A method for planning the lane-changing trajectory of an intelligent vehicle based on model predictive control.

Abstract

This invention proposes a lane-changing trajectory planning method for intelligent vehicles based on model predictive control. First, it senses environmental information in real time and calculates the emergency braking distance and desired safe distance to determine the intelligent vehicle's behavior. Then, based on the vehicle's point mass model as the prediction model, it combines an improved artificial potential field method with model predictive control to plan the lane-changing trajectory, which includes the currently optimal lane-changing reference trajectory selected from a sinusoidal lane-changing reference trajectory family. Finally, a trajectory tracking controller based on model predictive control is used to track the trajectory. This invention, employing an improved artificial potential field method, can fully adapt to changes in the position, speed, and acceleration of surrounding obstacle vehicles, resulting in significant dynamic obstacle avoidance and ensuring the safety of lane changes for the intelligent vehicle. This invention also considers scenarios such as following the vehicle in front and canceling a lane change, making it applicable to a wider range of scenarios.

Description

Technical Field

[0001] This invention relates to a method for planning the lane change trajectory of an intelligent vehicle based on model predictive control, belonging to the technical field of lane change trajectory planning for intelligent vehicles. Background Technology

[0002] In recent years, with the rapid development of computer technology and perception technology, autonomous driving technology is moving from partial to full automation. Adaptive cruise control, driver assistance, and lane change assist technologies have been widely applied in vehicles, effectively improving safety, comfort, and efficiency during driving. Lane changing, as the most common driving behavior, has become a key focus of current autonomous driving technology, as reasonable and safe lane changes improve driving efficiency, while unreasonable lane changes can lead to accidents. To enable intelligent vehicles to complete lane-changing tasks more safely and efficiently, it is necessary to plan lane-changing trajectories that meet safety, feasibility, and comfort requirements.

[0003] Artificial Potential Field (APF) and Model Predictive Control (MPC) have been widely applied in trajectory planning for intelligent vehicles. MPC predicts the vehicle's state over a finite time domain based on an existing model and the current state. It then designs an appropriate objective function and constraints to construct an optimization problem, solving which yields the optimal control variables to achieve planning and control. However, complex models increase the computational burden of MPC, significantly impacting the real-time performance of intelligent vehicle planning and control. APF is a classic obstacle avoidance method for intelligent vehicles. It utilizes attractive forces designed around the target point and repulsive forces from obstacles and road boundaries to enable the vehicle to avoid obstacles and reach the target point. However, existing research rarely considers the acceleration information of surrounding obstacles and road boundary information within the potential field, and there are few strategies for situations where lane-changing conditions are not met or the vehicle's safety is threatened during lane-changing. This leaves room for improvement in the safety and feasibility of lane-changing for intelligent vehicles. Summary of the Invention

[0004] The technical problem to be solved by this invention is to overcome the problem that the existing APF only considers the position and speed of surrounding obstacles without considering acceleration and road boundaries. In order to propose a model predictive control-based intelligent vehicle lane change trajectory planning method, a sine curve is used as the reference trajectory when changing lanes. Under the conditions of meeting safety and feasibility, a smoother and more efficient lane change trajectory is obtained.

[0005] This invention provides a method for intelligent vehicle lane-changing trajectory planning based on model predictive control. The system involved in this method includes an environmental perception module, a behavior decision module, a lane-changing trajectory planning module, and a trajectory tracking module. The method includes the following steps:

[0006] Step 1: Obtain environmental perception information based on LiDAR, GPS sensors, and vehicle speed sensors. The environmental perception information includes the status information of the intelligent vehicle and surrounding dynamic obstacles, as well as the information of the intelligent vehicle's current driving road. The status information of the intelligent vehicle and obstacles includes pose, speed, acceleration, etc., and the road information includes the center lines and boundary lines of each lane. Input the status information of the intelligent vehicle and surrounding dynamic obstacles into the emergency braking distance model and the expected safe distance model, and output behavioral decision information according to the decision scheme.

[0007] Step 2: Based on environmental perception information and behavioral decision-making information, construct the objective function for trajectory planning based on the improved artificial potential field method and model predictive control, and combine the constraints to construct the optimization problem of the trajectory planning module;

[0008] Step 3: Select the optimal sinusoidal curve lane change reference trajectory under the current conditions based on the lane change trajectory evaluation function to ensure the safety, efficiency and comfort of the lane change process.

[0009] This invention provides a method for decision-making, planning, and control based on real-time acquisition of relevant information by an environmental perception module. First, the invention senses environmental information in real time and calculates the emergency braking distance and desired safe distance to determine the behavior of the intelligent vehicle. Then, based on the vehicle's point mass model as a prediction model, it combines an improved artificial potential field method and model predictive control to plan a lane-changing trajectory, which includes the current optimal lane-changing reference trajectory selected from a sinusoidal lane-changing reference trajectory family. Finally, a trajectory tracking controller based on model predictive control is used to track the trajectory.

[0010] The following is a further optimized technical solution of the present invention:

[0011] In step 1, the dynamic obstacle vehicles and the vehicle's own status information around the intelligent vehicle are considered in real time, and the information is input into the emergency braking distance model and the expected safe distance model. Further, based on the emergency braking distance, the expected safe distance and the longitudinal distance between the vehicle and the surrounding obstacle vehicles, it is determined whether there is an intention to change lanes and whether the feasibility of changing lanes is met, and finally the behavioral decision information of the intelligent vehicle is output.

[0012] Emergency braking distance refers to the distance between a following vehicle and the preceding vehicle in the event of emergency braking, ensuring safe driving by preventing a collision. The emergency braking distance model is as follows:

[0013]

[0014] In the formula, S is the emergency braking distance of the vehicle in its current state, v0 is the vehicle's speed before emergency braking, t1 is the time during which the deceleration increases linearly, and S ε To correct for distance, b max For the maximum deceleration and b max =μg, where μ is the road surface adhesion coefficient, and g is the acceleration due to gravity and g = 9.8 m / s². 2 ;

[0015] The expected safe distance is mainly related to the speeds of the vehicles in front and behind, the road surface adhesion coefficient, gravitational acceleration, and emergency braking distance. It measures the minimum distance that the following vehicle is expected to maintain with the vehicle in front when traveling at a desired speed. The expected safe distance model is as follows:

[0016]

[0017] In the formula, D fs τ represents the expected safe distance for the vehicle in its current state. r v is the sensor delay time. M v is the longitudinal speed of the rear vehicle. F S is the longitudinal speed of the vehicle in front. M a is the emergency braking distance of the following vehicle. F Acceleration for the vehicle in front.

[0018] In step 1, the specific method for detecting the generation of lane-changing intention and the feasibility of lane changing is as follows:

[0019] (1) Generation of lane change intention - Determine whether the longitudinal distance between the intelligent vehicle and the vehicle in front of the current lane is less than the intelligent vehicle's expected safe distance. If the longitudinal distance between the intelligent vehicle and the vehicle in front of the current lane is less than the intelligent vehicle's expected safe distance, it means that the safety of the intelligent vehicle when driving at the expected speed cannot be guaranteed, so a lane change intention is generated; otherwise, it means that the safety performance of the intelligent vehicle when driving at the expected speed is guaranteed, so no lane change intention is generated.

[0020] (2) Lane change feasibility conditions - The longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is used to determine whether the lane change feasibility conditions are met. When the longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is less than the emergency braking distance of the vehicle behind in the target lane, it means that the lane change feasibility conditions are not met; otherwise, it means that the lane change feasibility conditions are met.

[0021] In step 1, the method for constructing intelligent vehicle behavior decision-making and planning based on lane-changing intention and lane-changing feasibility includes the following specific steps:

[0022] Step 1.1: Determine if there is an intention to change lanes. If there is an intention to change lanes, further determine if the conditions for lane change feasibility are met. If there is no intention to change lanes, maintain the desired speed.

[0023] Step 1.2: Determine whether the conditions for lane change feasibility are met. If the conditions are met, proceed with the lane change; otherwise, slow down and follow the vehicle in front.

[0024] Step 1.3: During the lane change process, the longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is detected in real time, and the safety conditions for lane change are determined based on the detection results. That is, whether the longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is greater than the emergency braking distance of the vehicle behind in the target lane. If the safety conditions for lane change are not met, the lane change is canceled and the vehicle continues to decelerate and follow the vehicle in front.

[0025] Step 1.4: Output the behavioral decision measurement information to the trajectory planning module.

[0026] The specific decision-making process is as follows: When the distance between the vehicle and the vehicle in front is less than the vehicle's desired safe distance, the vehicle will intend to change lanes; otherwise, it will maintain its desired driving state. The feasibility of changing lanes depends on the longitudinal distance between the vehicle and the vehicle behind it in the target lane. This invention only considers the case where there is only one vehicle in the target lane. After intending to change lanes, if the longitudinal distance between the vehicle and the vehicle in the target lane is greater than the emergency braking distance of the vehicle behind it, the feasibility of changing lanes is satisfied, and the lane change is executed; otherwise, the vehicle should follow the vehicle in front. During the lane change process, if the distance between the vehicle and surrounding obstacle vehicles is less than the emergency safe distance, the lane change should be cancelled.

[0027] In step 2, after receiving the behavioral decision information, the trajectory planning module will construct the objective function and optimization problem based on the behavioral decision information and environmental perception information. The improved artificial potential field method considers information such as the vehicle's position, speed, and the positions, speeds, and accelerations of surrounding obstacle vehicles, as well as road information. Specifically, it includes a road potential field function and an obstacle potential field function, the calculation method of which is shown below:

[0028] (1) The method for calculating the road potential field function is as follows:

[0029] U road =k r (yy l ) 2 (yy r ) 2

[0030] In the formula, U road Let k be the potential field function of the road. r Let y be the adjustment coefficient of the road potential field, and y be the lateral position of the vehicle. l y r These are the left and right center lines of a two-lane road, respectively.

[0031] (2) The method for calculating the potential field function of the obstacle is as follows:

[0032] U obs =w1k o f1(x|μ1,∑1)+w2k o f2(x|μ2,∑2)

[0033] In the formula, μ1=(x1,y1) T μ2 = (x2, y2) T

[0034]

[0035] Among them, U obs Let w1 and w2 be the obstacle potential field function, and w1 + w2 = 1. Let x be the position information of the vehicle and x = (x av ,y av ) T k o To measure the comprehensive coefficient of the obstacle's impact, f1 and f2 are both two-dimensional joint probability density functions, μ1 and μ2 are the mean values, and x1, y1, ∑1, ... To calculate the parameters related to the mean and covariance of f1, x2, y2, ∑2, Parameters related to the mean and covariance of f2;

[0036] Furthermore, the mean and covariance matrices of f1 and f2 are defined as follows:

[0037]

[0038] In the formula, (x obs ,y obs (v) represents the position coordinates of the obstacle vehicle. x_obs ,v y_obs (a) represents the longitudinal and lateral velocities of the obstacle vehicle. x_obs ,a y_obs S represents the longitudinal and lateral accelerations of the obstacle vehicle. x To measure the longitudinal safety distance, S y To measure the lateral safe distance, S o The minimum longitudinal distance, D, is determined by the vehicle's dimensions. o The minimum lateral distance is determined by the vehicle dimensions; Δx and Δy represent the adjustable distances in the longitudinal and lateral directions, respectively; T0 is the time interval; k m It is a positive parameter related to vehicle weight, and is defined as:

[0039]

[0040] m represents the mass of the obstacle vehicle.

[0041] In step 2, the objective function and optimization problem of the trajectory planning module can be arbitrarily switched under different decision conditions, ensuring the flexibility of the intelligent vehicle's driving trajectory. The optimization problem of the trajectory planning module is specifically described as follows:

[0042]

[0043]

[0044] U min ≤U i ≤U max

[0045] y rr ≤y t ≤y rl

[0046] |u(t)|<μg

[0047] In the formula, N p For prediction in the time domain, N c To control the time domain, η((t+i|t) is the predicted quantity in the prediction time domain. ref (t+i|t) represents the reference value in the prediction time domain, where t is the current sampling time, i is the i-th time in the current prediction time domain, and U i To control the quantity, U all The total potential function is constructed, where Q and R are weight matrices. Let ξ(t) be the derivative of the state variables in the vehicle point mass model, and u(t) be the control variable in the vehicle point mass model. min U max y represents the upper and lower bounds of the controlled variable. t Let y be the lateral position of the vehicle at time t. rr y rl All are lane boundaries, and |u(t)|<μg is the constraint condition for the control quantity.

[0048] In step 3, the reference value predicted in the time domain during the optimization problem of the trajectory planning module becomes the optimal sinusoidal curve lane change reference trajectory when executing lane change decisions. When a combination of lane change time and longitudinal acceleration is determined, a sinusoidal curve lane change reference trajectory can be determined. The formula for calculating the sinusoidal curve lane change reference trajectory is as follows:

[0049] The lateral acceleration at any time t is:

[0050]

[0051] The transverse velocity at any time t is:

[0052]

[0053] The lateral displacement model at any time t is:

[0054]

[0055] In the formula, t s t is the moment when the lane change begins. e W represents the lane width at the end of the lane change.

[0056] In step 3, the method for solving the optimal sinusoidal curve lane change reference trajectory is as follows:

[0057] Step 3.1: After receiving the lane change instruction from the decision module, based on the real-time acquired status information of surrounding vehicles, assuming that the acceleration of surrounding vehicles remains constant during the lane change process, predict their motion state during the lane change time.

[0058] Step 3.2: Sample the lane change time interval ΔT within the range of [2s, 6s], and within the range of [-3m / s]... 2 3m / s 2 Within the range, longitudinal acceleration intervals Δa are sampled, and any combination of lane change time and longitudinal acceleration can determine a lane change reference trajectory;

[0059] Step 3.3: Evaluate the lane change trajectory, considering the safety, comfort, and efficiency of the intelligent vehicle's lane change process. Linearly weight the three indicators to obtain the lane change reference trajectory evaluation function. Determine the optimal lane change time and longitudinal acceleration by finding the maximum and minimum values ​​of the function.

[0060] The evaluation function for measuring the safety, efficiency, and comfort of the lane change reference trajectory, and the final linearly weighted objective function are as follows:

[0061] (1) Safety evaluation function for lane change process

[0062]

[0063] In the formula, N is the total number of obstacle vehicles, k s Let x(t) and y(t) represent the safety coefficient, and their respective longitudinal and lateral positions at time t. Let be the longitudinal and lateral positions of the i-th surrounding obstacle vehicle at time t, respectively;

[0064] (2) Efficiency evaluation function for lane change process

[0065]

[0066] In the formula, k et k ev These are efficiency coefficients related to time and speed, respectively, v x(t), v y (t) represents the longitudinal and lateral velocities of the intelligent vehicle at time t, respectively, and T is the lane change time at the current sampling point;

[0067] (3) Acceleration is commonly used to measure comfort during lane changes. (Comfort evaluation function)

[0068]

[0069] In the formula, k cx k cy These are the longitudinal and lateral comfort coefficients, respectively, a x (t), a y (t) represents the longitudinal acceleration and lateral acceleration of the intelligent vehicle at time t, respectively;

[0070] The comprehensive evaluation function is obtained by linearly weighting the evaluation functions for safety, comfort, and efficiency, as shown below:

[0071] j = w s j s +w c j c +w e j e

[0072] In the formula, w s w c w e These are the weighting coefficients for safety, comfort, and efficiency, respectively.

[0073] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0074] (1) The present invention adopts an improved artificial potential field function, which not only considers the position and velocity of dynamic obstacles, but also incorporates acceleration information, and constructs a road potential field function based on road information; making the intelligent vehicle more sensitive to the dynamic changes of dynamic obstacles and driving on the center line of the lane as much as possible, thereby increasing the safety of the intelligent vehicle during driving.

[0075] (2) The present invention designs an evaluation function for selecting the optimal sinusoidal curve lane change reference trajectory, and selects the optimal lane change reference trajectory from a series of lane change reference trajectory clusters, thereby improving the safety, efficiency and comfort of the lane change process.

[0076] (3) This invention can not only plan lane-changing trajectories, but also construct different trajectory planning optimization problems based on decision information, thereby planning a reasonable trajectory that meets the current driving environment, effectively improving the flexibility of intelligent vehicles during lane-changing. Specifically, using the method of this invention, intelligent vehicles can cancel lane changes, decelerate and follow the vehicle in front, etc., improving the flexibility of intelligent vehicles in the face of sudden situations during lane changes.

[0077] In summary, the improved artificial potential field method of this invention can fully adapt to changes in the position, speed, and acceleration of surrounding obstacle vehicles, resulting in significant dynamic obstacle avoidance and ensuring the safety of intelligent vehicles changing lanes. This invention also considers scenarios such as following the vehicle in front and canceling lane changes, making it adaptable to a wider range of scenarios. Attached Figure Description

[0078] Figure 1 This is a structural diagram of the lane change trajectory planning system of the present invention.

[0079] Figure 2 This is a flowchart of the lane change trajectory planning method proposed in this invention.

[0080] Figure 3 This is a flowchart illustrating the behavioral decision-making process of this invention.

[0081] Figure 4 This is a road condition vehicle-road simulation diagram for the present invention.

[0082] Figure 5 The figure shows the simulation results of the lateral controller tracking at a speed of 60km / h according to the present invention.

[0083] Figure 6 The figure shows the simulation results of the longitudinal controller of this invention.

[0084] Figure 7 This is a simulation diagram of a lane change trajectory according to one embodiment of the present invention.

[0085] Figure 8 This is a diagram showing the vehicle positions at different times according to one embodiment of the present invention. Detailed Implementation

[0086] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings: This embodiment is implemented under the premise of the technical solution of the present invention, and provides detailed implementation methods and specific operation processes, but the protection scope of the present invention is not limited to the following embodiments.

[0087] like Figure 1 As shown, the intelligent vehicle lane-changing trajectory planning system based on the artificial potential field method and model predictive control includes an environmental perception module, a behavior decision-making module, a lane-changing trajectory planning module, and a trajectory tracking module. Among them:

[0088] The environmental perception module includes LiDAR, GPS sensors, and vehicle speed sensors, which are used to acquire information about the intelligent vehicle, surrounding obstacles, and roads.

[0089] The behavioral decision-making module includes an emergency braking distance model and a desired safe distance model, as well as a decision-making process built based on these two models;

[0090] The lane change trajectory module includes constructing a point mass model of the vehicle to predict the vehicle's state in the future finite time domain, designing a road potential field and an obstacle potential field based on a two-dimensional joint probability density function to ensure the safety of the intelligent vehicle during driving, solving for the optimal lane change time and longitudinal acceleration based on a sine curve-based lane change reference trajectory model, and finally designing an objective function that satisfies the safety, smoothness and efficiency of the intelligent vehicle's lane change, and solving for the optimal lane change trajectory at the current moment.

[0091] The trajectory tracking module includes a vehicle lateral tracking controller and a longitudinal tracking controller built based on a model predictive control algorithm to achieve trajectory tracking.

[0092] like Figure 2 As shown, the intelligent vehicle lane-changing trajectory planning method based on the artificial potential field method and model predictive control includes:

[0093] S1. Real-time acquisition of the intelligent vehicle's status information, the status information of surrounding dynamic obstacles, and the information of the intelligent vehicle's current driving road. The status information of the intelligent vehicle and obstacles includes pose, speed, acceleration, etc., and the road information includes the center line and boundary line of each lane.

[0094] The intelligent vehicle's position and orientation are obtained by GPS sensors; the intelligent vehicle's speed and acceleration information are obtained by vehicle speed sensors; road information is obtained by high-precision maps, including road boundary information, lane line information, etc.; and the status information and speed of surrounding obstacle vehicles are obtained by LiDAR.

[0095] S2. Based on the state information of the intelligent vehicle and obstacles obtained in step S1, input it into the emergency braking distance model and the expected safe distance model to determine if there is a lane change requirement. If there is a lane change requirement, determine if the lane change conditions are met. If the lane change conditions are not met, follow the vehicle in front; otherwise, execute the lane change. The decision results include maintaining the current state (no lane change requirement), following the vehicle in front (lane change requirement but lane change conditions not met), and executing the lane change (lane change requirement and lane change conditions met). Refer to... Figure 3 Provide behavioral decision-making options.

[0096] Specifically, emergency braking distance refers to the distance between a following vehicle and the preceding vehicle in the event of emergency braking, ensuring that the following vehicle will not collide with the preceding vehicle and guaranteeing driving safety. The emergency braking distance model is as follows:

[0097]

[0098] In the formula, S is the emergency braking distance of the vehicle in its current state, v0 is the vehicle's speed before emergency braking, t1 is the time during which the deceleration increases linearly, and S ε To correct for distance, the maximum deceleration is b. max =μg, where μ is the road surface adhesion coefficient, and g = 9.8 m / s 2This is the acceleration due to gravity.

[0099] The expected safe distance is mainly related to the speeds of the vehicles in front and behind, the road surface adhesion coefficient, gravitational acceleration, and emergency braking distance. It measures the minimum distance that the following vehicle is expected to maintain with the vehicle in front when traveling at the desired speed. The expected safe distance model is as follows:

[0100]

[0101] In the formula, τ r v is the sensor delay time; M v is the longitudinal speed of the rear vehicle. F S represents the longitudinal speed of the vehicle in front. M a is the emergency braking distance of the following vehicle. F Acceleration for the vehicle in front.

[0102] Based on the emergency braking distance, the expected safe distance, and the longitudinal distance between the vehicle and surrounding obstacle vehicles, the system determines whether the intelligent vehicle intends to change lanes and whether the lane change is feasible, and finally outputs the intelligent vehicle's behavioral decision information.

[0103] The specific methods for detecting the generation of lane-changing intentions and the feasibility of lane changes are as follows:

[0104] (1) Generation of lane change intention - Determine whether the longitudinal distance between the intelligent vehicle and the vehicle in front of the current lane is less than the intelligent vehicle's expected safe distance. If the longitudinal distance between the intelligent vehicle and the vehicle in front of the current lane is less than the intelligent vehicle's expected safe distance, it means that the safety of the intelligent vehicle when driving at the expected speed cannot be guaranteed, so a lane change intention is generated; otherwise, it means that the safety performance of the intelligent vehicle when driving at the expected speed is guaranteed, so no lane change intention is generated.

[0105] (2) Lane change feasibility conditions - The longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is used to determine whether the lane change feasibility conditions are met. When the longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is less than the emergency braking distance of the vehicle behind in the target lane, it means that the lane change feasibility conditions are not met; otherwise, it means that the lane change feasibility conditions are met.

[0106] The method for constructing intelligent vehicle behavior decision-making and planning based on lane-changing intention and feasibility involves the following specific steps:

[0107] Step 1.1: Determine if there is an intention to change lanes. If there is an intention to change lanes, further determine if the conditions for lane change feasibility are met. If there is no intention to change lanes, maintain the desired speed.

[0108] Step 1.2: Determine whether the conditions for lane change feasibility are met. If the conditions are met, proceed with the lane change; otherwise, slow down and follow the vehicle in front.

[0109] Step 1.3: During the lane change process, the longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is detected in real time, and it is determined whether the lane change safety conditions are met based on the detection results (i.e., whether the longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is greater than the emergency braking distance of the vehicle behind in the target lane). If the safety of the lane change process is not met, the lane change is canceled and the vehicle continues to decelerate and follow the vehicle in front.

[0110] Step 1.4: Output the behavioral decision measurement information to the trajectory planning module.

[0111] S3. Based on the decision information from step S2, construct different objective functions. The objective function for local trajectory planning should include the reference state under the decision. Specifically, when the decision is to perform a lane change, an evaluation function is established based on the sinusoidal lane change curve family to select the optimal lane change reference trajectory from the sinusoidal lane change reference trajectory, and to determine the lane change time and longitudinal acceleration. During the lane change process, the intelligent vehicle is monitored in real time for any safety threats. If a threat is detected, the lane change is canceled, and the vehicle follows the vehicle in front; otherwise, the lane change operation continues. Based on the decision information output from step S2, construct an objective function for trajectory planning based on improved APF and MPC.

[0112] The specific method for constructing the objective function for trajectory planning based on the improved APF and MPC is as follows:

[0113] (1) Establish the point mass model of the vehicle, abbreviated as:

[0114]

[0115] In the formula, φ, Y, X represent the lateral speed, longitudinal speed, heading angle, and x and y coordinates of the intelligent vehicle in Cartesian coordinates, respectively; u represents the control variable a. y .

[0116] (2) The improved artificial potential field method includes the road potential field function and the obstacle potential field function.

[0117] The road potential field function is constructed as follows:

[0118] U road =k r (yy l ) 2 (yy r ) 2

[0119] In the formula, U road Let k be the potential field function of the road. r Let y be the adjustment coefficient of the road potential field, and y be the lateral position of the vehicle. l y rThese are the left and right center lines of a two-lane road.

[0120] The specific method for constructing the obstacle potential field based on the two-dimensional joint probability density function is as follows:

[0121] The two-dimensional joint probability density function is:

[0122]

[0123] In the formula, x = (x1, x2) T x i Both are one-dimensional random variables, x i The expectation is μ i ∑ is the covariance matrix.

[0124] Define the obstacle potential field function as:

[0125] U obs =w1k o f1(x|μ1,∑1)+w2k o f2(x|μ2,∑2)

[0126] In the formula, μ1=(x1,y1) T μ2 = (x2, y2) T

[0127]

[0128] Among them, U obs Let w1 and w2 be the obstacle potential field function, and w1 + w2 = 1. Let x be the position information of the vehicle and x = (x av ,y av ) T k o To measure the comprehensive coefficient of the obstacle's impact, f1 and f2 are both two-dimensional joint probability density functions, μ1 and μ2 are the mean values, and x1, y1, ∑1, ... To calculate the parameters related to the mean and covariance of f1, x2, y2, ∑2, The parameters related to the calculation of the mean and covariance of f2 are given.

[0129] Furthermore, the mean and covariance matrices of f1 and f2 are defined as follows:

[0130]

[0131] In the formula, (x obs ,y obs (v) represents the position coordinates of the obstacle vehicle. x_obs ,v y_obs (a) represents the longitudinal and lateral velocities of the obstacle vehicle. x_obs ,ay_obs S represents the longitudinal and lateral accelerations of the obstacle vehicle. x S y S measures the safety distances in both the longitudinal and lateral directions. o D o These are the minimum longitudinal and lateral distances determined by the vehicle dimensions; Δx and Δy represent the adjustable longitudinal and lateral distances, respectively; T0 is the time interval; k m It is a positive parameter related to vehicle weight, and is defined as:

[0132]

[0133] m represents the mass of the obstacle vehicle.

[0134] Finally, the total potential field function, which is the sum of the road potential field and the obstacle potential field, is defined as:

[0135] U all =U road +U obs

[0136] The objective function of local trajectory planning should include the constructed artificial potential field function, with the aim of ensuring the safety of intelligent vehicle driving.

[0137] (3) If the decision result in step S2 is to change lanes, in order to ensure safety, efficiency and comfort during the lane change process, the optimal lane change reference trajectory will be selected from the sinusoidal curve family as the reference trajectory during the lane change process. The sinusoidal curve lane change reference trajectory model can be divided into lateral acceleration, lateral velocity and lateral displacement models, and the specific forms are as follows:

[0138] The lateral acceleration a at any time t y (t) is:

[0139]

[0140] The lateral velocity v at any time t y (t) is:

[0141]

[0142] The lateral displacement model y at any time t y (t) is:

[0143]

[0144] In the formula, t s t is the moment when the lane change begins. e At the end of the lane change, A y denoted as the peak value of lateral acceleration, and W as the lane width.

[0145] The optimal lane change reference trajectory is obtained based on the following three principles:

[0146] (1) Upon receiving the lane change instruction from the decision module, the motion state of the surrounding vehicles is predicted based on the real-time acquired status information of the surrounding vehicles. Assuming that the acceleration of the surrounding vehicles remains constant during the lane change process, the motion state of the surrounding vehicles during the lane change time is predicted.

[0147] (2) Sample the lane change time interval ΔT within [2s, 6s], and within [-3m / s 2 3m / s 2 By sampling the longitudinal acceleration interval Δa, a lane change reference trajectory can be determined by any combination of lane change time and longitudinal acceleration.

[0148] (3) Evaluate the lane change trajectory, taking into account the safety, comfort, and efficiency of the intelligent vehicle's lane change process. The three indicators are linearly weighted to obtain the lane change reference trajectory evaluation function, and the lane change time and longitudinal acceleration are determined by finding the minimum value of the function.

[0149] Furthermore, a safety evaluation function j is constructed during the lane change process. s for:

[0150]

[0151] In the formula, k s Let x(t) and y(t) represent the safety coefficient, and their respective longitudinal and lateral positions at time t. Let be the longitudinal and lateral positions of the i-th obstacle vehicle at time t, respectively, and N be the total number of obstacle vehicles.

[0152] Furthermore, an efficiency evaluation function j is constructed during the lane-changing process. e for:

[0153]

[0154] In the formula, k et k ev These are efficiency coefficients related to time and speed, respectively, v x (t), v y (t) represents the longitudinal and lateral velocities of the intelligent vehicle at time t, respectively, and T is the lane change time currently being sampled.

[0155] Furthermore, acceleration is commonly used as a metric for comfort, and a comfort evaluation function j is constructed. c for:

[0156]

[0157] In the formula, k cx k cyThese are the longitudinal and lateral comfort coefficients, respectively, a x (t), a y (t) represents the longitudinal acceleration and lateral acceleration of the intelligent vehicle at time t, respectively.

[0158] The comprehensive evaluation function is obtained by linearly weighting the evaluation functions for safety, comfort, and efficiency, as shown below:

[0159] j = w s j s +w c j c +w e j e

[0160] In the formula, w s w c w e These are the weighting coefficients for safety, comfort, and efficiency, respectively. By evaluating the objective function of the optimal sinusoidal lane change reference trajectory and finding its minimum value, the optimal lane change time and longitudinal acceleration can be obtained, thus yielding the optimal sinusoidal lane change reference trajectory.

[0161] (4) By finding the minimum value of j, the optimal lane change time and longitudinal acceleration can be determined, and thus the optimal sinusoidal lane change reference trajectory during the lane change process can be determined. A vehicle point mass model is constructed as a prediction model to predict the vehicle's state information in a finite time domain. Combined with the artificial potential field function and the reference trajectory, the objective function of the trajectory planning module can be constructed.

[0162] Furthermore, the objective function for constructing the local trajectory of the intelligent vehicle based on APF and MPC is as follows:

[0163]

[0164] In the formula, N p N c These represent the prediction time domain and the control time domain, respectively, η((t+i|t), η ref (t+i|t) represents the predicted quantity and the reference quantity within the prediction time domain, U i To control the quantity, U all The total function of the constructed artificial potential field.

[0165] S4. Based on the objective function and constraints established in step S3, construct the optimization problem.

[0166] Based on the objective function of the trajectory planning module, and considering the relevant constraints of the intelligent vehicle, the optimization problem of the trajectory planning module is constructed as follows:

[0167]

[0168]

[0169] U min ≤U i ≤U max

[0170] y rr ≤y t ≤y rl

[0171] |u(t)|<μg

[0172] In the formula, N p N c These represent the prediction time domain and the control time domain, respectively, η((t+i|t), η ref (t+i|t) represents the predicted quantity and the reference quantity within the prediction time domain, where t is the current sampling time, i is the i-th time within the current prediction time domain, and U i To control the quantity, U all The total potential function is constructed, where Q and R are weight matrices. Let ξ(t) be the derivative of the state variables in the vehicle point mass model, and u(t) be the control variable in the vehicle point mass model. min U max y represents the upper and lower bounds of the controlled variable. t Let y be the lateral position of the vehicle at time t. rr y rl Let |u(t)| < μg be the lane boundary, and |u(t)| < μg be the constraint condition for the control quantity.

[0173] Specifically, η in optimization problems ref (t+i|t) will change based on decision information, such as when there is a need to change lanes but no conditions for doing so. ref (t+i|t) At this point, the current lane and the speed of the vehicle in front should be used as the target trajectory to enable the intelligent vehicle to follow the vehicle. When the lane change condition is met, the calculated optimal sinusoidal curve lane change reference trajectory will be used as η. ref (t+i|t).

[0174] S5. Solve the optimization problem constructed in step S4 to obtain Nc-dimensional discrete local trajectory points.

[0175] S6. Smooth the discrete local trajectory output in step S5 using a polynomial fitting method, and send the fitting parameters to the trajectory tracking module.

[0176] The local trajectory obtained in step S5 is a discrete trajectory point. Directly inputting it into the trajectory tracking control layer is not conducive to tracking. Therefore, a polynomial fitting method is used to smooth the discrete local trajectory, and the fitting parameters are sent to the trajectory tracking module. The fifth-degree polynomial expression is as follows:

[0177] f(t) = a0 + a1t + a2t 2 +a3t 3 +a4t 4 +a5t 5

[0178] In the formula, a = [a0 a1 a2 a3 a4 a5] are the coefficients of the polynomial to be determined, and a0, a1, a2, a3, a4, and a5 are the coefficients of the constant term to the fifth degree term in the polynomial, respectively.

[0179] S7. Design a model predictive control-based horizontal and vertical trajectory tracking controller, and track the trajectory according to the parameters of the local trajectory output in step S6.

[0180] (1) Construct an MPC lateral trajectory tracking controller based on the vehicle's dynamic model. The vehicle's dynamic model is as follows:

[0181]

[0182] It can be abbreviated as:

[0183]

[0184] In the formula, These are the derivatives of the state variables and the control variables in the vehicle dynamics model, respectively. For the state variables of the system, Y and X represent the lateral velocity, longitudinal velocity, vehicle yaw angle, rate of change of vehicle yaw angle, vehicle lateral position, and vehicle longitudinal position, respectively, and the front wheel steering angle u. dyn =δ f This refers to the control variables of the system.

[0185] The dynamic model is linearized as follows:

[0186]

[0187] Furthermore, it is discretized as follows:

[0188] ξ dyn (k+1)=A dyn (k)ξ dyn (k)+B dyn (k)u dyn (k)

[0189] In the formula, A dyn (k)=I+T dyn A dyn (t), B dyn (k)=T dyn B dyn (t), where I is the identity matrix and T is...dyn The sampling period is [value]. The discretized system state equations are obtained as follows:

[0190]

[0191] In the formula, A dyn (k), ξ dyn (k), B dyn (k), u dyn (k), η dyn (k), C dyn (k) represents the state coefficient matrix, state variable, control coefficient matrix, control variable, output coefficient matrix, and output variable, respectively.

[0192] (2) To improve the stability of the intelligent vehicle's motion, the system's control increment is used as the controlled variable. Therefore, a new state vector is constructed as follows:

[0193] ξ dyn (k|t)=[x dyn u dyn (k-1)] T

[0194] In the formula, x dyn These are the state variables in the dynamic model.

[0195] The new state-space expression is obtained as follows:

[0196]

[0197] In the formula, Δu dyn (k) represents the control increment.

[0198] The output of the new system in the prediction time domain is shown in the following form:

[0199] Y dyn (t)=Ψ dyn ξ dyn (t∣t)+Θ dyn ΔU dyn (t)

[0200] In the formula, Ψ dyn Let ξ be the state coefficient matrix. dyn (t|t) is the state variable matrix, Θ dyn To control the incremental coefficient matrix, ΔU dyn (t) is the control increment matrix.

[0201] Considering the complexity of solving the problem, a relaxation factor is added to the objective function, and the objective function of the lateral trajectory tracking controller is constructed as follows:

[0202]

[0203] In the formula, η dyn (t+i|t) represents the predicted quantity in the time domain at time t, and η dyn,ref (t+i|t) represents the reference quantity in the time domain for prediction at time t, Δu dyn (t+i|t) represents the increment of the control quantity in the control time domain at time t, where ρ is a constant and ε is a relaxation factor.

[0204] The optimization problem to be solved in each control cycle is described as follows:

[0205]

[0206] stΔu dyn,min ≤Δu dyn,t ≤Δu dyn,max

[0207] u dyn,min ≤AΔu dyn,t +u dyn,t ≤u dyn,max

[0208] y hc,min ≤y hc ≤y hc，max

[0209] y sc,min -ε≤y sc ≤y sc,max +ε

[0210] ε>0

[0211] In the formula, N p N c These are the prediction time domain and the control time domain, respectively. hc Hard constraint output refers to output values ​​that cannot increase the constraint range. sc Soft-constrained output refers to an output range where the relaxation factor can dynamically change the output quantity. hc,min and y hc,max For the hard constraint limit value, y sc,min and y sc,max This represents the soft constraint limit value.

[0212] (3) Solving this problem yields a series of control increments in the control time domain, as shown below:

[0213]

[0214] In the formula, These are the control quantity increments in the control time domain, calculated at time t.

[0215] The first element in the control sequence is used as the actual control increment and applied to the system, as shown below:

[0216] u dyn (t)=u dyn (t-1)+Δu dyn,t

[0217] Through rolling optimization and feedback correction, intelligent vehicles can achieve lateral tracking of the trajectory.

[0218] Furthermore, the longitudinal control of the vehicle can be represented using a first-order inertial system, as shown below:

[0219]

[0220] In the formula, K is the system gain; τ d Let a be a time constant. des Let a be the reference acceleration and 'a' be the actual acceleration. The state equation of the continuous system of longitudinal motion of the vehicle can be expressed as:

[0221]

[0222] In the formula, x = [va] T Let u = a be the state variable of the system. des τ is the control input of the system. d Let a be a time constant. des For reference acceleration.

[0223] Furthermore, it is discretized as follows:

[0224] x(k+1)=A k x(k)+B k u(k)

[0225] In the formula,

[0226]

[0227] Where k is the current sampling time, T s The sampling period.

[0228] The control objective of longitudinal control is to track the given speed. Therefore, the speed v is taken as the system output, and the output equation is:

[0229]

[0230] Considering the smoothness of the control process, the change in the control quantity is used as the control quantity, and the cost function is constructed as shown below:

[0231]

[0232] yp (t+i|t) represents the predicted quantity in the time domain at time t, y ref (t+i|t) represents the reference value in the time domain for prediction at time t.

[0233] (4) Considering the relevant constraints, the optimization problem is constructed as follows:

[0234]

[0235] stu min ≤u(t+i)≤u max

[0236] Δu min ≤Δu(t+i)≤Δu max

[0237] In the formula, N p For prediction in the time domain, N c For predicting the time domain, u min u max For the maximum and minimum values ​​of acceleration, Δu min , Δu max This represents the maximum or minimum value of the acceleration increment.

[0238] Solving this optimization problem will yield a series of control increments in the control time domain. The first output value plus the control quantity at the previous moment will give the optimal control quantity at that moment. Inputting this value into the controlled vehicle will enable speed tracking.

[0239] In summary, based on the vehicle dynamics model and the first-order inertial system design model, predictive control is used to control the lateral and longitudinal trajectory tracking controllers. The optimization problem of the lateral tracking controller is as follows:

[0240]

[0241] stΔu dyn,min ≤Δu dyn,t ≤Δu dyn,max

[0242] u dyn,min ≤AΔu dyn,t +u dyn,t ≤u dyn,max

[0243] y hc,min ≤y hc ≤y hc，max

[0244] y sc,min -ε≤y sc ≤y sc,max +ε

[0245] ε>0

[0246] The optimization problem for the longitudinal tracking controller is as follows:

[0247]

[0248] stu min ≤u(t+i)≤u max

[0249] Δu min ≤Δu(t+i)≤Δu max

[0250] By solving the two optimization problems above and inputting the first control variable from the solution into the vehicle, the vehicle can be controlled to travel along the planned trajectory.

[0251] Example 1

[0252] like Figure 4 As shown, the driving environment of the intelligent vehicle in this embodiment is a straight road. There is an obstacle vehicle ahead on the current road and an obstacle vehicle behind on the left side of the road. In order to obtain environmental information in real time, this invention builds a Carsim and Simulink co-simulation platform to verify the feasibility of the method.

[0253] First, the effectiveness of the trajectory tracking module was verified, such as... Figure 5 , Figure 6 As shown.

[0254] Set the driving trajectories of two obstacle vehicles. The initial position of the vehicle in front in the current lane is (50,-2), and it moves forward at a constant speed of 45km / h. The initial position of the vehicle in the target lane is (10,2), and it moves forward at a constant speed of 45km / h. The desired speed of the vehicle is 60km / h.

[0255] First, the system acquires real-time environmental information. When the distance between the intelligent vehicle and the vehicle in front exceeds the vehicle's expected safe distance, it shows no intention to change lanes and maintains the expected speed. The expected safe distance is calculated as follows:

[0256]

[0257] When an intelligent vehicle intends to change lanes, it calculates whether the distance to the vehicle behind it in the target lane is greater than the emergency braking distance of the following vehicle. If it is greater than this distance, the lane change is considered feasible. The emergency braking distance is calculated as follows:

[0258]

[0259] Once the lane change feasibility condition is met, calculate the optimal sinusoidal lane change reference trajectory under the current state. The objective function is as follows:

[0260] j = w s j s +w c j c +w e j e

[0261] After obtaining the final lane change time and longitudinal acceleration, substituting them into the sinusoidal lane change reference trajectory model yields the optimal lane change reference trajectory. The calculation of the sinusoidal lane change reference trajectory is as follows:

[0262] The lateral acceleration at any given time is:

[0263]

[0264] The lateral velocity at any given time is:

[0265]

[0266] The lateral displacement model at any given time is:

[0267]

[0268] The trajectory planning optimization problem is constructed based on the artificial potential field method and model predictive control, and takes the following form:

[0269]

[0270]

[0271] U min ≤U i ≤U max

[0272] y rr ≤y t ≤y rl

[0273] |u(t)|<μg

[0274] Solve this problem, and fit the discrete trajectory with a fifth-order polynomial to obtain a smooth trajectory. The fitting equation is:

[0275] f(t) = a0 + a1t + a2t 2 +a3t 3 +a4t 4 +a5t 5

[0276] Inputting parameters into the trajectory tracking layer facilitates tracking. In this embodiment, the lane-changing trajectory of the intelligent vehicle is shown below. Figure 7 and Figure 8 In particular, Figure 8 The unfilled black boxes represent two obstacle vehicles. To distinguish different vehicles at the same time, letters and numbers are used near the boxes to present the simulation results: letter O represents the driver vehicle (controlled vehicle), letter L represents the vehicle in the left lane, and letter R represents the vehicle in the right lane.

[0277] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any transformations or substitutions that can be conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for planning the lane-changing trajectory of an intelligent vehicle based on model predictive control, characterized in that, Includes the following steps: Step 1: Obtain environmental perception information based on LiDAR, GPS sensor and vehicle speed sensor. The environmental perception information includes the status information of the intelligent vehicle and surrounding dynamic obstacles, as well as the information of the intelligent vehicle's current driving road. Input the status information of the intelligent vehicle and surrounding dynamic obstacles into the emergency braking distance model and the expected safe distance model, and output behavioral decision information according to the decision scheme. Step 2: Based on environmental perception information and behavioral decision-making information, construct the objective function for trajectory planning based on the improved artificial potential field method and model predictive control, and combine the constraints to construct the optimization problem of the trajectory planning module; Step 3: Select the optimal sinusoidal curve lane change reference trajectory under the current conditions based on the lane change trajectory evaluation function to ensure the safety, efficiency and comfort of the lane change process.

2. The intelligent vehicle lane-changing trajectory planning method based on model predictive control according to claim 1, characterized in that, In step 1, the emergency braking distance model is as follows: In the formula, S is the emergency braking distance of the vehicle in its current state, v0 is the vehicle's speed before emergency braking, t1 is the time during which the deceleration increases linearly, and S ε To correct for distance, b max For the maximum deceleration and b max =μg, where μ is the road surface adhesion coefficient, and g is the acceleration due to gravity and g = 9.8 m / s². 2 ; The desired safe distance model is as follows: In the formula, D fs τ represents the expected safe distance for the vehicle in its current state. r v is the sensor delay time. M v is the longitudinal velocity of the rear vehicle. F S is the longitudinal speed of the vehicle in front. M a is the emergency braking distance of the following vehicle. F < represents the acceleration of the vehicle in front.

3. The intelligent vehicle lane-changing trajectory planning method based on model predictive control according to claim 2, characterized in that, In step 1, the specific method for detecting the generation of lane-changing intention and the feasibility of lane changing is as follows: (1) Generation of lane change intention - Determine whether the longitudinal distance between the intelligent vehicle and the vehicle in front of the current lane is less than the intelligent vehicle's expected safe distance. If the longitudinal distance between the intelligent vehicle and the vehicle in front of the current lane is less than the intelligent vehicle's expected safe distance, it means that the safety of the intelligent vehicle when driving at the expected speed cannot be guaranteed, so a lane change intention is generated. Otherwise, it means that the safety performance of the intelligent vehicle is guaranteed when driving at the desired speed, so it does not intend to change lanes; (2) Lane change feasibility conditions - The longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is used to determine whether the lane change feasibility conditions are met. When the longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is less than the emergency braking distance of the vehicle behind in the target lane, it means that the lane change feasibility conditions are not met; otherwise, it means that the lane change feasibility conditions are met.

4. The intelligent vehicle lane-changing trajectory planning method based on model predictive control according to claim 3, characterized in that, In step 1, the method for constructing intelligent vehicle behavior decision-making and planning based on lane-changing intention and lane-changing feasibility includes the following specific steps: Step 1.1: Determine if there is an intention to change lanes. If there is an intention to change lanes, further determine if the conditions for lane change feasibility are met. If there is no intention to change lanes, maintain the desired speed. Step 1.2: Determine whether the conditions for lane change feasibility are met. If the conditions are met, proceed with the lane change; otherwise, slow down and follow the vehicle in front. Step 1.3: During the lane change process, the longitudinal distance between the intelligent vehicle and the vehicle behind in the target lane is detected in real time, and the safety conditions for lane change are determined based on the detection results. If the safety conditions for lane change are not met, the lane change is canceled and the vehicle continues to decelerate and follow the vehicle in front. Step 1.4: Output the behavioral decision measurement information to the trajectory planning module.

5. The intelligent vehicle lane-changing trajectory planning method based on model predictive control according to claim 4, characterized in that, In step 2, the improved artificial potential field method includes a road potential field function and an obstacle potential field function, and its calculation method is as follows: (1) The method for calculating the road potential field function is as follows: OR road =k r (yy l ) 2 (yy r ) 2 In the formula, U road Let k be the potential field function of the road. r Let y be the adjustment coefficient of the road potential field, and y be the lateral position of the vehicle. l y r These are the left and right center lines of a two-lane road, respectively. (2) The method for calculating the potential field function of the obstacle is as follows: U obs =w1k o f1(x|μ1,∑1)+w2k o f2(x|μ2,∑2) In the formula, μ1=(x1,y1) T μ2 = (x2, y2) T Among them, U obs Let w1 and w2 be the obstacle potential field function, and w1 + w2 = 1. Let x be the position information of the vehicle, and k be the obstacle potential field function. o To measure the comprehensive coefficient of the obstacle's impact, f1 and f2 are both two-dimensional joint probability density functions, μ1 and μ2 are the mean values, and x1, y1, ∑1, ... To calculate the parameters related to the mean and covariance of f1, x2, y2, ∑2, The parameters related to the calculation of the mean and covariance of f2 are given.

6. The intelligent vehicle lane-changing trajectory planning method based on model predictive control according to claim 5, characterized in that, In step 2, the optimization problem of the trajectory planning module is specifically described as follows: In the formula, N p For predicting the time domain, N c To control the time domain, η((t+i|t) is the predicted quantity in the prediction time domain. ref (t+i|t) represents the reference value in the prediction time domain, where t is the current sampling time, i is the i-th time in the current prediction time domain, and U i To control the quantity, U all The total potential function is constructed, where Q and R are weight matrices. Let ξ(t) be the derivative of the state variables in the vehicle point mass model, and u(t) be the control variable in the vehicle point mass model. min U max y represents the upper and lower bounds of the controlled variable. t Let y be the lateral position of the vehicle at time t. rr y rl All are lane boundaries, and |u(t)|<μg is the constraint condition for the control quantity.

7. The intelligent vehicle lane-changing trajectory planning method based on model predictive control according to claim 6, characterized in that, In step 3, the formula for calculating the sinusoidal curve lane change reference trajectory is as follows: The lateral acceleration at any time t is: The transverse velocity at any time t is: The lateral displacement model at any time t is: In the formula, t s t is the moment when the lane change begins. e W represents the lane width at the end of the lane change.

8. The intelligent vehicle lane-changing trajectory planning method based on model predictive control according to claim 7, characterized in that, In step 3, the method for solving the optimal sinusoidal curve lane change reference trajectory is as follows: Step 3.1: Based on the real-time acquired status information of surrounding vehicles, assuming that the acceleration of surrounding vehicles remains constant during the lane change process, predict their motion state during the lane change time. Step 3.2: Sample the lane change time interval ΔT within the range of [2s, 6s], and within the range of [-3m / s]... 2 3m / s 2 Within the range, longitudinal acceleration intervals Δa are sampled, and any combination of lane change time and longitudinal acceleration can determine a lane change reference trajectory; Step 3.3: Evaluate the lane change trajectory, considering the safety, comfort, and efficiency of the intelligent vehicle's lane change process. Linearly weight the three indicators to obtain the lane change reference trajectory evaluation function. Determine the optimal lane change time and longitudinal acceleration by finding the maximum and minimum values ​​of the function.

9. The intelligent vehicle lane-changing trajectory planning method based on model predictive control according to claim 8, characterized in that, The evaluation function for measuring the safety, efficiency, and comfort of the lane change reference trajectory, and the final linearly weighted objective function are as follows: (1) Safety evaluation function for lane change process In the formula, N is the total number of obstacle vehicles, k s Let x(t) and y(t) represent the safety coefficient, and their respective longitudinal and lateral positions at time t. Let be the longitudinal and lateral positions of the i-th surrounding obstacle vehicle at time t, respectively; (2) Efficiency evaluation function for lane change process In the formula, k et k ev These are efficiency coefficients related to time and speed, respectively, v x (t), v y (t) represents the longitudinal and lateral velocities of the intelligent vehicle at time t, respectively, and T is the lane change time currently being sampled; (3) Comfort evaluation function In the formula, k cx k cy These are the longitudinal and lateral comfort coefficients, respectively, a x (t), a y (t) represents the longitudinal acceleration and lateral acceleration of the intelligent vehicle at time t, respectively; The comprehensive evaluation function is obtained by linearly weighting the evaluation functions for safety, comfort, and efficiency, as shown below: j=w s j s +w c j c +w e j e In the formula, w s w c w e These are the weighting coefficients for safety, comfort, and efficiency, respectively.

10. The intelligent vehicle lane-changing trajectory planning method based on model predictive control according to claim 5, characterized in that, The mean and covariance matrices of f1 and f2 are defined as follows: In the formula, (x obs ,y obs (v) represents the position coordinates of the obstacle vehicle. x_obs ,v y_obs (a) represents the longitudinal and lateral velocities of the obstacle vehicle. x_obs ,a y_obs S represents the longitudinal and lateral accelerations of the obstacle vehicle. x To measure the longitudinal safety distance, S y To measure the lateral safe distance, S o The minimum longitudinal distance, D, is determined by the vehicle's dimensions. o The minimum lateral distance is determined by the vehicle dimensions; Δx and Δy represent the adjustable distances in the longitudinal and lateral directions, respectively; T0 is the time interval; k m It is a positive parameter related to vehicle weight, and is defined as: m represents the mass of the obstacle vehicle.

Images