Longitudinal formation control method for multiple autonomous platforms under dynamic communication topology

Abstract

The application discloses a kind of longitudinal formation control method of multiple autonomous platforms under dynamic communication topology, first constructs objective function, then multiple-platform formation control problem is described as optimization problem P1 of objective function, solves optimization problem P1 based on distributed model predictive control method, obtains optimal control sequence, realizes autonomous platform longitudinal formation control under dynamic communication topology condition.The application can still successfully carry out multiple-platform longitudinal formation control under the condition that communication topology changes in real time, while simulation test verifies the longitudinal formation control of autonomous platform under dynamic communication topology, realizes the engine torque tracking control of leader autonomous platform, the distance control between autonomous platform and front platform, the speed and engine torque of following autonomous platform servo control, can handle the state constraints, control input constraints and collision avoidance constraints inevitably in practical application, and optimization is carried out to performance index, with good practicability.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of longitudinal control of multiple autonomous platforms under dynamic communication topology, and particularly relates to a longitudinal formation control method of multiple autonomous platforms under dynamic communication topology. BACKGROUND

[0002] Longitudinal formation control of multiple autonomous platforms is an advanced traffic management technology, aiming to enable a group of autonomous platforms to form a tight formation on the road and maintain the desired spacing and the same speed through the communication and cooperation between intelligent systems and autonomous platforms. First, each platform in the multiple autonomous platforms perceives the position and other information of the front and surrounding autonomous platforms in real time through the sensor system carried by the platform, and then calculates the control input according to the algorithm. The autonomous platform adjusts its speed and direction through the automatic control system to maintain the desired spacing between the front and rear autonomous platforms, thereby forming a stable formation state. At the same time, the autonomous platforms need to maintain the same speed, and the speed synchronization and coordination between the autonomous platforms are realized through the communication system between the autonomous platforms to ensure the stability and safety of the formation operation.

[0003] The control of multiple autonomous platforms is usually divided into centralized control and distributed control. Centralized longitudinal formation control of multiple autonomous platforms refers to the formation and control of the entire multiple autonomous platforms by a central controller, which is responsible for guiding the driving speed and spacing of all autonomous platforms to realize the longitudinal formation operation of multiple autonomous platforms. The centralized controller can globally consider the running state and road conditions of the multiple autonomous platforms to realize more flexible scheduling and formation control. However, the reliability and processing capacity of the central controller are required to be high, and once the controller fails, the operation of the entire multiple autonomous platforms may be seriously affected, and good communication infrastructure and reliable communication connection are needed to transmit control instructions and autonomous platform state feedback information in time. Distributed longitudinal formation control of multiple autonomous platforms is to realize the autonomous adjustment and formation control of multiple autonomous platforms through the communication and coordination between autonomous platforms. The distributed control method has high fault tolerance, and each platform can make independent decisions and adjustments, and a platform failure will not have a great impact on other autonomous platforms, which reduces the complexity and centralization of the system. However, the distributed control method has high requirements for the communication and algorithm design of autonomous platforms, and needs to ensure the accuracy and reliability of information transmission and coordination between autonomous platforms. The negotiation and decision-making of distributed control design of autonomous platforms may cause delay and inconsistency problems in the formation process.

[0004] The formation control of the autonomous control multi-platform needs information interaction between the autonomous platforms, whether the control is centralized or distributed. Nowadays, based on the new communication technology represented by 5G, the intelligent transportation system also develops rapidly, and the existing research results all assume that the multi-autonomous platforms have a fixed communication topology, but the network communication topology occasionally fails in the actual environment, and once the failure occurs, the control method will no longer be applicable, resulting in the failure of the autonomous platform column control. Therefore, the existing research is not suitable for solving this actual situation.

[0005] The model predictive control is a kind of control that can process various constraint conditions such as input-output constraints, state constraints and the like to ensure that the system runs within the safety boundary. The distributed model predictive control combines the advantages of the model predictive control and the distributed control, can optimize the performance index under the premise of meeting the constraint conditions, and is an online solution method that can cope with the real-time changes and new demands of the system in the task process. SUMMARY

[0006] The purpose of the present application is to provide a kind of longitudinal formation control method of multi-autonomous platform under dynamic communication topology, to solve the above problems, adopt distributed model predictive control to realize the longitudinal formation control of autonomous platform under dynamic communication topology. The autonomous platform can be unmanned autonomous platform.

[0007] The present application is mainly realized by the following technical solutions:

[0008] A kind of longitudinal formation control method of multi-autonomous platform under dynamic communication topology, based on distributed model predictive control to realize the longitudinal formation control of multi-autonomous platform, including the following steps:

[0009] Step S1: construct target function based on control target:

[0010]

[0011] Wherein: j=1 indicates that it is the lead autonomous platform, at this time c 1j =1,c 2j =0;

[0012] j≠1 represents that it is the following autonomous platform, at this time c 1j =0,c 2j =1;

[0013] c 1j It is the coefficient of the difference item of engine torque and expected engine torque of the lead autonomous platform;

[0014] c 2j It is the coefficient of the difference item of engine torque of the following platform and the preceding platform;

[0015] d is the expected distance of the preceding platform and the following platform.

[0016] Qi-Q4 are weight matrices of each term in the objective function, respectively;

[0017] T q,Lead is the engine torque of the lead autonomous platform;

[0018] is the desired engine torque of the lead autonomous platform;

[0019] T q,j-1 is the engine torque of the j-1th autonomous platform;

[0020] T q,j is the engine torque of the jth autonomous platform;

[0021] v j-1 is the speed of the j-1th autonomous platform;

[0022] v j is the speed of the jth autonomous platform;

[0023] s j-1 is the longitudinal distance of the j-1th autonomous platform;

[0024] s j is the longitudinal distance of the jth autonomous platform;

[0025] u j is the control input;

[0026] At each sampling time k, the communication state is detected, and α(k), β(k) represent the communication indication signals; α(k) = 1 indicates that the autonomous platform can interact with the front platform at time k;

[0027] α(k) = 0 indicates that the autonomous platform cannot communicate with the front platform at time k;

[0028] β(k) = 1 indicates that the autonomous platform can interact with the lead autonomous platform at time k; β(k) = 0 indicates the case that the autonomous platform cannot communicate with the lead autonomous platform;

[0029] Step S2: describe the multi-autonomous platform control problem as an optimization problem P1 of an objective function:

[0030]

[0031] Define the state vector x(k) = [s v T q ] T , then the state update equation is:

[0032] x j '(t) = ax j (k) + buj (k) (7)

[0033] wherein: a and b are system matrix and control matrix of the continuous system respectively;

[0034] State constraint:

[0035] x j (k)∈X j (8)

[0036] wherein: X j is the state constraint set;

[0037] Control input constraint:

[0038] u j (k)∈U j (9)

[0039] wherein: U j is the control quantity constraint set;

[0040] State initialization equation:

[0041] x j (0|k)=x j (k) (10)

[0042] Safe distance constraint between autonomous platforms:

[0043] ||s j-1 -s j ||>R (11)

[0044] wherein: R is the safe distance;

[0045] Step S3: solving the optimization problem P1 based on the distributed model prediction control method, obtaining an optimal control sequence, and realizing longitudinal formation control of the autonomous platforms based on the optimal control sequence.

[0046] In order to better realize the present application, further, in the step S2, T is taken as a sampling period, and the following discrete system matrix A and B are obtained:

[0047]

[0048] wherein: a and b are system matrix and control matrix of the continuous system respectively;

[0049] I is a unit matrix;

[0050] T is a sampling period.

[0051] Therefore, the continuous system can be discretized as follows:

[0052] x j(k+1) = Ax j (k) + Bu j (k).

[0053] In order to better realize the present application, further, in the step S1, the target is to track the speed and engine torque of the predecessor while maintaining the predefined distance, and to consider the engine torque tracking of the leading autonomous platform, the leading autonomous platform considers the engine torque tracking control, and constructs the control target:

[0054]

[0055] Wherein: is the desired engine torque of the autonomous platform.

[0056] In order to better realize the present application, further, in the step S1, the control target further includes the control of the fuel performance index.

[0057] In order to better realize the present application, further, the cooperative formation control applied to multiple unmanned autonomous platforms includes the following steps:

[0058] Step A1: initialize state information x j (0|k) = x j (k) ;

[0059] Step A2: determine the communication indication signals alpha(k), beta(k) based on the detection of the communication condition by the airborne equipment, and receive the state information of the front unmanned platform or the leading platform;

[0060] Step A3: solve the optimization problem P1 to obtain the optimal control sequence;

[0061] Step A4: apply the first control quantity in the optimal control sequence to the current system, encapsulate the state information of itself, and transmit it to the rear unmanned platform;

[0062] Step A5: determine whether the end condition is met, if not, return to step A1, if yes, end.

[0063] In order to better realize the present application, further, the unmanned autonomous platform determines Q1, Q2, Q3, Q4, R, d, Parameters in the offline stage.

[0064] The beneficial effects of the present application are as follows:

[0065] (1) The application can be applied to the longitudinal formation control of multiple autonomous platforms in the case of dynamic communication topology, and adopts distributed model predictive control, which can process the inevitable state constraints, control input constraints and collision avoidance constraints in actual application, and optimizes the performance index. The application can still successfully perform the formation control of multiple autonomous platforms in the case of real-time changes of communication topology, and the longitudinal formation control of 5 platforms under dynamic communication topology is verified by simulation test, realizing the engine torque tracking control of the leading autonomous platform, the distance control between the autonomous platform and the front platform, and the speed and engine torque control of the following autonomous platform, which has good practicability.

[0066] (2) The application is based on distributed model predictive control, and each autonomous platform independently uses local model and prediction information to realize the formation control of multiple autonomous platforms. Model predictive control utilizes the dynamic model of the system to formulate the optimal control strategy by predicting the future state. This enables the system to optimize within the prediction range to achieve better performance. And model predictive control can handle nonlinear and complex systems, because it can use various models to describe the dynamic characteristics of the system, and can optimize for different models, which has good practicability. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 Communication topology diagram of multiple autonomous platforms for perfect communication;

[0068] Figure 2 Communication topology diagram of multiple autonomous platforms for only communicating with the leading autonomous platform;

[0069] Figure 3 Communication topology diagram of multiple autonomous platforms for only communicating with the front platform;

[0070] Figure 4 Communication topology diagram of multiple autonomous platforms for no communication at all;

[0071] Figure 5 Structure diagram of the distributed longitudinal control system of multiple autonomous platforms;

[0072] Figure 6 Motion trajectory curve of the unmanned autonomous platform;

[0073] Figure 7 Distance change curve between unmanned autonomous platforms;

[0074] Figure 8 Speed change curve of the unmanned autonomous platform;

[0075] Figure 9 Engine torque change curve of the leading platform;

[0076] Figure 10The torque variation curves for all unmanned autonomous platforms;

[0077] Figure 11 To control the input variation curve. Detailed Implementation

[0078] Example 1:

[0079] A method for longitudinal formation control of multiple autonomous platforms under dynamic communication topology includes the following steps:

[0080] (1) Communication Topology

[0081] like Figures 1-4 As shown, this invention assumes four communication topologies, including four scenarios: intact communication, communication only with the team leader's autonomous platform, communication only with the front platform, and no communication at all. Figure 1 As shown, in a multi-autonomous platform scenario, when each autonomous platform is functioning correctly, it can receive position, speed, and engine torque information from the leader and front platforms via communication and onboard sensors. However, considering real-world multi-autonomous platform driving scenarios where various communication issues may arise, potentially leading to information exchange failures or communication link interruptions, [further details will be needed]. Figure 1 Communication status degraded to Figure 2 The communication scenario shown indicates that each autonomous platform can only exchange information with the team leader platform, and has degenerated into... Figure 3 As shown in the communication diagram, the autonomous platform can only exchange information with the preceding platform. For example... Figure 4 As shown, this indicates a complete communication failure, and the autonomous platform can only perceive the speed and position information of the preceding platform through onboard sensors.

[0082] (2) Problem Modeling

[0083] Consider the following form of autonomous platform motion equation model:

[0084]

[0085] Where S j (t) represents the longitudinal distance of the j-th autonomous platform, v j (t) represents velocity, T q,j Represents the torque of the engine on the independent platform; Let be the derivative of the distance traveled by the j-th autonomous platform at time t;

[0086] Let be the derivative of the velocity of the j-th autonomous platform at time t;

[0087] Let be the derivative of the engine torque of the j-th autonomous platform at time t;

[0088] τ is the force coefficient;

[0089] i o,j It is a scalar coefficient value;

[0090] η j For mechanical efficiency;

[0091] m j For the quality of the j-th autonomous platform;

[0092] r j For the structural parameters of the autonomous platform;

[0093] u j For control input;

[0094] The above model can be rewritten in the following form:

[0095]

[0096] in Represents the state vector, u j Represents the control input, and the following system matrices a and b:

[0097]

[0098] Discretizing the above continuous system equations using the zero-order preservation method, they can be transformed into the following form:

[0099] x j (k+1)=Ax j (k)+Bu j (k)(3)

[0100] Taking T as the sampling period, the following discretized system matrices A and B can be obtained:

[0101]

[0102] Where: I is the identity matrix;

[0103] T is the sampling period.

[0104] (3) Design of multi-platform longitudinal formation control algorithm based on distributed model predictive control

[0105] In a distributed control architecture, a distributed model predictive controller transmits information between controllers, allowing local controllers to gain some understanding of the behavior of others. For example... Figure 5 As shown, in the multi-autonomous platform system of the present invention, these single-machine subsystems will make autonomous decisions and controls based on their respective local models and prediction information. In order to achieve coordination and overall optimization, these subsystems need to interact to jointly achieve the system-level control objectives.

[0106] An important goal of the multi-autonomous platform system control is to track the speed and engine torque of the predecessor while maintaining a predefined distance, also considering the engine torque tracking of the leading autonomous platform, which considers the engine torque tracking control, defining The desired engine torque curve for the autonomous platform, so the following desired target is constructed:

[0107]

[0108] Where: s j-1 is the longitudinal distance of the j-1th autonomous platform;

[0109] s j is the longitudinal distance of the jth autonomous platform;

[0110] v j-1 is the speed of the j-1th autonomous platform;

[0111] v j is the speed of the jth autonomous platform;

[0112] T q,j-1 is the engine torque of the j-1th autonomous platform;

[0113] T q,j is the engine torque of the jth autonomous platform;

[0114] T q,Lead is the engine torque of the leading autonomous platform;

[0115] is the desired engine torque of the leading autonomous platform;

[0116] At the same time, in order to save the fuel consumption of heavy autonomous platform application, the fuel performance index is also considered by the present application, so the objective function of the distributed model predictive control can be constructed in the following form:

[0117]

[0118] Where j = 1 represents the leading autonomous platform, at this time c 1j = 1, c 2j = 0, when j ≠ 1 represents the following autonomous platform, at this time c 1j = 0, c 2j = 1.

[0119] At each sampling time k, the communication state is detected by the platform-mounted device, and the communication indication signal is represented by α(k), β(k):

[0120] α(k) = 1 represents the case that the autonomous platform can interact with the front platform at time k,

[0121] α(k) = 0 represents that the follower platform cannot communicate with the leader platform,

[0122] β(k) = 1 represents that the follower platform can communicate with the leader platform at time k,

[0123] β(k) = 0 represents that the follower platform cannot communicate with the leader platform.

[0124] Based on the above, the longitudinal formation control problem of multiple platforms based on model predictive control can be described in the form of the following problem P1:

[0125]

[0126] s.t.x j (k+1) = Ax j (k) + Bu j (k) (7) x j (k) ∈ X j (8)

[0127] u j (k) ∈ U j (9)

[0128] x j (0|k) = x j (k) (10)

[0129] ||s j-1 -s j || > R (11)

[0130] where formula (7) is a state update equation, formulas (8)-(9) represent state constraints and control input constraints, formula (10) is state initialization, and formula (11) is a safety distance constraint between autonomous platforms.

[0131] Example 2:

[0132] A longitudinal formation control method for multiple autonomous platforms in a dynamic communication topology, to realize the cooperative formation control of multiple unmanned autonomous platforms, first, the problem is modeled, and the problem is constructed as a mathematical solving problem, which has been described in the above research content, and the solving steps of the problem are introduced in detail below taking the jth unmanned autonomous platform as an example.

[0133] Offline stage: determine the system parameters and control constant parameters, including Q1, Q2, Q3, Q4, R, d,

[0134] Online stage: the specific steps are as follows:

[0135] ① Initialize state information xj (0|k)=x j (k);

[0136] ②Based on the airborne equipment to detect the communication situation, determine the communication instruction signal α(k), β(k), receive the front unmanned platform or the state information of the leader;

[0137] ③Solve the optimization problem P1, obtain the optimal control sequence;

[0138] ④The first control quantity in the optimal control sequence is applied to the current system, and the state information of itself is encapsulated and transmitted to the rear unmanned platform;

[0139] ⑤Judge the simulation end condition, if not satisfied, return to step 1, if satisfied, end.

[0140] As Figure 6 shown, it can be seen that the five trajectories have no intersection, which shows that the collision avoidance between unmanned platforms is realized, the five curves are basically parallel, and the spacing is also basically equal, and the longitudinal formation control under the dynamic communication topology is realized under the algorithm. As Figure 7 shown, the distance between unmanned autonomous platforms is controlled to the expected 25 spacing. As Figure 8 shown, it can be seen that the speed values between the five unmanned platforms are very small, and the change trend is consistent, which shows that the speed coordination control under the dynamic communication topology is realized under the algorithm. As Figure 9 shown, the solid line is the expected change curve, and the dotted line is the actual value under the algorithm control, it can be seen that the engine torque control effect of the leader is very good. As Figure 10 shown, the unmanned autonomous platform will track the engine torque of the front unmanned platform or the leader, which can verify the effectiveness of the algorithm from the figure. As Figure 11 shown, the control input change satisfies the constraint condition [-120, 120].

[0141] The above is only the preferred embodiment of the present application, not any form of the application of the above-described embodiments of the application, any simple modification, equivalent change according to the technical essence of the application, falls within the scope of the present application.

Claims

1. A method for longitudinal formation control of multiple autonomous platforms in a dynamic communication topology, characterized in that, The longitudinal formation control of multiple autonomous platforms is realized based on a distributed model predictive control, including the following steps. Step S1: constructing a target function based on a control target: wherein: j = 1 indicates it is a lead autonomous platform, in which case c 1j = 1, c 2j = 0; j≠1 represents that it is a following autonomous platform, at this time c 1j = 0, c 2j = 1; c 1j a coefficient for a difference between engine torque of the lead autonomous platform and a desired engine torque c 2j is a coefficient for the engine torque difference term between the rear platform and the front platform; d is an expected distance between a front platform and a rear platform; Q1-Q4 are weight matrices of each term in the target function; T q,Lead Engine torque for the lead autonomous platform; a desired engine torque for the lead autonomous platform; T q,j-1 Tj-1is the engine torque for the j-1 autonomous platform; T q,j Tj is the engine torque for the jth autonomous platform; v j-1 vj-1is the velocity of the j-1th autonomous platform; v j vj is the velocity of the jth autonomous platform; s j-1 is the longitudinal range for the (j-1)th autonomous platform; s j longitudinal distance for the jth autonomous platform; u j u for control input; At each sampling time k, a communication state is detected, and alpha(k) and beta(k) represent communication indication signals; Alpha(k)=1 indicates that the autonomous platform can interact with the front platform at time k; Alpha(k)=0 indicates that the autonomous platform cannot communicate with the front platform at time k; Beta(k)=1 indicates that the autonomous platform can interact with the leading autonomous platform at time k; Beta(k)=0 indicates that the autonomous platform cannot communicate with the leading autonomous platform; Step S2: describing the multiple autonomous platform control problem as an optimization problem P1 of the target function: Define the state vector x(k) = [s vT q ] T Then the state update equation: x j '(t) = ax j (k) + bu j (k) (7) Wherein: a and b are system matrix and control matrix of a continuous system respectively; State constraint: x j (k) ∈ X j (8) wherein: X j is a set of state constraints; Control input constraint: u j (k)∈U j (9) where: U j is the set of control constraints; State initialization equation: x j (0 | k) = x j (k) (10) Safety distance constraint between autonomous platforms: ||s j-1 -s j ||>R (11) Wherein: R is a safety distance; Step S3: solving the optimization problem P1 based on a distributed model predictive control method to obtain an optimal control sequence, and realizing longitudinal formation control of the autonomous platform based on the optimal control sequence. 2.The longitudinal formation control method for multiple autonomous platforms in a dynamic communication topology according to claim 1, wherein, In the step S2, taking T as a sampling period, the following discrete system matrices A and B are obtained: Wherein: a and b are system matrix and control matrix of a continuous system respectively; I is an identity matrix; T is a sampling period. 3.The longitudinal formation control method for multiple autonomous platforms in a dynamic communication topology according to claim 1, wherein, In the step S1, the control target is to track the speed and engine torque of the predecessor while maintaining a predefined distance, and to consider engine torque tracking of the leading autonomous platform, and the leading autonomous platform considers engine torque tracking control. wherein: is the desired engine torque for the autonomous platform.

4. The longitudinal formation control method of multiple autonomous platforms under dynamic communication topology according to claim 3, characterized in that, In the step S1, the control target further includes a fuel performance index control.

5. The longitudinal formation control method of multiple autonomous platforms under dynamic communication topology according to any one of claims 1-4, characterized in that, The collaborative formation control of multiple unmanned autonomous platforms is applied, including the following steps. Step A1 : initialization of state information x j (0 | k) = x j (k); Step A2: detecting a communication state based on an airborne device, determining communication indication signals alpha(k) and beta(k), and receiving state information of a front unmanned platform or a leading platform; Step A3: solving the optimization problem P1 to obtain an optimal control sequence; Step A4: applying the first control amount in the optimal control sequence to the current system, encapsulating state information of the autonomous platform, and transmitting the state information to a rear unmanned platform; Step A5: determining whether an end condition is met, if not, returning to step A1, and if yes, ending.

6. The longitudinal formation control method of multiple autonomous platforms under dynamic communication topology according to claim 5, characterized in that, The unmanned autonomous platform determines Q1, Q2, Q3, Q4, R, d, parameters in an offline phase.

Images