Unmanned vehicle queue longitudinal control method considering communication delay problem

Abstract

The invention relates to the technical field of autonomous vehicle control, in particular to an unmanned vehicle queue longitudinal control method considering a communication delay problem, which comprises the following steps: firstly, based on a vehicle queue longitudinal kinematics model, considering the delay problem of a vehicle execution system, adding a first-order inertial link and establishing a longitudinal controller discrete state space equation; a Kalman filtering algorithm is improved by introducing Sage-Husa filtering, and an AEKF algorithm is designed to estimate vehicle state information in a queue. And the stability of the system is deduced and verified through the stability analysis of the Lyapunov-Krasovskii method and the stability margin of the LMI. And designing an upper-layer controller of the SMC algorithm based on the estimated vehicle state information, and outputting expected acceleration. And the lower layer PID controller outputs the driving torque and the braking pressure based on the expected acceleration, and queue following vehicle longitudinal control is achieved. According to the method, when the problem of communication delay is solved, the control precision can be improved, the performance reduction is reduced, and the longitudinal stability of the unmanned vehicle queue in an unstable communication environment is improved.

Description

Technical Field

[0001] This invention relates to the field of autonomous vehicle control technology, and more specifically to a longitudinal control method for unmanned vehicle platoons that takes into account communication delay issues. Background Technology

[0002] In recent years, with the rapid development of autonomous driving technology, the autonomous control of unmanned vehicles has received widespread attention. Queue control of unmanned vehicles can effectively improve task execution efficiency, enhance the stability and safety of the control system, and improve overall coordination capabilities.

[0003] In platoon control, vehicle-to-vehicle communication allows unmanned vehicles in the platoon to share information such as position and speed, enabling platoon control and reducing the occurrence of accidents during operation. However, during signal transmission between vehicles in the platoon, adverse weather conditions and signal interference can cause communication delays, data packet loss, and external interference, disrupting the normal operation of unmanned vehicles, reducing the stability and safety of platoon operation, and increasing energy consumption.

[0004] While significant progress has been made both domestically and internationally in addressing the communication delay issues in queuing longitudinal control, limitations remain. Existing research typically designs solutions based on fixed-frequency communication delays, but in practical applications, these problems occur randomly, exhibiting uncertainty and dynamism. These issues affect the accuracy of vehicle status information, reduce queuing stability, and hinder adaptation to complex and ever-changing driving environments. Summary of the Invention

[0005] The purpose of this invention is to provide a longitudinal control method for unmanned vehicle queues that takes into account communication delay issues, aiming to solve the problem of random communication delay in vehicle queue control under V2V communication environment.

[0006] To achieve the above objectives, the present invention provides a longitudinal control method for unmanned vehicle platoons that considers communication delay issues, comprising the following steps:

[0007] Step 1: Based on the longitudinal kinematic model of the vehicle platoon, considering the delay problem of the vehicle execution system, a first-order inertial element is added to establish the discrete state-space equation of the longitudinal controller;

[0008] Step 2: Introduce the Sage-Husa filter to improve the Kalman filter algorithm, and design the AEKF algorithm to estimate the vehicle state information in the queue;

[0009] Step 3: Prove the system stability based on stability analysis using the Lyapunov–Krasovskii method and the derivation of the LMI stability margin;

[0010] Step 4: Design the upper-level controller of the SMC algorithm based on the estimated vehicle state information, and output the desired acceleration;

[0011] Step 5: The lower-level PID controller outputs driving torque and braking pressure based on the desired acceleration to achieve longitudinal control of the convoy following the vehicle.

[0012] Optionally, in step 1, a fixed headway strategy is adopted as the vehicle platoon spacing strategy, and a longitudinal kinematic model of the vehicle platoon is established. The discrete time-domain relationship is as follows:

[0013]

[0014] in, For vehicle spacing error, To follow the relative speed of the vehicle in front, To follow the vehicle's actual acceleration, For the desired acceleration, This represents the first-order system gain, with a value of 1. The time constant of the inertial element, The current moment of the system. For the next moment of the system, This represents the system sampling time.

[0015] Selecting the state vector The discrete state-space equation of the system is:

[0016]

[0017] in, .

[0018] Optionally, the state prediction equation in step 2 is:

[0019]

[0020]

[0021] in, for The posterior error covariance matrix at time t. This is the noise covariance matrix in the state estimation process;

[0022] The measurement update equation is:

[0023]

[0024] in, Here is the Kalman gain matrix. To measure the noise covariance matrix, To observe the residuals, It is the identity matrix;

[0025] After introducing the Sage-Husa filter to improve the Kalman filter algorithm, the process noise covariance matrix Q and the observation noise covariance matrix R are estimated, and the update form is as follows:

[0026]

[0027]

[0028] in, The impact factor, which is weighted by the index, updates more slowly when it is closer to 1. These are small positive numbers used to maintain positive definiteness; To observe residuals; To predict the covariance matrix.

[0029] Optionally, in step 3, based on the Lyapunov–Krasovskii theorem, the following can be introduced:

[0030]

[0031]

[0032] Make: ,in .

[0033] Optionally, in step 4, the longitudinal distance between the two vehicles is selected as the error variable for the vehicle platoon longitudinal controller, and the expression is as follows:

[0034]

[0035] In the formula, Position of the vehicle in front. To follow the vehicle's position, Represents the minimum safe following distance. Represents the time distance between the front and rear of the vehicle. Represents the speed of the car in front. For the train commander;

[0036] The sliding surface is designed as follows:

[0037]

[0038] In the formula, This is the error proportional gain coefficient, which affects the system's response speed and convergence speed. It serves as the error integral gain coefficient, providing integral compensation and eliminating the system's steady-state error.

[0039] Optionally, in step 5, in drive mode, the desired torque of the drive motor is:

[0040]

[0041] in, It is the acceleration due to gravity. The rolling resistance coefficient, For the ramp angle, The air drag coefficient, For windward area, The radius of the wheel's rolling motion. The transmission ratio of the main reducer, The transmission efficiency of the power transmission system.

[0042] In braking mode, the desired brake master cylinder pressure is:

[0043]

[0044] in, This is the ratio coefficient between braking force and master cylinder pressure.

[0045] This invention provides a longitudinal control method for unmanned vehicle platoons that considers communication delay issues. First, based on the longitudinal kinematic model of the vehicle platoon, considering the delay problem of the vehicle execution system, a first-order inertial element is added to establish the discrete state-space equation of the longitudinal controller. An improved Kalman filter algorithm is introduced by introducing Sage-Husa filtering, and an AEKF algorithm is designed to estimate the vehicle state information within the platoon. The system stability is verified through stability analysis using the Lyapunov–Krasovskii method and the derivation of the LMI stability margin. Based on the estimated vehicle state information, a sliding mode algorithm upper-level controller is designed, outputting the desired acceleration. The lower-level PID controller, based on the desired acceleration, outputs driving torque and braking pressure to achieve longitudinal control of the platoon following the vehicles. This invention improves control accuracy, reduces performance degradation, and increases the longitudinal stability of the unmanned vehicle platoon in communication-unstable environments when addressing communication delay issues. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a simplified schematic diagram of the execution steps of a longitudinal control method for unmanned vehicle platoons that takes into account communication delay issues, according to the present invention.

[0048] Figure 2 This is a schematic diagram of the longitudinal motion model of the vehicle queue according to the present invention.

[0049] Figure 3 This is a schematic diagram of the AEKF algorithm flow of the present invention.

[0050] Figure 4 This is a diagram of the longitudinal controller framework based on AEKF-SMC of the present invention.

[0051] Figure 5 This is a simulation result diagram of the speed variation of different controllers under a communication delay-free environment, based on a specific embodiment of the present invention.

[0052] Figure 6 This is a simulation result diagram of the speed variation of different controllers under a random communication delay environment, which is a specific embodiment of the present invention.

[0053] Figure 7 This is a simulation result diagram of the vehicle spacing variation under a random communication delay environment according to a specific embodiment of the present invention. Detailed Implementation

[0054] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0055] The following explains the abbreviations of the English terms used in this invention:

[0056] AEKF (Adaptive Extended Kalman Filter): Adaptive Extended Kalman Filter;

[0057] SMC (Sliding Mode Control): Sliding mode control;

[0058] LMI (Linear Matrix Inequality): Linear matrix inequality;

[0059] PID (Proportional-Integral-Derivative): Proportional-Integral-Derivative;

[0060] Please see Figure 1 This invention provides a longitudinal control method for unmanned vehicle platoons that considers communication delay issues, comprising the following steps:

[0061] S1: Based on the longitudinal kinematic model of the vehicle platoon, considering the delay problem of the vehicle execution system, a first-order inertial element is added to establish the discrete state-space equation of the longitudinal controller.

[0062] Step 2: Introduce the Sage-Husa filter to improve the Kalman filter algorithm, and design the AEKF algorithm to estimate the vehicle state information in the queue;

[0063] Step 3: Prove the system stability based on stability analysis using the Lyapunov–Krasovskii method and the derivation of the LMI stability margin;

[0064] Step 4: Design the upper-level controller of the SMC algorithm based on the estimated vehicle state information, and output the desired acceleration;

[0065] Step 5: The lower-level PID controller outputs driving torque and braking pressure based on the desired acceleration to achieve longitudinal control of the convoy following the vehicle.

[0066] The following provides further explanation in conjunction with the implementation steps:

[0067] Step S1: Based on the longitudinal kinematic model of the vehicle platoon, considering the delay problem of the vehicle execution system, a first-order inertial element is added to establish the discrete state-space equation of the longitudinal controller.

[0068] Based on V2V communication, a fixed headway strategy is adopted to calculate the expected distance between adjacent vehicles. Considering the impact of communication delay, a first-order inertial element is introduced to describe the relationship between actual acceleration and expected acceleration, and a discrete time-domain relationship is established.

[0069] Please refer to the details. Figure 2 A simplified diagram of the longitudinal movement of the unmanned vehicle convoy is provided.

[0070] To ensure the normal operation of the unmanned vehicle platoon and to conform to the law of increasing safe braking distance, a fixed headway strategy is adopted as the vehicle platoon spacing strategy.

[0071]

[0072] In the formula, Represents the desired vehicle spacing. Represents the minimum safe following distance. Represents the time distance between the front and rear of the vehicle. This represents the speed of the vehicle in front.

[0073] Vehicle location information obtained from workshop communication can be used to determine the vehicle spacing. The details are as follows:

[0074]

[0075] In the formula: and This represents the position information of this vehicle relative to the vehicle in front. Representative of the train commander.

[0076] The distance error between adjacent vehicles can be obtained from the longitudinal kinematics. and relative velocity :

[0077]

[0078]

[0079] Considering the time delay issue in the vehicle's execution system, a first-order inertial element is added to describe the actual acceleration of the following vehicle. and expected acceleration The relationships between them are as follows:

[0080]

[0081] In the formula: This represents the first-order system gain, with a value of 1. This is the time constant of the inertial element.

[0082] Based on the above model and longitudinal kinematics, the discrete-time domain relation can be obtained as follows:

[0083]

[0084] In the formula: The current moment of the system. For the next moment of the system, This represents the system sampling time.

[0085] In summary, the state vector is selected. Then the discrete state-space equation of the system is:

[0086]

[0087] In the formula: .

[0088] Step S2: Introduce the Sage-Husa filter to improve the Kalman filter algorithm, and design the AEKF algorithm to estimate the vehicle state information in the queue;

[0089] For details, please refer to Figure 3 The AEKF state prediction algorithm is shown in the provided vertical controller framework diagram.

[0090] The system state equation and measurement equation established in this invention can be expressed as follows:

[0091]

[0092] Since higher-order linearization significantly increases algorithm complexity, this invention uses a first-order Taylor expansion to linearize the above model.

[0093]

[0094] In the formula, , This is the Jacobian matrix for taking the partial derivative with respect to the state after linearization.

[0095] To estimate and analyze vehicle state, it is necessary to establish its time update equation and measurement update equation. The time update equation is derived by forward extrapolating prior state estimates. Calculate the prior error covariance matrix The measurement update equation extrapolates prior state estimates. Calculate the measured variables Then, the posterior estimates of the state variables are calculated. Simultaneously update the posterior error covariance matrix of the estimated state. The algorithm continuously updates and iterates the estimated value of the current state variable after obtaining the estimated value of the state variable from the previous time step and the measured value of the current state variable.

[0096] The state prediction equation is:

[0097]

[0098]

[0099] In the formula, for The posterior error covariance matrix at time t. Let be the noise covariance matrix in the state estimation process.

[0100] The measurement update equation is:

[0101]

[0102] In the formula, Here is the Kalman gain matrix. To measure the noise covariance matrix, To observe the residuals, It is an identity matrix.

[0103] During vehicle motion, the system process noise covariance matrix Q and the observation noise covariance matrix... The process noise covariance matrix Q and the observation noise covariance matrix Q are affected by changes in the environment and communication delay. Traditional EKF algorithms assume these two covariance matrices are fixed values, leading to a decrease in filtering accuracy. The Sage-Husa adaptive extended Kalman filter algorithm designed in this paper, based on the traditional EKF algorithm, adjusts the process noise covariance matrix Q and the observation noise covariance matrix Q. The estimation and update are performed as follows:

[0104]

[0105]

[0106] In the formula, The impact factor, which is weighted by the index, updates more slowly when it is closer to 1. For state increments; These are small positive numbers used to maintain positive definiteness; To observe residuals; To predict the covariance matrix.

[0107] By adjusting based on residuals and The filter can respond in real time to the uncertainty caused by communication delay, thus improving the robustness of the estimation.

[0108] Step S3: Prove the system stability based on stability analysis and LMI stability margin derivation using the Lyapunov–Krasovskii method;

[0109] The first The control law for each node is defined as follows:

[0110]

[0111] in, and To compensate for local synchronization errors with the leading / adjacent vehicle, and For time-varying communication delay; the adaptive law is:

[0112]

[0113] From this, we can understand the error dynamics of the following vehicle, and further introduce a block matrix:

[0114]

[0115] It can be known that the time-delay closed-loop system for multiple vehicles is:

[0116]

[0117] Assuming the lead vehicle is reachable, define:

[0118]

[0119] For each delay function, we have:

[0120]

[0121] Equivalently rewritten as a form containing only the current state and the derivative integral:

[0122]

[0123] Define the combinatorial functional as:

[0124]

[0125] in:

[0126]

[0127] According to the Lyapunov–Krasovskii theorem, we introduce:

[0128]

[0129]

[0130] Make: ,in .

[0131] Calculate along the closed-loop system We can obtain:

[0132]

[0133] Applying integral-type quadratic inequalities to the mixed terms and the upper bound of the maximum time delay, and then using Jensen's inequality to estimate the upper bound of the integral terms, we can... Rewrite it as an upper bound form containing only the current state, the lag term, the quadratic form, and its weight matrix. Further, all delays are... This means that we can obtain:

[0134]

[0135] in:

[0136]

[0137] This is a matrix constructed diagonally along the delay block.

[0138]

[0139] Based on the above estimates, Sufficient condition for transformation into the following linear matrix inequality:

[0140]

[0141] If the above LMIs have solutions, then:

[0142]

[0143] That is, the multi-vehicle formation can achieve synchronization with the lead vehicle under multiple time delays, and the adaptive gain converges to a constant value.

[0144] When the adaptive gain converges Let the corresponding limit matrix be , Then the upper bound of the system's maximum tolerable delay can be obtained as:

[0145] .

[0146] Step S4: Design the upper-level controller of the sliding membrane algorithm based on the estimated vehicle state information, and output the desired acceleration;

[0147] For details, please refer to Figure 4 The provided invention provides a longitudinal controller framework diagram based on AEKF-SMC.

[0148] The longitudinal distance between the two vehicles is selected as the error variable for the longitudinal controller of the vehicle platoon.

[0149]

[0150] In the formula, Position of the vehicle in front. To follow the vehicle's position, Represents the minimum safe following distance. Represents the time distance between the front and rear of the vehicle. Represents the speed of the car in front. For the train commander.

[0151] The sliding surface is designed as follows:

[0152]

[0153] In the formula, This is the error proportional gain coefficient, which affects the system's response speed and convergence speed. It serves as the error integral gain coefficient, providing integral compensation and eliminating the system's steady-state error.

[0154] The standard quadratic Lyapunov function is selected as follows:

[0155]

[0156] Substituting the first-order inertial model of the system, the sliding membrane control law is constructed as follows:

[0157]

[0158] To avoid the system's stability being affected by chattering, a saturation function is used to design the reaching law. The specific saturation reaching law is as follows:

[0159]

[0160] In the formula, This represents the inflection point of the saturation function.

[0161] The derivative of the Lyapunov function is as follows:

[0162]

[0163] The controller's sliding diaphragm dynamics satisfy:

[0164]

[0165] In the formula, The total disturbance term consists of all bounded disturbances, including Estimation error, feedforward model error, etc., satisfy:

[0166]

[0167] We can obtain:

[0168]

[0169] when Sometimes, Therefore:

[0170]

[0171] like ,but ,Right now Setting a value greater than the upper limit of the total perturbation terms satisfies the negative definiteness condition of the Lyapunov derivative.

[0172] when ,have Therefore:

[0173]

[0174] like ,but ,Right now Setting a value greater than the upper limit of the total perturbation terms satisfies the negative definiteness condition of the Lyapunov derivative.

[0175] Therefore, the approach law parameters satisfy... Under these conditions, the designed control law ensures that the derivative of the Lyapunov function is strictly negative definite, the system state exhibits stable sliding motion on the sliding surface, and the control system maintains good robustness and stability even under bounded disturbances. The controller theoretically satisfies the requirements for finite-time convergence and disturbance rejection performance.

[0176] Step S5: The lower-level PID controller outputs driving torque and braking pressure based on the desired acceleration to achieve longitudinal control of the queuing following the vehicle.

[0177] Based on the vehicle's longitudinal kinematics model, the equations for the vehicle's longitudinal acceleration and deceleration are obtained.

[0178] accelerate:

[0179]

[0180] In the formula: For vehicles The overall vehicle quality This is the conversion factor for the rotational mass of a vehicle. As the driving force, For air resistance, For rolling resistance, Ramp resistance.

[0181] slow down:

[0182]

[0183] In the formula: For the desired braking force.

[0184] In drive mode, the desired torque of the drive motor can be derived from the above formula as follows:

[0185]

[0186] In the formula: It is the acceleration due to gravity. The rolling resistance coefficient, For the ramp angle, The air drag coefficient, For windward area, The radius of the wheel's rolling motion. The transmission ratio of the main reducer, The transmission efficiency of the power transmission system.

[0187] In braking mode, the desired brake master cylinder pressure is derived from the previous formula as follows:

[0188]

[0189] In the formula: This is the ratio coefficient between braking force and master cylinder pressure.

[0190] Based on the vehicle's longitudinal acceleration and deceleration motion equations, the vehicle's driving torque and braking pressure are obtained.

[0191] For further details, please refer to Figures 5 to 7This invention provides a specific embodiment to verify the performance of the AEKF-SMC queue longitudinal controller. A co-simulation system was built, and the control accuracy and robustness of the built AEKF-SMC queue longitudinal controller were explained through the analysis of the simulation results.

[0192] Under conditions of no communication delay, the result is as follows Figure 5 As shown, the following vehicle speed can stably follow the preceding vehicle speed under both controllers, and the vehicle spacing error between the two controllers is not significantly different, with the maximum vehicle spacing error between the two longitudinal controllers being approximately 0.4m. Therefore, under conditions without communication delay, the AEKF-SMC longitudinal controller designed in this paper, compared to the SMC longitudinal controller, has no impact on the control accuracy of the longitudinal movement of the vehicle platoon after the addition of state estimation. Under conditions with communication delay, such as... Figure 6 As shown, the speeds under both controllers can generally follow the speed of the vehicle in front, but the SMC longitudinal controller exhibits severe fluctuations in vehicle speed tracking, while the AEKF-SMC longitudinal controller maintains high speed tracking accuracy. Figure 7 As shown, the maximum vehicle spacing error under the SMC longitudinal controller can reach 6m, and significant jitter occurs overall, while the maximum vehicle spacing error under the AEKF-SMC longitudinal controller remains stable at around 0.4m. This indicates that the AEKF-SMC controller can effectively handle the random communication delay problem in vehicle platoon communication, thereby ensuring the stability and safety of the platoon system under communication delay conditions.

[0193] Table 1 Comparison of Mean Square Error (MSE) between vehicles with and without communication delay for different controllers

[0194]

[0195] The results in Table 1 show that the AEKF-SMC longitudinal controller maintains high control accuracy even when facing communication delay interference. Compared with the SMC longitudinal controller, the AEKF-SMC controller shows little difference in vehicle spacing under no communication delay conditions. However, under communication delay conditions, the AEKF-SMC controller exhibits significantly lower MSE and IAE values ​​for vehicle spacing error, indicating that it is less affected by communication delay interference. This is attributed to the AEKF state estimation algorithm introduced into the controller, which can effectively estimate the vehicle state under communication delay conditions, thereby improving the robustness of the system. This demonstrates that the AEKF-SMC controller of this invention can effectively solve the communication delay problem and improve the stability and safety of the queuing system.

[0196] The above description discloses only one or more preferred embodiments of the present invention, and should not be construed as limiting the scope of the present invention. Those skilled in the art will understand that all or part of the processes of the above embodiments can be implemented, and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A longitudinal control method for unmanned vehicle platoons considering communication delay issues, characterized in that, Includes the following steps: Step 1: Based on the longitudinal kinematic model of the vehicle platoon, considering the delay problem of the vehicle execution system, a first-order inertial element is added to establish the discrete state-space equation of the longitudinal controller; Step 2: Introduce the Sage-Husa filter to improve the Kalman filter algorithm, and design the AEKF algorithm to estimate the vehicle state information in the queue; Step 3: Prove the system stability based on stability analysis using the Lyapunov–Krasovskii method and the derivation of the LMI stability margin; Step 4: Design the upper-level controller of the SMC algorithm based on the estimated vehicle state information, and output the desired acceleration; Step 5: The lower-level PID controller outputs driving torque and braking pressure based on the desired acceleration to achieve longitudinal control of the convoy following the vehicle.

2. The longitudinal control method for unmanned vehicle platoons considering communication delay issues as described in claim 1, characterized in that, In step 1, a fixed headway strategy is adopted as the vehicle platoon spacing strategy, and a longitudinal kinematic model of the vehicle platoon is established. The discrete time-domain relationship is as follows: ； in, For vehicle spacing error, To follow the relative speed of the vehicle in front, To follow the vehicle's actual acceleration, For the desired acceleration, This represents the first-order system gain, with a value of 1. The time constant of the inertial element, The current moment of the system. For the next moment of the system, The system sampling time; Selecting the state vector The discrete state-space equation of the system is: ； in, .

3. The longitudinal control method for unmanned vehicle platoons considering communication delay issues as described in claim 2, characterized in that, The state prediction equation in step 2 is: ； ； in, for The posterior error covariance matrix at time t. This is the noise covariance matrix in the state estimation process; The measurement update equation is: ； in, Here is the Kalman gain matrix. To measure the noise covariance matrix, To observe the residuals, It is the identity matrix; After introducing the Sage-Husa filter to improve the Kalman filter algorithm, the process noise covariance matrix Q and the observation noise covariance matrix R are estimated, and the update form is as follows: ； ； in, The impact factor, which is weighted by the index, updates more slowly when it is closer to 1. These are small positive numbers used to maintain positive definiteness; To observe residuals; To predict the covariance matrix.

4. The longitudinal control method for unmanned vehicle platoons considering communication delay issues as described in claim 3, characterized in that, In step 3, based on the Lyapunov–Krasovskii theorem, we introduce: ； ； Make: ,in .

5. The longitudinal control method for unmanned vehicle platoons considering communication delay issues as described in claim 4, characterized in that, In step 4, the longitudinal distance between the two vehicles is selected as the error variable for the vehicle platoon longitudinal controller, and the expression is as follows: ； In the formula, Position of the vehicle in front. To follow the vehicle's position, Represents the minimum safe following distance. Represents the time distance between the front and rear of the vehicle. Represents the speed of the car in front. For the train commander; The sliding surface is designed as follows: ； In the formula, This is the error proportional gain coefficient, which affects the system's response speed and convergence speed. It serves as the error integral gain coefficient, providing integral compensation and eliminating the system's steady-state error.

6. The longitudinal control method for unmanned vehicle platoons considering communication delay issues as described in claim 5, characterized in that, In step 5, under drive mode, the desired torque of the drive motor is: ； in, It is the acceleration due to gravity. The rolling resistance coefficient, For the ramp angle, The air drag coefficient, For windward area, The radius of the wheel's rolling motion. The transmission ratio of the main reducer, For the transmission efficiency of the power transmission system; In braking mode, the desired brake master cylinder pressure is: ； in, This is the ratio coefficient between braking force and master cylinder pressure.

Images