Lane multi-body dynamic cooperative game method and device, electronic equipment and storage medium

Abstract

The invention provides a lane multi-agent dynamic cooperative game method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining the game earnings corresponding to the passing behaviors of all game participants in a target lane under different cooperation intention conditions; determining respective replicated dynamic equations of the plurality of game participants based on the game earnings; and determining game equilibrium points of the plurality of duplicated dynamic equations, and performing stability analysis on the game equilibrium points to obtain a dynamic cooperative game result of the traffic behavior when each game participant travels. By using the method, the non-motor vehicle and the unmanned distribution vehicle can safely and smoothly run in the non-motor vehicle lane at the same time, the safety risk existing when the unmanned distribution vehicle runs in the non-motor vehicle lane is effectively evaluated, a deterministic relationship exists between the safety risk and the passing behavior of the non-motor vehicle, and sustainable development of the unmanned distribution vehicle can also be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic safety management, and in particular to a lane multi-agent dynamic cooperative game method, device, electronic equipment and storage medium. Background Art

[0002] At present, with the rapid development of intelligent network technology, intelligent network vehicles have gradually become the representative of new transportation tools. Some cities have begun to see unmanned delivery vehicles on the road, and can deliver small items such as express delivery and take-out on non-motorized vehicle lanes; thereby reducing labor costs, improving delivery efficiency, providing 24-hour convenient services, and saving energy and reducing emissions. However, considering that more and more cities have a large number of electric bicycles and more residents ride non-motorized vehicles, non-motorized vehicle lanes are often very crowded when non-motorized vehicles and unmanned delivery vehicles are driving at the same time, and some sections are extremely congested. Therefore, how to achieve safe and smooth driving of non-motorized vehicles and unmanned delivery vehicles in non-motorized vehicle lanes at the same time has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The purpose of the present invention is to provide a lane multi-agent dynamic cooperative game method, device, electronic device and storage medium to solve the technical problem that the existing technology cannot solve the safe and smooth driving of non-motor vehicles and unmanned delivery vehicles in non-motor vehicle lanes at the same time. By establishing a game model of the traffic behavior of unmanned delivery vehicles and non-motor vehicles, it provides support for optimizing road management and resource utilization, which can not only realize the safe and smooth driving of non-motor vehicles and unmanned delivery vehicles in non-motor vehicle lanes at the same time, but also effectively evaluate the safety risks of unmanned delivery vehicles when driving in non-motor vehicle lanes, and have a deterministic relationship with the traffic behavior of non-motor vehicles, but also realize the sustainable development of unmanned delivery vehicles. The many technical effects that can be produced by the preferred technical scheme among the many technical schemes provided by the present invention are described in detail below.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] The present invention provides a lane multi-agent dynamic cooperative game method, comprising:

[0006] Obtain the corresponding game benefits of each game participant in the target lane under different cooperation intentions;

[0007] Determine the replication dynamic equations of each of the plurality of game participants based on the game gains;

[0008] The game equilibrium points of the plurality of replicated dynamic equations are determined, and stability analysis is performed on the game equilibrium points to obtain the dynamic cooperative game results of the traffic behavior of each of the game participants during driving.

[0009] According to a lane multi-agent dynamic cooperative game method provided by the present invention, the method of determining the respective replication dynamic equations of the plurality of game participants based on the game gains includes:

[0010] Based on the game benefits, determining the first expectation of each of the game participants if they choose to cooperate, the second expectation if they choose not to cooperate, and the average expectation;

[0011] Based on each of the first expectations, each of the second expectations and each of the average expectations, a plurality of the replicated dynamic equations are determined.

[0012] According to a lane multi-agent dynamic cooperative game method provided by the present invention, the first expectation of each of the game participants choosing to cooperate, the second expectation of choosing not to cooperate, and the average expectation are determined based on the game benefits, including:

[0013] Based on the game benefits, and the first probability of each of the game participants choosing to cooperate and the second probability of each of the game participants choosing not to cooperate, a game benefit matrix is ​​constructed, wherein the game benefit matrix is ​​used to characterize the game benefits of the different game participants under different cooperation intentions;

[0014] Based on the game payoff matrix, each of the first expectations, each of the second expectations and each of the average expectations are calculated.

[0015] According to a lane multi-agent dynamic cooperative game method provided by the present invention, the determining of the game equilibrium points of the plurality of replicated dynamic equations comprises:

[0016] Based on the evolutionary assumption, the equilibrium point is solved by setting each of the replicated dynamic equations equal to 0 to obtain the game equilibrium point; wherein the evolutionary assumption is used to characterize that any initial point and the point after the initial point evolves belong to a preset two-dimensional space.

[0017] According to a lane multi-agent dynamic cooperative game method provided by the present invention, the stability analysis of the game equilibrium point is performed to obtain the dynamic cooperative game result of the traffic behavior of each of the game participants during driving, including:

[0018] Constructing Jacobian matrices of multiple game participants based on multiple replicated dynamic equations;

[0019] Based on the rank and determinant of the Jacobian matrix, the stability analysis of the game equilibrium point is performed to obtain the dynamic cooperative game result.

[0020] According to a lane multi-agent dynamic cooperative game method provided by the present invention, the stability analysis of the game equilibrium point is performed based on the rank and determinant of the Jacobian matrix to obtain the dynamic cooperative game result, including:

[0021] The game equilibrium point is brought into the Jacobian matrix, and the stability analysis of the game equilibrium point is performed based on the sign of the rank of the Jacobian matrix and the sign of the determinant to obtain the dynamic cooperative game result.

[0022] According to a lane multi-agent dynamic cooperative game method provided by the present invention, the obtaining of the game benefits corresponding to the passing behaviors of each game participant in the target lane under different cooperation intentions includes:

[0023] Obtaining the game benefit of each of the game participants' passing behavior in the target lane when each of the game participants chooses to cooperate; and

[0024] The game benefit of the passing behavior of each of the game participants is obtained when at least one of the game participants chooses not to cooperate in the target lane.

[0025] The present invention also provides a lane multi-agent dynamic cooperative game device, comprising the following modules.

[0026] The profit acquisition module is used to obtain the corresponding game profit of each game participant in the target lane under different cooperation intentions;

[0027] An equation determination module, used to determine the respective replication dynamic equations of the plurality of game participants based on the respective game gains;

[0028] The cooperative game module is used to determine the game equilibrium points of the multiple replicated dynamic equations, and perform stability analysis on the game equilibrium points to obtain the dynamic cooperative game results of the traffic behavior of each game participant during driving.

[0029] The present invention provides a lane multi-agent dynamic cooperative game method, device, electronic device and storage medium, wherein the lane multi-agent dynamic cooperative game method determines the respective replicated dynamic equations of multiple game participants through the game benefits corresponding to the traffic behavior of each game participant in the target lane under different cooperation intentions, and further determines the game equilibrium points of the multiple replicated dynamic equations, and performs stability analysis on the game equilibrium points to obtain the dynamic cooperative game results of the traffic behavior of each game participant when driving. In this way, by establishing a traffic behavior game model of unmanned delivery vehicles and non-motor vehicles, support is provided for optimizing road management and resource utilization, which can not only realize the safe and smooth driving of non-motor vehicles and unmanned delivery vehicles in non-motor vehicle lanes at the same time, but also effectively evaluate the safety risks of unmanned delivery vehicles when driving in non-motor vehicle lanes, and there is a deterministic relationship between the traffic behavior of non-motor vehicles, and the sustainable development of unmanned delivery vehicles can also be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 A schematic diagram of the process flow of a lane multi-agent dynamic cooperative game method provided by an embodiment of the present invention;

[0032] Figure 2 A game dynamic evolution diagram provided by an embodiment of the present invention;

[0033] Figure 3 A schematic diagram of the structure of a lane multi-agent dynamic cooperative game device provided by an embodiment of the present invention;

[0034] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0036] In the embodiments of the present invention, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B may be singular or plural. In the textual description of the present invention, the character " / " generally indicates that the previous and next associated objects are in an "or" relationship. In addition, it should be noted that the serial numbers themselves, such as "first", "second", etc., used to distinguish the objects described in the present invention are only used to distinguish the objects described, and do not have any order or technical meaning.

[0037] At present, with the rapid development of intelligent network technology, intelligent networked vehicles have gradually become the representative of new transportation tools. Unmanned delivery vehicles have begun to appear on the road in some cities, and can deliver express, take-out and other small items on non-motorized lanes; thereby reducing labor costs, improving delivery efficiency, providing 24-hour convenient services, and saving energy and reducing emissions.

[0038] Due to the lack of relevant research on the traffic behavior of unmanned delivery vehicles at this stage, their driving generally occupies non-motorized vehicle lanes. However, considering that more and more cities have a large number of electric bicycles and many residents ride non-motorized vehicles, there is a lack of effective assessment of whether there are safety risks when unmanned delivery vehicles are driving on non-motorized vehicle lanes, and there is an uncertain relationship between the traffic behavior of non-motorized vehicles. As a result, when non-motorized vehicles and unmanned delivery vehicles are driving on non-motorized vehicle lanes at the same time, it is often very crowded and even some sections are extremely congested. Therefore, how to achieve safe and smooth driving of non-motorized vehicles and unmanned delivery vehicles on non-motorized vehicle lanes at the same time has become a technical problem that needs to be solved urgently.

[0039] In order to solve the above technical problems, the present invention provides a lane multi-agent dynamic cooperative game method, device, electronic device and storage medium. By establishing a game model of the traffic behavior of unmanned delivery vehicles and non-motor vehicles, support is provided for optimizing road management and resource utilization. It can not only realize the safe and smooth driving of non-motor vehicles and unmanned delivery vehicles in non-motor vehicle lanes at the same time, but also effectively evaluate the safety risks of unmanned delivery vehicles when driving in non-motor vehicle lanes, and there is a deterministic relationship between the traffic behavior of non-motor vehicles, but also realize the sustainable development of unmanned delivery vehicles.

[0040] Combine the following Figure 1-Figure 4A lane multi-agent dynamic cooperative game method, device, electronic device and storage medium are described, wherein the execution subject of the lane multi-agent dynamic cooperative game method is an electronic device or a server, and the electronic device or the server can communicate with the unmanned delivery vehicle and the non-motor vehicle in the non-motor vehicle lane; the electronic device can be a personal computer (PC), a portable device, a laptop, a smart phone, a tablet computer and a portable wearable device and other devices cloud server, and the server can refer to a server, or a server cluster composed of multiple servers, a cloud server, etc. The present invention does not specifically limit the specific form of the electronic device or the server. Furthermore, the lane multi-agent dynamic cooperative game method can also be applied to a lane multi-agent dynamic cooperative game device set in an electronic device or a server, and the lane multi-agent dynamic cooperative game device can be implemented by software, hardware or a combination of the two. The following takes the execution subject of the lane multi-agent dynamic cooperative game method as an electronic device as an example to describe the lane multi-agent dynamic cooperative game method.

[0041] In order to facilitate understanding of the lane multi-agent dynamic cooperative game method provided by the embodiment of the present invention, the lane multi-agent dynamic cooperative game method provided by the present invention will be described in detail below through the following exemplary embodiments. It can be understood that the following exemplary embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0042] Reference Figure 1 , which is a flow chart of a lane multi-agent dynamic cooperative game method provided by an embodiment of the present invention, such as Figure 1 As shown, the lane multi-agent dynamic cooperative game method includes the following steps 110 to 130.

[0043] Step 110: Obtain the game benefits corresponding to the passing behaviors of each game participant in the target lane under different cooperation intentions.

[0044] Among them, the target lane can be a non-motor vehicle lane where non-motor vehicles and unmanned delivery vehicles are running at the same time, and the number of non-motor vehicles and unmanned delivery vehicles running in the target lane is at least 1.

[0045] The specific cooperation intention situations may include the following four situations: both unmanned delivery vehicles and non-motor vehicles choose to cooperate; the unmanned delivery vehicle chooses to cooperate and the non-motor vehicle chooses not to cooperate; the unmanned delivery vehicle chooses not to cooperate and the non-motor vehicle chooses to cooperate; and both the unmanned delivery vehicle and the non-motor vehicle choose not to cooperate.

[0046] It should be noted that the game participants here include unmanned delivery vehicles and non-motor vehicles, and both are limited rational subjects. Each participant makes rational strategies in the game process. In this way, evolutionary game theory can be based on the following assumptions: emphasizing limited rational decision-making, believing that the game participants are both limited rational parties, and the behavior choices of each game participant will be in a process of constant adjustment and change, and will eventually tend to local stability; thereby establishing a traffic behavior game model.

[0047] Therefore, based on the above assumptions, the present invention sets "traffic behavior" as two aspects of the game, that is, the traffic behavior of the unmanned delivery vehicle and the traffic behavior of non-motor vehicles; and the decision-making of the unmanned delivery vehicle's participation in the traffic behavior of non-motor vehicles can be "cooperation" and "non-cooperation"; the strategy of non-motor vehicles' participation in the traffic behavior of the unmanned delivery vehicle can also be "cooperation" and "non-cooperation".

[0048] Considering that the game between the two parties is an evolutionary game, there are multiple cooperation intention situations in the prevailing behavior game model. Therefore, we can first obtain the prevailing behavior game model parameters and then obtain the game benefits of each game participant in all cooperation intention situations based on the game model parameters.

[0049] Exemplarily, the synergistic benefit value generated by the cooperation between the two parties (i.e., the unmanned delivery vehicle and the non-motor vehicle) is preset to be P, and the additional benefits are distributed according to the preset participation ratio; that is, the benefit distribution ratio of the unmanned delivery vehicle is a, and the benefit distribution ratio of the non-motor vehicle is 1-a, and their respective benefits are aP and (1-a)P respectively.

[0050] Based on this, the game model parameters can be obtained through field surveys, electronic maps, satellite remote sensing images, etc. on the basis of analyzing the traffic flow of the target lane; the game model parameters include but are not limited to the revenue R1 of the unmanned delivery vehicle's traffic behavior when driving on the non-motorized vehicle lane, the revenue R2 of the non-motorized vehicle's traffic behavior when driving on the non-motorized vehicle lane, the input cost C1 of the unmanned delivery vehicle's traffic behavior when driving on the non-motorized vehicle lane, the input cost C2 of the non-motorized vehicle's traffic behavior when driving on the non-motorized vehicle lane, the additional revenue aP of the unmanned delivery vehicle's traffic behavior when the two parties cooperate, the additional revenue (1-a)P of the non-motorized vehicle's traffic behavior when the two parties cooperate, and the additional revenue coefficient of the cooperation between the two parties at different time periods and different sections The length of each road segment in the target lane is L rood , the number of unmanned delivery vehicles driving in the target lane b auto , the number of intersections in the target lane n cross , the number of access points in the target lane n char , the number of roadside units in the target lane n RSU , the number of high-precision positioning devices in the target lane n gps, the number of pedestrian crossings in the target lane n walk and the travel time t of each game participant in the target lane pass In this way, the game benefit corresponding to the passing behavior of each game participant in the target lane under each cooperation intention can be calculated in combination with the pre-configured game benefit algorithm.

[0051] Step 120: Determine the replication dynamic equations of the multiple game participants based on the gains of each game.

[0052] Specifically, based on each game participant in the target lane, each game participant can be taken as a random variable to analyze its probability distribution, and based on the game benefits corresponding to the passage behavior of each game participant in the target lane under each cooperation intention situation, all game benefits of the passage behavior of each game participant can be taken as a set of data to analyze its overall level or central position; thus, based on all the above analysis results, the replication dynamic equation of each game participant is derived, that is, the replication dynamic equation of the unmanned delivery vehicle and the replication dynamic equation of the non-motor vehicle are derived.

[0053] Step 130: determine the game equilibrium points of multiple replicated dynamic equations, and perform stability analysis on the game equilibrium points to obtain the dynamic cooperative game results of the traffic behavior of each game participant during driving.

[0054] Specifically, after deriving the replication dynamic equation of each game participant, multiple game equilibrium points are found by solving the replication dynamic equations jointly, and then the stability analysis of the multiple game equilibrium points is further carried out, so as to determine the game equilibrium stable point that represents the simultaneous implementation of an integrated strategy or a non-integrated strategy by two game participant groups from the multiple game equilibrium points, that is, the game equilibrium stable point where the traffic behavior between unmanned delivery vehicles and non-motor vehicles tends to a stable state after game evolution.

[0055] The lane multi-agent dynamic cooperative game method provided by the present invention determines the respective replicated dynamic equations of multiple game participants through the game benefits corresponding to the traffic behavior of each game participant in the target lane under different cooperation intentions, and further determines the game equilibrium points of the multiple replicated dynamic equations, and performs stability analysis on the game equilibrium points to obtain the dynamic cooperative game results of the traffic behavior of each game participant when driving. In this way, by establishing a traffic behavior game model of unmanned delivery vehicles and non-motor vehicles, support is provided for optimizing road management and resource utilization, which can not only realize the safe and smooth driving of non-motor vehicles and unmanned delivery vehicles in non-motor vehicle lanes at the same time, effectively evaluate the safety risks of unmanned delivery vehicles when driving in non-motor vehicle lanes, and have a deterministic relationship with the traffic behavior of non-motor vehicles, but also realize the sustainable development of unmanned delivery vehicles.

[0056] Based on the above Figure 1 In the lane multi-agent dynamic cooperative game method shown, in an exemplary embodiment, the specific acquisition process of step 110 can be implemented through the following steps.

[0057] The game benefits of each game participant's passing behavior are obtained when each game participant in the target lane chooses to cooperate; and the game benefits of each game participant's passing behavior are obtained when at least one game participant in the target lane chooses not to cooperate.

[0058] Specifically, when the game participants are unmanned delivery vehicles and non-motor vehicles, all the game benefits in the following four cases are obtained, that is, the game benefits of the unmanned delivery vehicle's passage behavior when both the unmanned delivery vehicle and the non-motor vehicle choose to cooperate are obtained as follows:

[0059] R1+δaP-(L road +n auto +n cross +n char +n RSU +n gps +n walk +t pass )C1, and the game benefits of non-motor vehicles’ traffic behavior when both the unmanned delivery vehicle and the non-motor vehicle choose to cooperate are

[0060] R2+δ(1-a)P-(L road +n cross +n char +n walk +t pass )C2;

[0061] The game payoff of the unmanned delivery vehicle’s traffic behavior when the unmanned delivery vehicle chooses to cooperate and the non-motor vehicle chooses not to cooperate is

[0062] R1-(L road +n auto +n cross +n char +n walk +tp ass )C1, and the game payoff of the non-motor vehicle’s traffic behavior when the unmanned delivery vehicle chooses to cooperate and the non-motor vehicle chooses not to cooperate is R2;

[0063] The game payoff of the unmanned delivery vehicle when the unmanned delivery vehicle chooses not to cooperate and the non-motor vehicle chooses to cooperate is R1, and the game payoff of the non-motor vehicle when the unmanned delivery vehicle chooses not to cooperate and the non-motor vehicle chooses to cooperate is R2-(L road +n cross +n char +n walk+t pass )C2;

[0064] The game payoff of the unmanned delivery vehicle's passage behavior when both the unmanned delivery vehicle and the non-motor vehicle choose not to cooperate is R1, and the game payoff of the non-motor vehicle's passage behavior when both the unmanned delivery vehicle and the non-motor vehicle choose not to cooperate is R2.

[0065] Based on the above Figure 1 In the lane multi-agent dynamic cooperative game method shown, in an exemplary embodiment, the specific determination process of step 120 can be implemented through the following steps.

[0066] First, based on the game benefits, the first expectation of each game participant choosing cooperation, the second expectation of choosing non-cooperation, and the average expectation are determined; then, based on the first expectations, the second expectations, and the average expectations, multiple replication dynamic equations are determined.

[0067] Specifically, when the game participants are unmanned delivery vehicles and non-motor vehicles, the expected value U of the unmanned delivery vehicle to choose cooperation can be calculated based on the game benefits corresponding to the traffic behavior of each game participant in the target lane under each cooperation intention. 11 , choose not to cooperate with the expectation U 12 and the average expectation U1, and the expectation U of non-motor vehicle cooperation 21 , choose not to cooperate with the expectation U 22 and the average expectation U2.

[0068] Furthermore, based on the above expectations and average expectations, the replicated dynamic equation F(x) of the unmanned delivery vehicle and the replicated dynamic equation F(y) of the non-motor vehicle are derived, and their expressions are shown in Equation (1) and Equation (2) respectively.

[0069]

[0070]

[0071] In equations (1) and (2), dx / dt represents the derivative of the probability function x of the unmanned delivery vehicle choosing cooperation with respect to the independent variable t, and dy / dt represents the derivative of the probability function y of the non-motor vehicle choosing cooperation with respect to the independent variable t.

[0072] Based on the above Figure 1 The lane multi-agent dynamic cooperative game method shown, in an exemplary embodiment, determines the first expectation of each game participant for choosing cooperation, the second expectation for choosing non-cooperation, and the average expectation based on the game benefits. The specific determination process can be achieved through the following steps.

[0073] Based on the game benefits, as well as the first probability of each game participant choosing to cooperate and the second probability of choosing not to cooperate, a game benefit matrix is ​​constructed. The game benefit matrix is ​​used to characterize the game benefits of the common behaviors of different game participants under different cooperation intentions. Then, based on the game benefit matrix, the first expectations, second expectations and average expectations are calculated.

[0074] Specifically, when the game participants are unmanned delivery vehicles and non-motor vehicles, the game benefit matrix shown in Table 1 can be constructed based on all the game benefits obtained in the above four cases, the first probability of the unmanned delivery vehicle choosing to cooperate (that is, the probability function x of the unmanned delivery vehicle choosing to cooperate) and the second probability of the non-motor vehicle choosing to cooperate (that is, the probability function y of the non-motor vehicle choosing to cooperate).

[0075] Table 1

[0076]

[0077] It should be noted that for Table 1,

[0078] s1=(L road +n auto +n cross +n char +n RSU +n gps +n walk +t pass )C1,

[0079] s2=(L road +n cross +n char +n walk +t pass )C2.

[0080] At this time, the expected value U of the unmanned delivery vehicle to choose cooperation is calculated by equations (3) to (5): 11 , choose not to cooperate with the expectation U 12 and the average expected value U 1。

[0081] U 11 =y(R1+δaP-s1)+(1-y)(R1-s1) (3).

[0082] U 12 =yR1+(1-y)R1=R1 (4).

[0083] U1=xU 11 +(1-x)U 12 (5).

[0084] And, the expected value U of non-motor vehicle selection cooperation is calculated by equations (6) to (8): 21 , choose not to cooperate with the expectation U 22 and the average expected value U 2。

[0085] U 21 =x(R2+δ(1-a)P-s2)+(1-x)(R2-s2) (6).

[0086] U 22 =xR2+(1-x)R2=R2 (7).

[0087] U2=yU 21 +(1-y)U 22 (8).

[0088] Based on the above Figure 1 In the lane multi-agent dynamic cooperative game method shown, in an exemplary embodiment, the game equilibrium points of multiple replicated dynamic equations are determined in step 130, and the specific determination process can be implemented through the following steps.

[0089] Based on the evolutionary assumptions, the equilibrium point is solved by setting each replica dynamic equation equal to 0 to obtain the game equilibrium point; wherein, the evolutionary assumptions are used to characterize that any initial point and the point after the initial point evolves belong to the preset two-dimensional space.

[0090] Specifically, by simultaneously assuming that the two replicated dynamic equations of equations (1) and (2) are both equal to 0, and according to the preset evolution assumptions, that is, any initial point and the point after the evolution of the initial point belong to the preset two-dimensional space V = {(x, y)|0≤x≤1, 0≤y≤1}, the equations are solved, and multiple game equilibrium points can be obtained.

[0091] Exemplarily, the multiple game equilibrium points are specifically 5 game equilibrium points, namely A(0,1), B(0,0), C(1,0), D(1,1), and E(s1 / δaP, s2 / δ(1-a)P).

[0092] Based on the above Figure 1 In the lane multi-agent dynamic cooperative game method shown, in an exemplary embodiment, in step 130, a stability analysis is performed on the game equilibrium point to obtain a dynamic cooperative game result of the traffic behavior of each game participant during driving. The specific process can be implemented through the following steps.

[0093] Firstly, the Jacobian matrices of multiple game participants are constructed based on multiple replicated dynamic equations. Then, based on the rank and determinant of the Jacobian matrix, the stability of the game equilibrium point is analyzed to obtain the dynamic cooperative game result.

[0094] Specifically, the Jacobian matrix ( ) of multiple game participants is constructed through the two replicated dynamic equations of formula (1) and formula (2), and its expression is shown in formula (9).

[0095]

[0096] In formula (9), a 11 =(1-2x)(R1-s1+yδaP)-R1,a 12 =x(1-x)δaP,a 13 =y(1-y)δ(1-a)P,a 14 =(1-2y)(R2-s2+xδaP)-R2.

[0097] At this point, we further conduct stability analysis on the game equilibrium points based on the rank and determinant of the Jacobian matrix. That is, we bring each game equilibrium point into the Jacobian matrix, and conduct stability analysis based on the determinant and trace of the Jacobian matrix to obtain the dynamic cooperative game results.

[0098] Based on the above Figure 1 The lane multi-agent dynamic cooperative game method shown, in an exemplary embodiment, performs stability analysis on the game equilibrium point based on the rank and determinant of the Jacobian matrix to obtain the dynamic cooperative game result, and the specific analysis process can be implemented through the following steps.

[0099] The game equilibrium point is brought into the Jacobian matrix, and the stability of the game equilibrium point is analyzed based on the sign of the rank and determinant of the Jacobian matrix to obtain the dynamic cooperative game result.

[0100] Specifically, when the five game equilibrium points are A(0,1), B(0,0), C(1,0), D(1,1), and E(s1 / δaP, s2 / δ(1-a)P), each game equilibrium point can be brought into the Jacobian matrix, and the determinant det(J) and trace tr(J) of the Jacobian matrix corresponding to each game equilibrium point can be calculated. The specific results are shown in Table 2.

[0101] Table 2

[0102]

[0103] Furthermore, based on the sign of the determinant det(J) of the Jacobian matrix corresponding to each game equilibrium point and the sign of the trace tr(J) of the Jacobian matrix, the equilibrium result corresponding to each game equilibrium point is obtained, and the specific results are shown in Table 3.

[0104] Table 3

[0105]

[0106] Reference Figure 2 The game dynamic evolution diagram shown in Figure 2 Among the five game equilibrium points shown, only two game equilibrium points, B and D, are evolutionarily stable strategies (ESS), corresponding to the two game participant groups simultaneously implementing an integrated strategy or a non-integrated strategy, that is, the traffic behavior game between unmanned delivery vehicles and non-motor vehicles tends to a stable equilibrium.

[0107] also, Figure 2 Among the five game equilibrium points shown, there are two unstable game equilibrium points A and C and the saddle point E. The final result of the evolution of these three points may be complete integration or incomplete integration. Their evolutionary development depends on the scene and mechanism guidance of the traffic behavior of unmanned delivery vehicles and non-motor vehicles. The evolutionary development data obtained according to the actual situation can be brought into the behavioral game model to obtain the game equilibrium point under the actual situation.

[0108] The lane multi-agent dynamic cooperative game device provided by the present invention is described below. The lane multi-agent dynamic cooperative game device described below and the lane multi-agent dynamic cooperative game method described above can be referenced to each other.

[0109] Reference Figure 3 , which is a schematic diagram of the structure of a lane multi-agent dynamic cooperative game device provided by an embodiment of the present invention, such as Figure 3 As shown, the lane multi-agent dynamic cooperative game device 300 includes: a profit acquisition module 310, an equation determination module 320 and a cooperative game module 330.

[0110] The profit acquisition module 310 is used to obtain the game profits corresponding to the passing behaviors of each game participant in the target lane under different cooperation intentions.

[0111] The equation determination module 320 is used to determine the replication dynamic equations of the multiple game participants based on the game benefits.

[0112] The cooperative game module 330 is used to determine the game equilibrium points of multiple replicated dynamic equations, and perform stability analysis on the game equilibrium points to obtain the dynamic cooperative game results of the traffic behavior of each game participant during driving.

[0113] Optionally, the equation determination module 320 is specifically used to determine the first expectation of each game participant for choosing cooperation, the second expectation for choosing non-cooperation, and the average expectation based on each game payoff; and determine multiple replication dynamic equations based on each first expectation, each second expectation, and each average expectation.

[0114] Optionally, the equation determination module 320 is specifically used to construct a game benefit matrix based on the game benefits and the first probability of each game participant choosing to cooperate and the second probability of choosing not to cooperate. The game benefit matrix is ​​used to characterize the game benefits of the common behaviors of different game participants under different cooperation intentions; based on the game benefit matrix, each first expectation, each second expectation and each average expectation is calculated.

[0115] Optionally, the cooperative game module 330 is specifically used to obtain the game equilibrium point based on the evolutionary hypothesis and by solving the equilibrium point by setting each replicated dynamic equation equal to 0; wherein the evolutionary hypothesis is used to characterize that any initial point and the point after the initial point evolves belong to a preset two-dimensional space.

[0116] Optionally, the cooperative game module 330 is specifically used to construct the Jacobian matrix of multiple game participants based on multiple replicated dynamic equations; based on the rank and determinant of the Jacobian matrix, perform stability analysis on the game equilibrium point to obtain the dynamic cooperative game result.

[0117] Optionally, the cooperative game module 330 is specifically used to bring the game equilibrium point into the Jacobian matrix, and perform stability analysis on the game equilibrium point based on the sign of the rank of the Jacobian matrix and the sign of the determinant to obtain a dynamic cooperative game result.

[0118] Optionally, the profit acquisition module 310 is specifically used to obtain the game profit of the passing behavior of each game participant in the target lane when each game participant chooses to cooperate; and to obtain the game profit of the passing behavior of each game participant in the target lane when at least one game participant in the target lane chooses not to cooperate.

[0119] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the lane multi-agent dynamic cooperation game method, which includes: obtaining the game benefits corresponding to the passing behavior of each game participant in the target lane under different cooperation intentions; determining the respective replication dynamic equations of the multiple game participants based on the game benefits; determining the game equilibrium points of the multiple replication dynamic equations, and performing stability analysis on the game equilibrium points to obtain the dynamic cooperation game results of the passing behavior of each game participant when driving.

[0120] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0121] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the lane multi-agent dynamic cooperative game method provided by the above methods. The method includes: obtaining the game benefits corresponding to the passage behavior of each game participant in the target lane under different cooperation intentions; determining the respective replicated dynamic equations of multiple game participants based on each game benefit; determining the game equilibrium points of multiple replicated dynamic equations, and performing stability analysis on the game equilibrium points to obtain the dynamic cooperative game results of the passage behavior of each game participant when driving.

[0122] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the lane multi-agent dynamic cooperative game method provided by the above-mentioned methods, the method comprising: obtaining the game benefits corresponding to the passage behavior of each game participant in the target lane under different cooperation intentions; determining the respective replicated dynamic equations of the multiple game participants based on the game benefits; determining the game equilibrium points of the multiple replicated dynamic equations, and performing stability analysis on the game equilibrium points to obtain the dynamic cooperative game results of the passage behavior of each game participant when driving.

[0123] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0124] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A lane multi-agent dynamic cooperative game method, characterized in that: include: Obtain the corresponding game benefits of each game participant in the target lane under different cooperation intentions; Determine the replication dynamic equations of each of the plurality of game participants based on the game gains; The game equilibrium points of the plurality of replicated dynamic equations are determined, and stability analysis is performed on the game equilibrium points to obtain the dynamic cooperative game results of the traffic behavior of each of the game participants during driving.

2. The lane multi-agent dynamic cooperative game method according to claim 1 is characterized in that: The step of determining the respective replication dynamic equations of the plurality of game participants based on the respective game gains comprises: Based on the game benefits, determining the first expectation of each of the game participants if they choose to cooperate, the second expectation if they choose not to cooperate, and the average expectation; Based on each of the first expectations, each of the second expectations and each of the average expectations, a plurality of the replicated dynamic equations are determined.

3. The lane multi-agent dynamic cooperative game method according to claim 2 is characterized in that: Determining the first expectation of each of the game participants for choosing cooperation, the second expectation for choosing non-cooperation, and the average expectation based on the game benefits includes: Based on the game benefits, and the first probability of each of the game participants choosing to cooperate and the second probability of each of the game participants choosing not to cooperate, a game benefit matrix is ​​constructed, wherein the game benefit matrix is ​​used to characterize the game benefits of the different game participants under different cooperation intentions; Based on the game payoff matrix, each of the first expectations, each of the second expectations and each of the average expectations are calculated.

4. The lane multi-agent dynamic cooperative game method according to any one of claims 1 to 3, characterized in that: The determining of the game equilibrium points of the plurality of replicated dynamic equations comprises: Based on the evolutionary assumption, the equilibrium point is solved by setting each of the replicated dynamic equations equal to 0 to obtain the game equilibrium point; wherein the evolutionary assumption is used to characterize that any initial point and the point after the initial point evolves belong to a preset two-dimensional space.

5. The lane multi-agent dynamic cooperative game method according to any one of claims 1 to 3, characterized in that: The stability analysis of the game equilibrium point is performed to obtain the dynamic cooperative game result of the traffic behavior of each of the game participants during driving, including: Constructing Jacobian matrices of multiple game participants based on multiple replicated dynamic equations; Based on the rank and determinant of the Jacobian matrix, the stability analysis of the game equilibrium point is performed to obtain the dynamic cooperative game result.

6. The lane multi-agent dynamic cooperative game method according to claim 5 is characterized in that: The stability analysis of the game equilibrium point based on the rank and determinant of the Jacobian matrix to obtain the dynamic cooperative game result includes: The game equilibrium point is brought into the Jacobian matrix, and the stability analysis of the game equilibrium point is performed based on the sign of the rank of the Jacobian matrix and the sign of the determinant to obtain the dynamic cooperative game result.

7. The lane multi-agent dynamic cooperative game method according to any one of claims 1 to 3, characterized in that: The obtaining of the game benefits corresponding to the passing behaviors of each game participant in the target lane under different cooperation intentions includes: Obtaining the game benefit of each of the game participants' passing behavior in the target lane when each of the game participants chooses to cooperate; and The game benefit of the passing behavior of each of the game participants is obtained when at least one of the game participants chooses not to cooperate in the target lane.

8. A lane multi-agent dynamic cooperative game device, characterized in that: include: The profit acquisition module is used to obtain the corresponding game profit of each game participant in the target lane under different cooperation intentions; An equation determination module, used to determine the respective replication dynamic equations of the plurality of game participants based on the respective game gains; The cooperative game module is used to determine the game equilibrium points of the multiple replicated dynamic equations, and perform stability analysis on the game equilibrium points to obtain the dynamic cooperative game results of the traffic behavior of each game participant during driving.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, it implements the lane multi-agent dynamic cooperative game method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the lane multi-agent dynamic cooperative game method as described in any one of claims 1 to 7 is implemented.

Images