Cooperative control and evolution method for heterogeneous mobile agents

Abstract

The invention discloses a heterogeneous mobile agent cooperative control and evolution method, which comprises the steps of embedding an emotion cognition module in an artificial mobile agent, and obtaining emotion clues with different granularities to construct an evolution rule of the artificial mobile agent, comprising an acceleration rule, a deceleration rule, a random moderation rule in which a random moderation probability function is introduced, and a position updating rule; constructing automatic mobile agent evolution rules based on the improved interaction potential energy field function, wherein the automatic mobile agent evolution rules comprise an acceleration rule, a deceleration rule and a position updating rule; the acceleration rule is used for adjusting speed; the deceleration rule is used for avoiding collision; the random slowing-down rule is used for randomly slowing down the manual moving agent; the position updating rule is used for determining positions; and the motion path of the heterogeneous mobile agent is controlled and planned through a field force planning method. According to the method, the problem of cooperative control of the heterogeneous mobile intelligent agents in a complex dynamic environment is solved, the cooperative capability among the heterogeneous intelligent agents is enhanced, and the collision avoidance safety is improved.

Description

Technical Field

[0001] The present invention relates to a method for cooperative control and evolution of heterogeneous mobile intelligent bodies, belonging to the technical field of formation and cooperative control of mobile intelligent bodies. Background Art

[0002] The idea of ​​the mobile agent system is that when multiple agents realize the desired formation movement, they comprehensively consider the working characteristics of artificial mobile agents and automatic mobile agents, in order to complete the integrated collaborative control and formation management tasks. It has always received widespread attention because it is widely used in traffic flow operation control, aerial equipment formation and space exploration. Most of the previous studies did not fully consider the impact of the emotional factors of artificial mobile agents and the uncertain factors in the external environment on the operation of multi-agent collaborative control systems. In practical applications, the influence of individual emotional factors and uncertain environments will not only cause each agent to deviate from the desired motion path, but also have an adverse effect on the collaborative formation operation of multiple agents. Therefore, the research on the problem of multi-agent collaborative control in uncertain environments has extremely important theoretical and practical value.

[0003] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgement or any form of suggestion that the information constitutes the prior art already known to ordinary technicians in this field. Summary of the invention

[0004] The technical problem to be solved by the present invention is that the existing technology does not fully consider the individual emotional factors of artificial mobile agents and the impact of uncertain factors in the local environment on the collaborative control of multiple agents, resulting in the problem that the collaborative control and evolution of heterogeneous mobile agent systems cannot approach the actual situation.

[0005] In order to solve the above technical problems, the present invention is implemented by adopting the following technical solutions.

[0006] The present invention provides a method for cooperative control and evolution of heterogeneous mobile agents, comprising:

[0007] Using continuous cellular automata to mix artificial mobile agent cells and automatic mobile agent cells according to a preset ratio, and generating artificial mobile agents and automatic mobile agents through an iterative process;

[0008] Embedding an emotion recognition module in the artificial mobile agent to obtain emotion clues of different granularities based on fuzzy theory;

[0009] According to the emotional clues of different granularities, constructing evolution rules of the artificial mobile agent and updating the state of the artificial mobile agent;

[0010] The evolution rules of the artificial mobile agent include an acceleration rule, a deceleration rule, a random slowing-down rule introducing a random slowing-down probability function, and a position updating rule;

[0011] Obtain the position, velocity and acceleration of the automatic mobile agent, construct the automatic mobile agent evolution rules based on the improved interaction potential energy field function, and update the state of the automatic mobile agent;

[0012] The automatic mobile agent evolution rules include acceleration rules, deceleration rules and position update rules;

[0013] The acceleration rule is used to adjust the speed of the automatic mobile agent or the artificial mobile agent;

[0014] The deceleration rule is used to avoid collision between the automatic mobile agent and the artificial mobile agent;

[0015] The random slowing down rule introducing the random slowing down probability function is used to randomly slow down the artificial mobile agent based on the emotion influencing factor and the random slowing down probability function;

[0016] The position update rule is used to determine the position of the automatic mobile agent or the artificial mobile agent;

[0017] The movement and motion path of the automatic mobile agent are controlled through the field force planning method, and the motion path of the artificial mobile agent is planned by utilizing the perception information sharing mechanism of the automatic mobile agent, combining the movement and motion path of the automatic mobile agent with the emotional clues of different granularities in the operation scenario of the artificial mobile agent.

[0018] Based on the aforementioned heterogeneous mobile agent collaborative control and evolution method, the emotion recognition module includes an emotion perception layer and a feature fusion network. The emotion perception layer adopts a dual-channel fusion attention mechanism network structure. According to fuzzy theory, the method of obtaining emotion clues of different granularities includes:

[0019] The emotion perception layer is used to capture and extract character emotion clues and scene emotion clues in the running scene through the character channel and the scene channel respectively to form a preliminary feature vector;

[0020] The emotion values ​​of continuous dimensions in the preliminary feature vector are quantified using fuzzy theory to obtain emotion influencing factors.

[0021] The emotion influencing factors are input into the feature fusion network for fusion processing to obtain emotion clues of different granularities.

[0022] Based on the aforementioned heterogeneous mobile agent collaborative control and evolution method, the method of obtaining the emotion influencing factor by quantifying the emotion value of the continuous dimension in the preliminary feature vector using fuzzy theory includes:

[0023] Based on the Mamdani model, the emotional pleasure, emotional arousal and emotional dominance are taken as input variables, and the emotional influencing factors are taken as output variables to construct a three-value input and single-value output emotional fuzzy reasoning model.

[0024] According to the emotion measurement values ​​of emotion pleasure, emotion arousal and emotion dominance, input variable fuzzy sets are constructed respectively, including emotion pleasure fuzzy set 1, emotion pleasure fuzzy set 2 and emotion pleasure fuzzy set 3, emotion arousal fuzzy set 1, emotion arousal fuzzy set 2 and emotion arousal fuzzy set 3, and emotion dominance fuzzy set 1, emotion dominance fuzzy set 2 and emotion dominance fuzzy set 3;

[0025] Constructing output variable fuzzy sets, including emotion factor fuzzy set 1, emotion factor fuzzy set 2 and emotion factor fuzzy set 3;

[0026] Using Gaussian membership function, the input variable fuzzy set and the output variable fuzzy set are fuzzified;

[0027] Analyze the data samples of emotional pleasure, emotional arousal, and emotional dominance in the Emotic dataset to obtain an artificial empirical rule library that can reflect the relationship between input variables and output variables;

[0028] According to the artificial experience rule base, the emotion representation fuzzy rule matrix is ​​constructed;

[0029] The continuous sentiment values ​​in the preliminary feature vector are converted into the membership of the input set fuzzy set through fuzzification;

[0030] According to the emotion representation fuzzy rule matrix, the input fuzzy set is mapped to the output fuzzy set using a fuzzy reasoner, and the final output fuzzy set is obtained by calculation;

[0031] The final output fuzzy set is defuzzified using the centroid method to obtain the emotion influence factor.

[0032] Based on the aforementioned heterogeneous mobile agent collaborative control and evolution method, the emotion representation fuzzy rule matrix is ​​constructed according to the artificial experience rule base, and the constructed emotion representation fuzzy rule matrix is ​​expressed as:

[0033] (1);

[0034] Among them, the first three columns of the emotion representation fuzzy rule matrix represent the fuzzy set indexes corresponding to the input variables emotion pleasure, emotion arousal and emotion dominance, respectively; the fourth column represents the fuzzy set indexes corresponding to the output variable emotion influencing factor; the fifth column represents the rule weights of the emotion representation fuzzy rule matrix; the sixth column represents the connectives of the rules of the emotion representation fuzzy rule matrix; when the rule weight is 1, the logical connective is “and”; when the rule weight is 2, the logical connective is “or”;

[0035] Among them, the rules of the first row of the emotion representation fuzzy rule matrix are: when the input emotion pleasure belongs to the emotion pleasure fuzzy set 1, the input emotion arousal belongs to the emotion arousal fuzzy set 3, and the input emotion dominance belongs to the emotion dominance fuzzy set 3, the output emotion factor fuzzy set 1, the rule weight of the first row is 1, and the logical connective is "and";

[0036] The rules in the second row of the fuzzy rule matrix are: when the input emotional pleasure belongs to the emotional pleasure fuzzy set 2, the input emotional arousal belongs to the emotional arousal fuzzy set 1, and the input emotional dominance belongs to the emotional dominance fuzzy set 3, the output emotional factor fuzzy set 2, the rule weight in the second row is 1, and the logical connective is "and";

[0037] The rule in the third row of the fuzzy rule matrix is: when the input emotional pleasantness belongs to emotional pleasantness fuzzy set one and the input emotional arousal belongs to emotional arousal fuzzy set two and the input emotional dominance belongs to emotional dominance fuzzy set one, the output emotional factor fuzzy set three, the rule weight in the third row is 1, and the logical connective is "and".

[0038] Based on the aforementioned heterogeneous mobile agent collaborative control and evolution method, when the operation scenario is road traffic flow, the continuous cellular automaton uses vehicles as cells to generate manually driven vehicles and automatically connected vehicles through an iterative process. The state information of the continuous cellular automaton , expressed as:

[0039] (2);

[0040] in, , and Respectively indicate this vehicle The position, velocity and acceleration at the current time t, Indicates this vehicle The ideal state vector of the vehicle is Indicates this vehicle The driver's emotional state vector, , , Respectively indicate this vehicle The expected speed of the vehicle, the vehicle The expected following distance of the vehicle and the vehicle The vehicle safety time interval, , and Respectively indicate at the current moment This car Measures of drivers’ emotional pleasure, emotional arousal, and emotional dominance.

[0041] The continuous cellular automaton uses vehicles as cells and generates manually driven vehicles and automatically connected vehicles through an iterative process. It is expressed as:

[0042] (3);

[0043] in, represents the maximum acceleration of the vehicle, represents the vehicle acceleration index, and Respectively indicate the front vehicle and this car The relative distance and relative speed, Indicates the absolute value of the vehicle's comfortable deceleration. Indicates the static safety distance, Indicates the vehicle ahead The velocity at the current time t.

[0044] Based on the aforementioned heterogeneous mobile agent collaborative control and evolution method, the final output fuzzy set is defuzzified using the centroid method to obtain the emotion influence factor, and the obtained emotion influence factor k is expressed as:

[0045] (4);

[0046] in, , and Respectively indicate at the current moment This car Measures of driver emotional pleasure, emotional arousal, and emotional dominance.

[0047] Based on the aforementioned heterogeneous mobile agent cooperative control and evolution method, the evolution rules of the artificial mobile agent include acceleration rules, deceleration rules, random slowing-down rules introducing random slowing-down probability functions, and position update rules; wherein the acceleration rule of the artificial mobile agent is expressed as:

[0048] (5);

[0049] in, Indicates this vehicle The velocity at the current time t, Indicates this vehicle The velocity at the current time t, Indicates this vehicle The acceleration at the current time t, Indicates the maximum speed limit for the road.

[0050] The deceleration rule of the artificial mobile agent is expressed as:

[0051] (6);

[0052] in, Indicates safe following distance. Indicates the vehicle ahead and this car The relative distance Indicates the stationary safety distance.

[0053] Based on this car Driver's emotional influence factor and random slowing probability function, the vehicle The driver has The probability of random deceleration is, where represents the random slowing probability of vehicle n, and the random slowing rule introducing the random slowing probability function is expressed as:

[0054] (7);

[0055] in, Indicates this vehicle The velocity at the current time t, Indicates this vehicle The velocity at the current time t, Indicates the deceleration value;

[0056] The position of vehicle n at the current time t and speed Together determine the position of vehicle n at the next moment t+1 ;

[0057] The expression of the position update rule of the artificial mobile agent is:

[0058] (8);

[0059] in, Indicates in this vehicle The position at the next moment (t+1), Indicates this vehicle At the current time t, Indicates this vehicle The velocity at the current time t.

[0060] Based on the aforementioned heterogeneous mobile agent cooperative control and evolution method, the evolution rules of the automatic mobile agent include acceleration rules, deceleration rules and position update rules;

[0061] Among them, the acceleration rule of the automatic mobile agent is expressed as:

[0062] (9);

[0063] in, Indicates this vehicle The velocity at the current time t, Indicates this vehicle The velocity at the current time t, Indicates the maximum speed limit of the road. represents the vehicle calculated according to the double integral motion equation The acceleration of Indicates this vehicle Quality, represents the safe following distance, and F represents the total virtual force.

[0064] The deceleration rule of the autonomous mobile agent is: when the repulsive force generated by the virtual potential field of the current vehicle n+1 is greater than the gravitational force, the total virtual force on the vehicle n is expressed as repulsive force;

[0065] The position update rules of the automatic mobile agent are consistent with those of the artificial mobile agent.

[0066] Based on the aforementioned heterogeneous mobile agent cooperative control and evolution method, a method for controlling the motion and motion path of an automatic mobile agent through a field force planning method includes:

[0067] The position, velocity and acceleration of the autonomous mobile agent are acquired in real time through the sensors of the autonomous mobile agent, and the interaction potential energy field function is defined;

[0068] Introduce virtual force of position, virtual force of velocity and virtual force of acceleration to improve the interaction potential energy field function;

[0069] According to the improved interaction potential energy field function, the virtual forces on each autonomous mobile agent from other agents are calculated, including position virtual force, velocity virtual force and acceleration virtual force;

[0070] Perform vector synthesis of all virtual forces on each autonomous mobile agent to obtain the total virtual force of each autonomous mobile agent;

[0071] The total virtual force is used as a control signal to adjust the speed and direction of the autonomous mobile agent through a control algorithm.

[0072] Use the virtual force of potential energy field to plan the driving path of the autonomous mobile agent.

[0073] Based on the aforementioned heterogeneous mobile agent cooperative control and evolution method, the position virtual force includes position virtual gravity and position virtual repulsion, the velocity virtual force includes velocity virtual gravity and velocity virtual repulsion, and the acceleration virtual force includes acceleration virtual gravity and acceleration virtual repulsion;

[0074] The virtual repulsion force It is expressed as:

[0075] (10);

[0076] in, represents the virtual repulsive potential energy coefficient of the position, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, represents the distance vector from the autonomous connected vehicle to the target point, || || represents the modulus of the distance vector from the autonomous connected vehicle to the target point.

[0077] The virtual gravity of the position It is expressed as:

[0078] (11);

[0079] in, represents the virtual gravitational potential energy coefficient of the position, Indicates the distance from the autonomous connected vehicle to the target point.

[0080] The velocity virtual repulsion It is expressed as:

[0081] (12);

[0082] in, represents the velocity virtual repulsive potential energy coefficient, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, represents the speed difference vector between the preceding vehicle and the vehicle itself, || || represents the modulus of the speed difference vector between the preceding vehicle and the vehicle itself.

[0083] The velocity is virtual gravity It is expressed as:

[0084] (13);

[0085] in, represents the velocity virtual gravitational potential energy coefficient, Indicates the speed difference between the desired speed and the current speed. represents the distance from the autonomous networked vehicle to the target point, tanh represents the hyperbolic tangent function, || || represents the modulus of the speed difference vector between the desired speed and the current speed.

[0086] The acceleration virtual repulsion It is expressed as:

[0087] (14);

[0088] in, represents the acceleration virtual repulsive potential energy coefficient, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, represents the acceleration difference vector between the preceding vehicle and the vehicle itself, || || represents the modulus of the acceleration difference vector between the preceding vehicle and the vehicle itself.

[0089] The acceleration virtual gravity It is expressed as:

[0090] ; (15);

[0091] in, represents the virtual gravitational potential energy coefficient of acceleration, Indicates the difference between the expected acceleration and the current acceleration. represents the distance from the autonomous connected vehicle to the target point, || || represents the modulus of the difference vector between the expected acceleration and the current acceleration.

[0092] The present invention introduces emotion influencing factors, accurately quantifies continuous dimensional emotion values ​​and integrates them into the random slowing-down probability function, thereby improving the authenticity and accuracy of the collaborative control of heterogeneous mobile agents and making the behavior of artificial mobile agents closer to reality.

[0093] At the same time, the method based on potential energy field virtual force converts the position, velocity and acceleration information of the autonomous mobile agent into virtual field force, effectively controls its movement through field force planning, and enhances the autonomous decision-making and path planning capabilities of the autonomous mobile agent.

[0094] In response to the problem of collaborative control in complex environments, the present invention significantly improves the adaptability and overall intelligence of collaborative control of heterogeneous mobile agents through emotion recognition, deceleration rules and potential field optimization.

[0095] Compared with the prior art, the present invention has the following beneficial effects:

[0096] The present invention embeds an emotion recognition module in the artificial mobile agent and uses fuzzy theory to process emotion clues of different granularities, so that the agent can perceive and understand emotional information in complex environments. The evolution rules constructed based on emotion clues of different granularities make the behavior of the artificial mobile agent more flexible and changeable. In particular, the introduction of random slowing-down probability functions and emotion influencing factors can dynamically adjust the movement speed according to the emotional state, which not only increases the diversity of the agent's behavior, but also improves the agent's survivability and adaptability in complex environments.

[0097] At the same time, the present invention effectively realizes dynamic collision avoidance and coordinated motion between agents by modeling virtual repulsive fields and virtual gravitational fields, and constructing the evolution rules of automatic mobile agents based on improved interaction potential field functions. The present invention also uses field force planning methods to control the motion and path of automatic mobile agents, and combines the perception information sharing mechanism and emotional clues to plan a more reasonable and efficient motion path for artificial mobile agents. This solves the problem that the existing technology does not fully consider the individual emotional factors of artificial mobile agents and the influence of uncertain factors in the local environment on the collaborative control of multiple agents, resulting in the inability of the collaborative control and evolution of heterogeneous mobile agent systems to approach the real situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0098] Figure 1 is a schematic diagram of the structure of a heterogeneous mobile agent provided by an embodiment of the present invention;

[0099] Figure 2 It is a schematic diagram of the process of constructing the emotion fuzzy reasoning model provided by an embodiment of the present invention;

[0100] Figure 3 It is a simulation schematic diagram of a virtual force model of a potential energy field in an operation scenario of collaborative control of heterogeneous mobile agents provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0101] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0102] Example 1

[0103] like Figure 1 As shown, this embodiment introduces a method for cooperative control and evolution of heterogeneous mobile agents, including:

[0104] Step 1: Using continuous cellular automata to mix artificial mobile agent cells and automatic mobile agent cells according to a preset ratio, and generate artificial mobile agents and automatic mobile agents through an iterative process;

[0105] Through the continuous cellular automaton model, artificial mobile agents and automatic mobile agents are mixed in preset proportions to simulate heterogeneous mobile agents in a real interactive environment, providing a basic simulation environment for subsequent collaborative control research, making the experimental results closer to actual application scenarios, and enhancing the practicality and reliability of the research.

[0106] Step 2: embedding an emotion recognition module in the artificial mobile agent, and obtaining emotion clues of different granularities according to fuzzy theory;

[0107] By embedding the emotion recognition module and using fuzzy theory to obtain and quantify the emotional cues of artificial mobile agents, the emotional cues will affect the movement of artificial mobile agents. Considering emotional cues of different granularities can make the behavioral decisions of artificial mobile agents more in line with human reality, improve the authenticity and accuracy of the simulation model, and also introduce a new dimension for the collaborative control of heterogeneous agents - emotional factors.

[0108] Step 3: constructing an evolution rule of the artificial mobile agent according to the emotional clues of different granularities, and updating the state of the artificial mobile agent;

[0109] The evolution rules of the artificial mobile agent include an acceleration rule, a deceleration rule, a random slowing-down rule introducing a random slowing-down probability function, and a position updating rule;

[0110] Step 4: Construct the evolution rules of the artificial mobile agent including acceleration rules, deceleration rules, random slowing rules and position update rules according to the emotional clues to update the state of the artificial mobile agent.

[0111] The above rules enable the behavior of the artificial mobile agent to be dynamically adjusted according to emotional changes, increasing the complexity and authenticity of the simulation environment. At the same time, the random slowing down rule introduces uncertainty, making the behavior more difficult to predict.

[0112] Step 5: Obtain the position, velocity and acceleration of the automatic mobile agent, construct the automatic mobile agent evolution rule based on the improved interaction potential energy field function, and update the state of the automatic mobile agent;

[0113] Based on the improved interaction potential energy field function, the acceleration rules, deceleration rules and position update rules of the autonomous mobile agent are constructed to optimize the motion path and obstacle avoidance capability of the autonomous mobile agent, thereby improving the adaptability and safety of the autonomous mobile agent in complex environments, and enabling the autonomous mobile agent to respond to various situations more intelligently, such as avoiding pedestrians and adjusting speed.

[0114] The automatic mobile agent evolution rules include acceleration rules, deceleration rules and position update rules;

[0115] The acceleration rule is used to adjust the speed of the automatic mobile agent or the artificial mobile agent.

[0116] The deceleration rule is used to avoid collision between the automatic mobile agent and the artificial mobile agent.

[0117] The acceleration rule is used to adjust the speed of the agent to match its movement requirements, and the deceleration rule is used to avoid collisions and ensure safety. Together, they maintain the stability and safety of the simulation environment, reduce the occurrence of collision events, and improve the efficiency of the agent's movement.

[0118] The random slowing down rule introducing the random slowing down probability function is used to randomly slow down the artificial mobile agent based on the emotion influencing factor and the random slowing down probability function; the artificial mobile agent is randomly slowed down by the emotion influencing factor and the random slowing down probability function, which increases the authenticity and unpredictability of the simulation environment and simulates the random behavior pattern of humans or organisms under the influence of emotions.

[0119] The position update rule is used to determine the position of the automatic mobile agent or the artificial mobile agent. The position of the agent is updated according to the speed and direction to promote its movement in the simulation environment, thereby realizing the continuous movement and interaction of the agent in the virtual space and providing a basis for subsequent path planning and decision-making.

[0120] Step 6: Control the movement and movement path of the automatic mobile agent through the field force planning method, and use the perception information sharing mechanism of the automatic mobile agent, combined with the movement and movement path of the automatic mobile agent and the emotional clues of different granularities in the operation scenario of the artificial mobile agent, to plan the movement path of the artificial mobile agent.

[0121] The motion path of the automatic mobile agent is controlled by modeling virtual repulsive fields and virtual gravitational fields, and the motion path of the artificial mobile agent is planned by using the perception information sharing mechanism combined with emotional cues of different granularities. This improves the path planning efficiency and obstacle avoidance capabilities of the automatic mobile agent, while making the motion path of the artificial mobile agent more reasonable and safe.

[0122] Example 2

[0123] Based on the same inventive concept as Example 1, this example introduces a specific application example of a method for collaborative control and evolution of heterogeneous mobile agents.

[0124] This embodiment takes traffic flow operation control as an example. The road traffic flow is composed of a mixture of connected and automated vehicles (CAV) and human-driven vehicles (HDV), which share road resources. Compared with human-driven vehicles HDV, connected and automated vehicles CAV have a smaller reaction delay time. During driving, the connected and automated vehicles CAV automatically maintain the headway with the vehicle in front. At the same time, the connected and automated vehicles CAV have environmental perception and autonomous driving capabilities, supporting more flexible and intelligent decision-making. After the human-driven vehicle HDV is embedded with the emotion recognition module, a driving strategy that incorporates the driver's emotional influencing factors is generated. The combination of human-driven vehicles HDV and connected and automated vehicles CAV can simulate mixed traffic flow characteristics that are close to reality.

[0125] When the operating scenario is road traffic flow, the continuous cellular automaton takes the vehicle as a cell and uses the continuous cellular automaton to mix the HDV cells of the manually driven vehicle and the CAV cells of the automatically connected vehicle in a preset proportion, generating manually driven vehicles and automatically connected vehicles through an iterative process.

[0126] The state information of the continuous cellular automaton , expressed as:

[0127] ;

[0128] in, , and Respectively indicate this vehicle The position, velocity and acceleration at the current time t, Indicates this vehicle The ideal state vector of the vehicle is Indicates this vehicle The driver's emotional state vector, , , Respectively indicate this vehicle The expected speed of the vehicle, the vehicle The expected following distance of the vehicle and the vehicle The vehicle safety time interval, , and Respectively indicate at the current moment This car Measures of drivers’ emotional pleasure, emotional arousal, and emotional dominance.

[0129] Among them, the car At the current time t the acceleration It is expressed as:

[0130] ;

[0131] in, represents the maximum acceleration of the vehicle, represents the vehicle acceleration index, and Respectively indicate the front vehicle and this car The relative distance and relative speed, Indicates the absolute value of the vehicle's comfortable deceleration. Indicates the static safety distance, Indicates the vehicle ahead The velocity at the current time t.

[0132] Among them, the car It refers to a manually driven vehicle (HDV) or an autonomous networked vehicle (CAV) operating in mixed traffic flow.

[0133] An emotion recognition module is embedded in the HDV body, and emotion clues of different granularities are obtained according to fuzzy theory. The specific steps include:

[0134] (1) The emotion perception layer is used to capture and extract character emotion clues and scene emotion clues in the running scene through the character channel and the scene channel respectively to form a preliminary feature vector.

[0135] (2) Use fuzzy theory to quantify the emotion values ​​of the continuous dimensions in the preliminary feature vector to obtain the emotion influencing factors, such as Figure 2 As shown, the specific steps include:

[0136] Based on the Mamdani model, the emotional pleasure, emotional arousal and emotional dominance are taken as input variables, and the emotional influencing factors are taken as output variables to construct a three-value input and single-value output emotional fuzzy reasoning model.

[0137] According to the emotion measurement values ​​of emotion pleasure, emotion arousal and emotion dominance, input variable fuzzy sets are constructed respectively, including emotion pleasure fuzzy set 1, emotion pleasure fuzzy set 2 and emotion pleasure fuzzy set 3, emotion arousal fuzzy set 1, emotion arousal fuzzy set 2 and emotion arousal fuzzy set 3, and emotion dominance fuzzy set 1, emotion dominance fuzzy set 2 and emotion dominance fuzzy set 3;

[0138] Constructing output variable fuzzy sets, including emotion factor fuzzy set 1, emotion factor fuzzy set 2 and emotion factor fuzzy set 3;

[0139] Using Gaussian membership function, the input variable fuzzy set and the output variable fuzzy set are fuzzified;

[0140] Analyze the data samples of emotional pleasure, emotional arousal, and emotional dominance in the Emotic dataset to obtain an artificial empirical rule library that can reflect the relationship between input variables and output variables;

[0141] According to the artificial experience rule base, the emotion representation fuzzy rule matrix is ​​constructed, and the constructed emotion representation fuzzy rule matrix is ​​expressed as:

[0142] ;

[0143] Among them, the first three columns of the emotion representation fuzzy rule matrix represent the fuzzy set indexes corresponding to the input variables emotion pleasure, emotion arousal and emotion dominance, respectively; the fourth column represents the fuzzy set indexes corresponding to the output variable emotion influencing factor; the fifth column represents the rule weights of the emotion representation fuzzy rule matrix; the sixth column represents the connectives of the rules of the emotion representation fuzzy rule matrix; when the rule weight is 1, the logical connective is “and”; when the rule weight is 2, the logical connective is “or”;

[0144] Among them, the rules of the first row of the emotion representation fuzzy rule matrix are: when the input emotion pleasure belongs to the emotion pleasure fuzzy set 1, the input emotion arousal belongs to the emotion arousal fuzzy set 3, and the input emotion dominance belongs to the emotion dominance fuzzy set 3, the output emotion factor fuzzy set 1, the rule weight of the first row is 1, and the logical connective is "and";

[0145] The rules in the second row of the fuzzy rule matrix are: when the input emotional pleasure belongs to the emotional pleasure fuzzy set 2, the input emotional arousal belongs to the emotional arousal fuzzy set 1, and the input emotional dominance belongs to the emotional dominance fuzzy set 3, the output emotional factor fuzzy set 2, the rule weight in the second row is 1, and the logical connective is "and";

[0146] The rule in the third row of the fuzzy rule matrix is: when the input emotional pleasantness belongs to emotional pleasantness fuzzy set one and the input emotional arousal belongs to emotional arousal fuzzy set two and the input emotional dominance belongs to emotional dominance fuzzy set one, the output emotional factor fuzzy set three, the rule weight in the third row is 1, and the logical connective is "and".

[0147] The continuous sentiment values ​​in the preliminary feature vector are converted into the membership of the input set fuzzy set through fuzzification;

[0148] According to the emotion representation fuzzy rule matrix, the input fuzzy set is mapped to the output fuzzy set using a fuzzy reasoner, and the final output fuzzy set is obtained by calculation;

[0149] The final output fuzzy set is defuzzified using the centroid method to obtain the emotional impact factor, which is expressed as:

[0150] ;

[0151] Among them, k represents the emotion influencing factor, , and Respectively indicate at the current moment This car Measures of driver emotional pleasure, emotional arousal, and emotional dominance.

[0152] (3) Inputting the emotion influencing factors into the feature fusion network for fusion processing to obtain emotion clues of different granularities.

[0153] According to the emotional clues of different granularities, the evolution rules of the artificial mobile agent are constructed, and the state of the artificial mobile agent is updated.

[0154] The evolution rules of the artificial mobile agent include an acceleration rule, a deceleration rule, a random slowing-down rule introducing a random slowing-down probability function, and a position updating rule.

[0155] The acceleration rule of the constructed artificial mobile agent is expressed as:

[0156] ;

[0157] in, Indicates this vehicle The speed at time t, Indicates this vehicle The speed at time t, Indicates this vehicle The acceleration at time t is Indicates the maximum speed limit for the road.

[0158] The deceleration rule of the constructed artificial mobile agent is expressed as:

[0159] ;

[0160] in, Indicates safe following distance. Indicates the vehicle ahead and this car The relative distance Indicates the stationary safety distance.

[0161] Considering that the driver's emotional factors will affect the operation of the HDV, based on this vehicle Driver's emotional influence factor and random slowing probability function, the vehicle The driver has The probability of random deceleration is, where represents the random slowing probability of vehicle n, and the random slowing rule constructed by introducing the random slowing probability function is expressed as:

[0162] ;

[0163] in, Indicates this vehicle The speed at time t, Indicates this vehicle The speed at time t, Indicates the deceleration value;

[0164] The position of vehicle n at the current time t and speed Together determine the position of vehicle n at the next moment t+1 , the expression of the position update rule of the constructed artificial mobile agent is:

[0165] ;

[0166] in, Indicates in this vehicle The position at the next moment (t+1), Indicates this vehicle At the current time t, Indicates this vehicle The velocity at the current time t.

[0167] The position, velocity and acceleration of the automatic mobile agent are obtained, and based on the improved interaction potential energy field function, the evolution rules of the automatic mobile agent are constructed, and the state of the automatic mobile agent is updated. The specific steps include:

[0168] A virtual potential energy field is defined, and the movement and movement path of the automatic mobile intelligent body are controlled by a field force planning method. The virtual potential energy field contains a potential energy field virtual force, and the potential energy field virtual force is generated by a potential energy gradient in the potential energy field.

[0169] The virtual potential energy field usually consists of two parts: the attraction field generated by the target point and the repulsion field generated by the obstacle.

[0170] like Figure 3As shown, the potential energy field virtual force can be divided into position virtual force, velocity virtual force and acceleration virtual force according to the generation mechanism; the potential energy field virtual force can be divided into position virtual gravity and position virtual repulsion, velocity virtual gravity and velocity virtual repulsion, and acceleration virtual gravity and acceleration virtual repulsion according to the action relationship.

[0171] Virtual Position Repulsion Field It is expressed as:

[0172] ;

[0173] Velocity repulsive field It is expressed as:

[0174] ;

[0175] The acceleration repulsive field It is expressed as:

[0176] ;

[0177] in, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, Indicates the speed difference between the front vehicle and the vehicle itself. Indicates the acceleration difference between the front vehicle and the vehicle itself. represents the virtual repulsive potential energy coefficient of the position, represents the velocity virtual repulsive potential energy coefficient, Represents the virtual repulsive potential energy coefficient of acceleration.

[0178] Since the vehicle is always expected to travel at the maximum speed, if the maximum expected speed is not reached, it will accelerate until it reaches the speed and moves at a uniform speed.

[0179] So the gravitational field of the virtual position It is expressed as:

[0180] ;

[0181] So the virtual velocity gravitational field It is expressed as:

[0182] ;

[0183] So the gravitational field of the virtual acceleration It is expressed as:

[0184] ;

[0185] in, represents the distance from the autonomous networked vehicle to the target point, Indicates the speed difference between the desired speed and the current speed. Indicates the difference between the expected acceleration and the current acceleration. represents the virtual gravitational potential energy coefficient of the position, represents the velocity virtual gravitational potential energy coefficient, Represents the virtual gravitational potential energy coefficient of acceleration.

[0186] Since the motion planning of the autonomous connected vehicle in the potential energy field is related to the potential energy virtual force, the relationship between the potential energy field virtual force and the potential energy field is expressed as:

[0187] ;

[0188] in, Represents potential energy field The generated virtual force has an index of i, and the negative sign indicates that the direction of the virtual force of the potential energy field extends from the high potential field energy to the low potential field energy; Represents the gradient function.

[0189] The position virtual force includes position virtual gravity and position virtual repulsion, the velocity virtual force includes velocity virtual gravity and velocity virtual repulsion, and the acceleration virtual force includes acceleration virtual gravity and acceleration virtual repulsion.

[0190] According to the relationship between the potential energy field virtual force and the potential energy field, the position virtual repulsion It is expressed as:

[0191] (10);

[0192] in, represents the virtual repulsive potential energy coefficient of the position, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, represents the distance vector from the autonomous connected vehicle to the target point, || || represents the modulus of the distance vector from the autonomous connected vehicle to the target point.

[0193] The virtual gravity of the position It is expressed as:

[0194] (11);

[0195] in, represents the virtual gravitational potential energy coefficient of the position, Indicates the distance from the autonomous connected vehicle to the target point.

[0196] The velocity virtual repulsion It is expressed as:

[0197] (12);

[0198] in, represents the velocity virtual repulsive potential energy coefficient, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, represents the speed difference vector between the preceding vehicle and the vehicle itself, || || represents the modulus of the speed difference vector between the preceding vehicle and the vehicle itself.

[0199] The velocity is virtual gravity It is expressed as:

[0200] (13);

[0201] in, represents the velocity virtual gravitational potential energy coefficient, Indicates the speed difference between the desired speed and the current speed. represents the distance from the autonomous networked vehicle to the target point, tanh represents the hyperbolic tangent function, || || represents the modulus of the speed difference vector between the desired speed and the current speed.

[0202] The acceleration virtual repulsion It is expressed as:

[0203] (14);

[0204] in, represents the acceleration virtual repulsive potential energy coefficient, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, represents the acceleration difference vector between the preceding vehicle and the vehicle itself, || || represents the modulus of the acceleration difference vector between the preceding vehicle and the vehicle itself.

[0205] The acceleration virtual gravity It is expressed as:

[0206] ; (15);

[0207] in, represents the virtual gravitational potential energy coefficient of acceleration, Indicates the difference between the expected acceleration and the current acceleration. represents the distance from the autonomous connected vehicle to the target point, || || represents the modulus of the difference vector between the expected acceleration and the current acceleration.

[0208] The position, velocity and acceleration of the autonomous mobile agent are acquired in real time through the sensors of the autonomous mobile agent, and the interaction potential energy field function is defined;

[0209] Position virtual force, velocity virtual force and acceleration virtual force are introduced to improve the interaction potential energy field function.

[0210] According to the improved interaction potential energy field function, the evolution rules of the automatic mobile agent are constructed and the state of the automatic mobile agent is updated.

[0211] The evolution rules of the automatic mobile agent include acceleration rules, deceleration rules and position update rules;

[0212] Among them, the acceleration rule of the automatic mobile agent is expressed as:

[0213] ;

[0214] in, Indicates this vehicle The speed at time t, Indicates this vehicle The speed at time t, Indicates the maximum speed limit of the road. represents the vehicle calculated according to the double integral motion equation The acceleration of Indicates this vehicle Quality, represents the safe following distance, and F represents the total virtual force.

[0215] The deceleration rule of the autonomous mobile agent is: when the repulsive force generated by the virtual potential field of the current vehicle n+1 is greater than the gravitational force, the total virtual force on the vehicle n is expressed as repulsive force;

[0216] The position update rules of the automatic mobile agent are consistent with those of the artificial mobile agent.

[0217] The movement and motion path of the automatic mobile agent are controlled through the field force planning method, and the motion path of the artificial mobile agent is planned by utilizing the perception information sharing mechanism of the automatic mobile agent, combining the movement and motion path of the automatic mobile agent with the emotional clues of different granularities in the operation scenario of the artificial mobile agent.

[0218] Among them, the method of controlling the movement and movement path of the automatic mobile intelligent body through the field force planning method includes:

[0219] According to the improved interaction potential energy field function, the virtual forces on each autonomous mobile agent from other agents are calculated, including position virtual force, velocity virtual force and acceleration virtual force;

[0220] Perform vector synthesis of all virtual forces on each autonomous mobile agent to obtain the total virtual force of each autonomous mobile agent;

[0221] The total virtual force is used as a control signal to adjust the speed and direction of the autonomous mobile agent through a control algorithm.

[0222] Use the virtual force of potential energy field to plan the driving path of the autonomous mobile agent.

[0223] In summary, the present invention improves the authenticity and accuracy of the collaborative control of heterogeneous mobile agents by introducing emotion influencing factors, accurately quantifying the continuous dimension emotion values ​​and integrating them into the random deceleration probability function, making the behavior of artificial mobile agents closer to reality. At the same time, the method based on the virtual force of the potential energy field converts the position, velocity and acceleration information of the automatic mobile agent into a virtual field force, effectively controls its movement through field force planning, and enhances the autonomous decision-making and path planning capabilities of the automatic mobile agent. In response to the problem of collaborative control in complex environments, the present invention significantly improves the adaptability and overall intelligence of the collaborative control of heterogeneous mobile agents through emotion recognition, deceleration rules and potential energy field optimization.

[0224] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0225] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0226] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0227] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0228] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A method for cooperative control and evolution of heterogeneous mobile agents, characterized in that: include: Using continuous cellular automata to mix artificial mobile agent cells and automatic mobile agent cells according to a preset ratio, and generating artificial mobile agents and automatic mobile agents through an iterative process; Embedding an emotion recognition module in the artificial mobile agent to obtain emotion clues of different granularities based on fuzzy theory; According to the emotional clues of different granularities, constructing evolution rules of the artificial mobile agent and updating the state of the artificial mobile agent; The evolution rules of the artificial mobile agent include an acceleration rule, a deceleration rule, a random slowing-down rule introducing a random slowing-down probability function, and a position updating rule; Obtain the position, velocity and acceleration of the automatic mobile agent, model it as a virtual repulsive field and a virtual gravitational field, obtain the position, velocity and acceleration of the automatic mobile agent, construct the automatic mobile agent evolution rule based on the improved interaction potential energy field function, and update the state of the automatic mobile agent; The automatic mobile agent evolution rules include acceleration rules, deceleration rules and position update rules; The acceleration rule is used to adjust the speed of the automatic mobile agent or the artificial mobile agent; The deceleration rule is used to avoid collision between the automatic mobile agent and the artificial mobile agent; The random slowing down rule introducing the random slowing down probability function is used to randomly slow down the artificial mobile agent based on the emotion influencing factor and the random slowing down probability function; The position update rule is used to determine the position of the automatic mobile agent or the artificial mobile agent; The movement and motion path of the automatic mobile agent are controlled through the field force planning method, and the motion path of the artificial mobile agent is planned by utilizing the perception information sharing mechanism of the automatic mobile agent, combining the movement and motion path of the automatic mobile agent with the emotional clues of different granularities in the operation scenario of the artificial mobile agent.

2. The method for cooperative control and evolution of heterogeneous mobile agents according to claim 1, characterized in that: The emotion recognition module includes an emotion perception layer and a feature fusion network. The emotion perception layer adopts a dual-channel fusion attention mechanism network structure. According to fuzzy theory, the method of obtaining emotion clues of different granularities includes: The emotion perception layer is used to capture and extract character emotion clues and scene emotion clues in the running scene through the character channel and the scene channel respectively to form a preliminary feature vector; The emotion values ​​of continuous dimensions in the preliminary feature vector are quantified using fuzzy theory to obtain emotion influencing factors. The emotion influencing factors are input into the feature fusion network for fusion processing to obtain emotion clues of different granularities.

3. The method for cooperative control and evolution of heterogeneous mobile agents according to claim 2, characterized in that: The method of quantifying the emotion values ​​of the continuous dimensions in the preliminary feature vector using fuzzy theory to obtain the emotion influencing factors includes: Based on the Mamdani model, the emotional pleasure, emotional arousal and emotional dominance are taken as input variables, and the emotional influencing factors are taken as output variables to construct a three-value input and single-value output emotional fuzzy reasoning model. According to the emotion measurement values ​​of emotion pleasure, emotion arousal and emotion dominance, input variable fuzzy sets are constructed respectively, including emotion pleasure fuzzy set 1, emotion pleasure fuzzy set 2 and emotion pleasure fuzzy set 3, emotion arousal fuzzy set 1, emotion arousal fuzzy set 2 and emotion arousal fuzzy set 3, and emotion dominance fuzzy set 1, emotion dominance fuzzy set 2 and emotion dominance fuzzy set 3; Constructing output variable fuzzy sets, including emotion factor fuzzy set 1, emotion factor fuzzy set 2 and emotion factor fuzzy set 3; Using Gaussian membership function, the input variable fuzzy set and the output variable fuzzy set are fuzzified; Analyze the data samples of emotional pleasure, emotional arousal, and emotional dominance in the Emotic dataset to obtain an artificial empirical rule library that can reflect the relationship between input variables and output variables; According to the artificial experience rule base, the emotion representation fuzzy rule matrix is ​​constructed; The continuous sentiment values ​​in the preliminary feature vector are converted into the membership of the input set fuzzy set through fuzzification; According to the emotion representation fuzzy rule matrix, the input fuzzy set is mapped to the output fuzzy set using a fuzzy reasoner, and the final output fuzzy set is obtained by calculation; The final output fuzzy set is defuzzified using the centroid method to obtain the emotion influence factor.

4. The method for cooperative control and evolution of heterogeneous mobile agents according to claim 3 is characterized in that: According to the artificial experience rule base, the emotion representation fuzzy rule matrix is ​​constructed, and the constructed emotion representation fuzzy rule matrix is ​​expressed as: （1）； Among them, the first three columns of the emotion representation fuzzy rule matrix represent the fuzzy set indexes corresponding to the input variables emotion pleasure, emotion arousal and emotion dominance, respectively; the fourth column represents the fuzzy set indexes corresponding to the output variable emotion influencing factor; the fifth column represents the rule weights of the emotion representation fuzzy rule matrix; the sixth column represents the connectives of the rules of the emotion representation fuzzy rule matrix; when the rule weight is 1, the logical connective is "and", and when the rule weight is 2, the logical connective is "or"; Among them, the rules of the first row of the emotion representation fuzzy rule matrix are: when the input emotion pleasure belongs to the emotion pleasure fuzzy set 1, the input emotion arousal belongs to the emotion arousal fuzzy set 3, and the input emotion dominance belongs to the emotion dominance fuzzy set 3, the output emotion factor fuzzy set 1, the rule weight of the first row is 1, and the logical connective is "and"; The rules in the second row of the fuzzy rule matrix are: when the input emotional pleasure belongs to the emotional pleasure fuzzy set 2, the input emotional arousal belongs to the emotional arousal fuzzy set 1, and the input emotional dominance belongs to the emotional dominance fuzzy set 3, the output emotional factor fuzzy set 2, the rule weight in the second row is 1, and the logical connective is "and"; The rule in the third row of the fuzzy rule matrix is: when the input emotional pleasantness belongs to emotional pleasantness fuzzy set one, the input emotional arousal belongs to emotional arousal fuzzy set two, and the input emotional dominance belongs to emotional dominance fuzzy set one, the output emotional factor fuzzy set three, the rule weight in the third row is 1, and the logical connective is "and".

5. The method for cooperative control and evolution of heterogeneous mobile agents according to claim 1, characterized in that:When the operation scenario is road traffic flow, the continuous cellular automaton uses vehicles as cells and generates manually driven vehicles and automatically connected vehicles through an iterative process. The state information of the continuous cellular automaton , expressed as: （2）； in, , and Respectively indicate this vehicle The position, velocity and acceleration at the current time t, Indicates this vehicle The ideal state vector of the vehicle is Indicates this vehicle The driver's emotional state vector, , , Respectively indicate this vehicle The expected speed of the vehicle, the vehicle The expected following distance of the vehicle and the vehicle The vehicle safety time interval, , and Respectively indicate at the current moment This car measures of drivers’ emotional pleasure, emotional arousal, and emotional dominance; The continuous cellular automaton uses vehicles as cells and generates manually driven vehicles and automatically connected vehicles through an iterative process. It is expressed as: （3）； in, represents the maximum acceleration of the vehicle, represents the vehicle acceleration index, and Respectively indicate the front vehicle and this car The relative distance and relative speed, Indicates the absolute value of the vehicle's comfortable deceleration. Indicates the static safety distance, Indicates the vehicle ahead The velocity at the current time t.

6. The method for cooperative control and evolution of heterogeneous mobile agents according to claim 5, characterized in that: The final output fuzzy set is defuzzified using the centroid method to obtain the emotional impact factor, which is expressed as: （4）； in, , and Respectively indicate at the current moment This car Measures of driver emotional pleasure, emotional arousal, and emotional dominance.

7. The method for cooperative control and evolution of heterogeneous mobile agents according to claim 6, characterized in that: The evolution rules of the artificial mobile agent include an acceleration rule, a deceleration rule, a random slowing-down rule introducing a random slowing-down probability function, and a position update rule; wherein the acceleration rule of the artificial mobile agent is expressed as: （5）； in, Indicates this vehicle The velocity at the current time t, Indicates this vehicle The velocity at the current time t, Indicates this vehicle The acceleration at the current time t, Indicates the maximum speed limit of the road; The deceleration rule of the artificial mobile agent is expressed as: （6）； in, Indicates safe following distance. Indicates the vehicle ahead and this car The relative distance Indicates the static safety distance; Based on this car Driver's emotional influence factor and random slowing probability function, the vehicle The driver has The probability of random deceleration is, where represents the random slowing probability of vehicle n, and the random slowing rule introducing the random slowing probability function is expressed as: （7）； in, Indicates this vehicle The velocity at the current time t, Indicates this vehicle The velocity at the current time t, Indicates the deceleration value; The position of vehicle n at the current time t and speed Together determine the position of vehicle n at the next moment t+1 ; The expression of the position update rule of the artificial mobile agent is: （8）； in, Indicates in this vehicle The position at the next moment (t+1), Indicates this vehicle At the current time t, Indicates this vehicle The velocity at the current time t.

8. The method for cooperative control and evolution of heterogeneous mobile agents according to claim 5, characterized in that: The evolution rules of the automatic mobile agent include acceleration rules, deceleration rules and position update rules; Among them, the acceleration rule of the automatic mobile agent is expressed as: （9）； in, Indicates this vehicle The velocity at the current time t, Indicates this vehicle The velocity at the current time t, Indicates the maximum speed limit of the road. represents the vehicle calculated according to the double integral motion equation The acceleration of Indicates this vehicle Quality, represents the safe following distance, and F represents the total virtual force; The deceleration rule of the autonomous mobile agent is: when the repulsive force generated by the virtual potential field of the current vehicle n+1 is greater than the gravitational force, the total virtual force on the vehicle n is expressed as repulsive force; The position update rules of the automatic mobile agent are consistent with those of the artificial mobile agent.

9. The method for cooperative control and evolution of heterogeneous mobile agents according to claim 1, characterized in that: The method of controlling the motion and motion path of an automatic mobile intelligent body by a field force planning method includes: According to the improved interaction potential energy field function, the virtual forces on each autonomous mobile agent from other agents are calculated, including position virtual force, velocity virtual force and acceleration virtual force; Perform vector synthesis of all virtual forces on each autonomous mobile agent to obtain the total virtual force of each autonomous mobile agent; The total virtual force is used as a control signal to adjust the speed and direction of the autonomous mobile agent through a control algorithm. Use the virtual force of potential energy field to plan the driving path of the autonomous mobile agent.

10. The method for cooperative control and evolution of heterogeneous mobile agents according to claim 9, characterized in that: The position virtual force includes position virtual gravity and position virtual repulsion, the velocity virtual force includes velocity virtual gravity and velocity virtual repulsion, and the acceleration virtual force includes acceleration virtual gravity and acceleration virtual repulsion; The virtual repulsion force It is expressed as: （10）； in, represents the virtual repulsive potential energy coefficient of the position, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, represents the distance vector from the autonomous connected vehicle to the target point, || || represents the modulus of the distance vector from the autonomous networked vehicle to the target point; The virtual gravity of the position It is expressed as: （11）； in, represents the virtual gravitational potential energy coefficient of the position, Indicates the distance from the autonomous connected vehicle to the target point; The velocity virtual repulsion It is expressed as: （12）； in, represents the velocity virtual repulsive potential energy coefficient, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, represents the speed difference vector between the preceding vehicle and the vehicle itself, || || represents the modulus of the speed difference vector between the preceding vehicle and the vehicle itself; The velocity is virtual gravity It is expressed as: （13）； in, represents the velocity virtual gravitational potential energy coefficient, Indicates the speed difference between the desired speed and the current speed. represents the distance from the autonomous networked vehicle to the target point, tanh represents the hyperbolic tangent function, || || represents the modulus of the speed difference vector between the expected speed and the current speed; The acceleration virtual repulsion It is expressed as: （14）； in, represents the acceleration virtual repulsive potential energy coefficient, represents the distance from the autonomous networked vehicle to the target point, represents the influence range of the repulsive potential field, represents the acceleration difference vector between the preceding vehicle and the vehicle itself, || || represents the modulus of the acceleration difference vector between the preceding vehicle and the vehicle itself; The acceleration virtual gravity It is expressed as: （15）； in, represents the virtual gravitational potential energy coefficient of acceleration, Indicates the difference between the expected acceleration and the current acceleration. represents the distance from the autonomous connected vehicle to the target point, || || represents the modulus of the difference vector between the expected acceleration and the current acceleration.

Images