fuzzy set [-w] whose corresponding values ​​are sorted from smallest to largest. m ,w1), NB fuzzy set [w1,w2), NM fuzzy set [w2,w3), NS fuzzy set [w3,0], ZO fuzzy set [0], PS fuzzy set (0,w4], PM fuzzy set (w4,w5], PB fuzzy set (w5,w6] and PL fuzzy set (w6,w4,w5], m w1, w2, w3, w4, w5, and w6 are all desired angular velocities w c The corresponding threshold for fuzzy set partitioning of variables.

[0070] The center point offset M o The corresponding variable fuzzy sets are XS fuzzy set [0, m1), S fuzzy set [m2, m3), M fuzzy set [m3, m4), L fuzzy set [m4, m5) and XL fuzzy set [m6, m7]. m and the expected angular velocity w c The corresponding variable fuzzy set NL fuzzy set [-w m The NS fuzzy set [w3,0], the ZO fuzzy set after interval expansion [0-δ, 0+δ], the PS fuzzy set (0,w4], and the PL fuzzy set (w6,w1), are all fuzzy sets. m The membership function of ] is a bell-shaped membership function; δ is the magnified difference; the center point offset M o The corresponding variable fuzzy sets VS fuzzy set [m1, m2) and VL fuzzy set [m5, m6), and the distance information l with the formation reference object. d The corresponding variable fuzzy sets are VS fuzzy set [0, l1), S fuzzy set [l1, l2), M fuzzy set [l2, l3), L fuzzy set [l3, l4) and VL fuzzy set [l4, l1]. m ] and the desired angular velocity w c The membership functions of the corresponding fuzzy sets NB (w1, w2), NM (w2, w3), PM (w4, w5), and PB (w5, w6) are triangular membership functions.

[0071] The fuzzy control rules of the fuzzy controller are as follows:

[0072]

[0073] Among them, H(w) c () is based on the center point offset M o and distance information from the formation reference object l d The expected angular velocity w obtained from the corresponding fuzzy set of variables c The corresponding fuzzy set of variables, the distance information l between the behavior in the matrix and the formation reference object. d The corresponding variable fuzzy sets are, in order, VL fuzzy set, L fuzzy set, M fuzzy set, S fuzzy set, and VS fuzzy set, with the columns in the matrix representing the center point offset M. o The corresponding variable fuzzy sets are, in order, XS fuzzy set, VS fuzzy set, S fuzzy set, M fuzzy set, L fuzzy set, VL fuzzy set, and XL fuzzy set.

[0074] The fuzzy controller is based on the Mandani fuzzy inference method to determine the center point offset M. o Distance information relative to the formation reference object d and expected angular velocity w c The ternary fuzzy relation R.

[0075] In this embodiment, taking pedestrian-vehicle platooning (pedestrian-priority follower) as an example, the fuzzy controller construction process is illustrated: using the marker center point offset M o Distance information between people and vehicles rh To control the angular velocity of the transport platform; center point offset M o This refers to the absolute value of the difference between the horizontal coordinate pixel (X-pixel) of the marker and the center point in the depth camera. For example, the D455 depth camera used in this invention has a range of [0, 1280] dp, with 640 dp as the center point. When the marker is within the range of [0, 640] dp, the angular velocity is negative, and the transport platform turns left; when the marker is within the range of [640, 1280] dp, the angular velocity is positive, and the transport platform turns right. When the distance between the person and the vehicle is constant, the offset is positively correlated with the angular velocity; when the offset is constant, the distance between the person and the vehicle is also positively correlated with the angular velocity.

[0076] A two-dimensional fuzzy controller with dual inputs and single outputs is used, with the center point offset M as the input. o Distance information between people and vehicles rh The output is angular velocity w c The specific design steps for a fuzzy controller are as follows:

[0077] ① Define the input and output fuzzy sets, with the horizontal coordinate pixel values ​​ranging from [0, 1280]dp, and set as 7 fuzzy sets, namely XS (minimal), VS (very small), S (small), M (medium), L (large), VL (very large), and XL (maximum); the distance between the person and the vehicle is l. rh The value range is [0, 10]m, and it is set into 5 fuzzy sets, namely VS (very small), S (small), M (medium), L (large), and VL (very large); angular velocity w c The value range is [-0.3, 0.3] rad / s, and it is set into 9 fuzzy sets, namely NL (negative maximum), NB (negative large), NM (negative medium), NS (negative small), ZO (zero), PS (positive small), PM (positive medium), PB (positive large), and PL (positive maximum).

[0078] ② Define the input and output membership functions, and the center point offset M. o The membership functions of XS (minimum), S (small), M (medium), L (large), and XL (maximum) are set to gbellmf (bell-shaped membership function), and the angular velocity w c The membership functions for NL (negative maxima), NS (negative minima), ZO (zero), PS (positive minima), and PL (positive maxima) are set to gbellmf (bell-shaped membership function), and the interval of ZO (zero) is expanded. The membership functions for the remaining fuzzy sets are set to trimf (triangular membership function), such as... Figure 4 and Figure 5 As shown.

[0079] ③ Establish fuzzy control rules:

[0080] If(M o is XS)and(l rh is VL)then(w c is NL);

[0081] If the center point offset M o It is XS (very small) and the distance between the person and the vehicle is l rh If VL (very large), then the angular velocity w c Take NL (negative maximum);

[0082] If(M o is VS)and(l rh is VL)then(w c is NB);

[0083] If the center point offset M o It is VS (very small) and the distance between people and vehicles is l rh If VL (very large), then the angular velocity w c Take NB (negative larger);

[0084] If(M o is S)and(l rh is VL)then(w c is NS);

[0085] If the center point offset M o It is S (small) and the distance between people and vehicles is l rh If VL (very large), then the angular velocity w c Take NS (negative smaller value);

[0086] ...

[0087] ④ Establish a fuzzy control table (as shown in Table 1).

[0088] Table 1

[0089]

[0090] ⑤ Fuzzy reasoning: In the designed fuzzy controller, the relationship of the fuzzy reasoning statement "If A and B then C" is (A∧B→C). Using the Mandani fuzzy reasoning method, the ternary fuzzy relation R is determined as follows:

[0091] R = (A × B) T1 ×C

[0092] In the formula, (A×B) T1 It consists of A×B, with dimensions m×n, where n and m are the number of elements in the universes of discourse of A and B, respectively; where A is the offset of the center point M. o The universe of discourse is set [XS,VS,S,M,L,VL,XL]. The membership value of a certain center point offset is determined by the membership function after multiple adjustments, and is [a1,a2,a3,a4,a5,a6,a7]. B is the distance between the person and the vehicle, l. rh The universe of discourse is set [VL,L,M,S,VS]. The membership value of a certain distance between a person and a vehicle is determined by the membership function after multiple adjustments, and is [b1,b2,b3,b4,b5]. C is the angular velocity w. c The universe of discourse is set as [NL, NB, NM, NS, ZO, PS, PM, PB, PL]. The membership values ​​corresponding to the angular velocities of the center point offset and the distance between the person and the vehicle are determined by the membership functions after multiple adjustments. The membership values ​​are [c1, c2, c3, c4, c5, c6, c7, c8, c9].

[0093] The formula for calculating (A×B) is as follows:

[0094]

[0095] In the formula, m(a1,b1) represents taking the smaller value of a1 and b1.

[0096] Expand (A×B) to (A×B) T1 The column vectors are shown below:

[0097] (A×B) T1 =[m(a1,b1)m(a1,b2)m(a1,b3)m(a1,b4)m(a1,b5)m(a2,b1)…m(a7,b5)] T

[0098] Therefore, the formula for calculating the ternary fuzzy relation R can be obtained as follows:

[0099]

[0100] In the formula, m(a1,b1,c1) represents taking the smaller value among a1,b1, andc1.

[0101] Based on the fuzzy inference rules and R, we can find C1 corresponding to given inputs A1 and B1 as follows:

[0102] C1 = (A1 × B1) T2 ×R

[0103] In the formula, A1 is the membership value [d1,d2,d3,d4,d5,d6,d7] corresponding to the actual measured center point offset; B1 is the membership value [e1,e2,e3,e4,e5] corresponding to the actual measured distance between people and vehicles; C1 is the membership value [f1,f2,f3,f4,f5,f6,f7,f8,f9] obtained by fuzzy inference after the actual measured center point offset and distance between people and vehicles; T2 is the row and column vector transformation.

[0104] The formula for (A1×B1) is as follows:

[0105]

[0106] In the formula, m(d1,e1) represents taking the smaller value of d1 and e1.

[0107] Expand (A1×B1) to (A1×B1) T2 The row vectors are shown below:

[0108] (A1×B1) T2 =[m(d1,e1)m(d1,e2)m(d1,e3)m(d1,e4)m(d1,e5)m(d2,e1)…m(d7,e5)]

[0109] Based on the actual inputs A1 and B1, C1, obtained through reasoning, is the membership value [f1, f2, f3, f4, f5, f6, f7, f8, f9]. Further defuzzification is needed to obtain the angular velocity w of the launch platform.c The precise value. The centroid method is often used for defuzzification, and the formula is as follows:

[0110]

[0111] In the formula, μ(w) is the membership function of the output; w is the specific value corresponding to the membership function of the output.

[0112] This yields the result based on the center point offset M. o Distance information between people and vehicles rh angular velocity w of the launch platform c .

[0113] Offset the center point by M o and distance information from the formation reference object l d The desired angular velocity of the unmanned transport vehicle is obtained by inputting the fuzzy controller, specifically:

[0114] Based on the center point offset M o and distance information from the formation reference object l d Using a fuzzy controller, the membership value C1 obtained through fuzzy inference is obtained; and the centroid method is used to defuzzify the membership value C1 to obtain the angular velocity of the unmanned transport vehicle.

[0115] Determine whether the unmanned transport vehicle is in a non-sensitive area. If so, set the expected angular velocity of the unmanned transport vehicle to 0; otherwise, set the expected angular velocity of the unmanned transport vehicle to the angular velocity of the unmanned transport vehicle.

[0116] The expression for the insensitive region is:

[0117]

[0118] Among them, I ss This is an insensitive region; i1, i2, and i3 are all constants; l d This represents the distance information to the formation reference object; e is a natural constant.

Claims

1. A method for vehicle platooning control based on an identifier, characterized by, include: Establish the formation and affix identification marks to the pedestrians and vehicles participating in the formation; The leader and followers of the queue are determined, and the formation reference objects of the followers are determined according to the formation; the marker is a double concentric circle; For unmanned transport vehicles acting as followers, the lightweight YOLOv5s target recognition algorithm is used to identify the markers of the formation reference objects in real time, and the center point offset is obtained based on a visual method. and distance information from the formation reference object ; and offset the center point and distance information from the formation reference object The desired linear velocity and desired angular velocity of the unmanned transport vehicle are obtained by inputting a virtual spring model and a fuzzy controller, respectively. These values ​​are then input into the vehicle's kinematic model to obtain the desired rotational speeds of the wheels. Based on these wheel rotational speeds, the wheel rotation is controlled by a motor controller. The expression for the desired linear velocity of the unmanned transport vehicle is as follows: in, for linear velocity at time; for linear velocity at time; This represents the virtual spring tension state. The net force acting on the unmanned transport vehicle; For the quality of unmanned transport vehicles; The differential symbol; For time; This refers to the distance information relative to the formation reference object. To maintain a safe distance; This is the spring relaxation distance; The weighted dynamic damping coefficient; The spring constant of the virtual spring; and All are weighting coefficients; The fuzzy controller is a two-dimensional fuzzy controller with dual inputs and a single output. The input to the controller is the center point offset. and distance information from the formation reference object The output is the desired angular velocity. ; center point offset in fuzzy controller The corresponding fuzzy set of variables is set as follows: based on the range of pixel values ​​at the horizontal coordinates. The partition consists of 7 fuzzy sets, including the XS fuzzy set. ), VS Fuzzy Sets[ ), S fuzzy set[ ), M-fuzzy set[ ), L fuzzy set[ ), VL fuzzy set[ ) and XL fuzzy sets[ ]; , , , , and All are center point offsets The corresponding variable fuzzy set partitioning threshold; The distance information between the fuzzy controller and the formation reference object The corresponding fuzzy set of variables is set as follows: based on the range of the distance between people and vehicles [0, ... Five equally divided fuzzy sets, including the VS fuzzy set whose corresponding values ​​are sorted from smallest to largest. ), S fuzzy set[ ), M-fuzzy set[ ), L fuzzy set[ ) and VL fuzzy sets[ ]; , , and All of these are distance information relative to the formation reference object. The corresponding variable fuzzy set partitioning threshold; The desired angular velocity in the fuzzy controller The corresponding fuzzy set of variables is set as follows: based on the range of angular velocity values ​​[ The data is divided into nine equal fuzzy sets, including the NL fuzzy set whose corresponding values ​​are sorted from smallest to largest. , ), NB fuzzy set[ , ), NM fuzzy set[ , ), NS fuzzy set[ ,0), ZO fuzzy set [0], PS fuzzy set (0, 0), ], PM fuzzy set ( , ], PB fuzzy set ( , ] and PL fuzzy set ( , ]; , , , , and All are desired angular velocities The corresponding threshold for fuzzy set partitioning of variables.

2. The vehicle-pedestrian platooning control method based on markers according to claim 1, characterized in that, The resultant force on the unmanned transport vehicle The expression is: in, The combined force received by the priority follower; The combined force experienced by ordinary followers; The force is the traction force of the spring. Number the obstacles; The total number of obstacles; For when The elastic repulsive force generated when entering the area of ​​influence of an obstacle; The distance between the priority follower and the obstacle when the follower is within the obstacle's influence range; The extent of the obstacle's influence; The distance between the priority follower and the formation reference object; The spring-like traction force between ordinary followers and the formation reference object; For when The elastic repulsive force generated when entering the area of ​​influence of an obstacle; This refers to the distance between a normal follower and the obstacle when the follower is within the obstacle's influence range. The number of the adjacent ordinary follower; The total number of adjacent ordinary followers; For when The elastic repulsive force generated when entering the range of an adjacent ordinary follower; The distance between a regular follower and the formation reference object; This refers to the distance between ordinary followers and their adjacent ordinary followers when they are within the influence range; The influence range of adjacent ordinary followers; the priority follower is the unmanned transport vehicle whose formation reference object is a pedestrian; the ordinary follower is the unmanned transport vehicle whose formation reference object is a vehicle.

3. The vehicle-pedestrian platooning control method based on markers according to claim 1, characterized in that, The center point offset The corresponding variable fuzzy set XS fuzzy set [ ), S fuzzy set[ ), M-fuzzy set[ ), L fuzzy set[ ) and XL fuzzy sets[ and desired angular velocity The corresponding variable fuzzy set NL fuzzy set [ , ), NS fuzzy set[ ,0), the ZO fuzzy set after interval expansion [0 0 PS fuzzy set (0, ] and PL fuzzy set ( , The membership function of ] is a bell-shaped membership function; To amplify the difference; the center point offset Corresponding variable fuzzy set VS fuzzy set [ ) and VL fuzzy sets[ Distance information relative to the formation reference object Corresponding variable fuzzy set VS fuzzy set [ ), S fuzzy set[ ), M-fuzzy set[ ), L fuzzy set[ ) and VL fuzzy sets[ ] and desired angular velocity The corresponding variable fuzzy set NB fuzzy set [ , ), NM fuzzy set[ , ), PM fuzzy set ( , ], PB fuzzy set ( , The membership function of ] is a triangular membership function.

4. The vehicle-pedestrian platooning control method based on markers according to claim 1, characterized in that, The fuzzy control rules of the fuzzy controller are as follows: in, Based on center point offset and distance information from the formation reference object The expected angular velocity obtained from the corresponding fuzzy set of variables The corresponding fuzzy set of variables, the distance information between the behavior in the matrix and the formation reference object. The corresponding variable fuzzy sets are, in order, VL fuzzy set, L fuzzy set, M fuzzy set, S fuzzy set, and VS fuzzy set, with the columns in the matrix representing the center point offsets. The corresponding variable fuzzy sets are, in order, XS fuzzy set, VS fuzzy set, S fuzzy set, M fuzzy set, L fuzzy set, VL fuzzy set, and XL fuzzy set.

5. The vehicle-pedestrian platooning control method based on markers according to claim 1, characterized in that, The fuzzy controller determines the center point offset based on the Mandani fuzzy inference method. Distance information relative to the formation reference object and expected angular velocity The ternary fuzzy relation R.

6. The vehicle-pedestrian platooning control method based on markers according to claim 1, characterized in that, offset the center point and distance information from the formation reference object The desired angular velocity of the unmanned transport vehicle is obtained by inputting the fuzzy controller, specifically: Based on the center point offset and distance information from the formation reference object Using a fuzzy controller, the membership values ​​obtained through fuzzy inference are obtained. ; And use the centroid method to determine membership values Defuzzification is performed to obtain the angular velocity of the unmanned transport vehicle; Determine whether the unmanned transport vehicle is in a non-sensitive area. If so, set the expected angular velocity of the unmanned transport vehicle to 0; otherwise, set the expected angular velocity of the unmanned transport vehicle to the angular velocity of the unmanned transport vehicle.

7. The vehicle-pedestrian platooning control method based on markers according to claim 6, characterized in that, The expression for the insensitive region is: in, This is a non-sensitive area; , and All are constants; This refers to the distance information relative to the formation reference object. It is a natural constant.

Images