A method for controlling the self-organized merging of intelligent connected vehicles into a queue

Abstract

This invention discloses a method for controlling the self-organized merging of intelligent connected vehicles into a queue, comprising: acquiring the speed and position of all target queue vehicles and merging vehicles within the communication distance; calculating the headway between each merging vehicle and the merging vehicle in front of it; determining whether each merging vehicle and the merging vehicle in front of it belong to the same local merging queue; counting the number of different local merging queues; determining whether the number of local merging queues is greater than the set maximum number of local merging queues; selecting the candidate merging position with the smallest distance as the target merging position; and considering the safety space and acceleration constraints of the local merging queues and the target queue, controlling the vehicles in the local merging queues to merge as a whole into the target merging position in the target queue. This method can iteratively organize merging vehicles into local merging queues based on their speed and spacing, and control the local merging queues to merge as a whole into the target queue, reducing the computational complexity of multi-vehicle merging queue problems.

Description

Technical Field

[0001] This invention belongs to the field of autonomous driving technology, specifically relating to a method for controlling the self-organized merging of intelligent connected vehicles into a queue. Background Technology

[0002] In a networked world, vehicles on the road are no longer independent of each other, but connected through wireless communication networks to form a multi-vehicle system, of which vehicle platooning is one type. In platooning system design, addressing the dynamic merging of vehicles will facilitate a wide range of collaborative applications, such as flexible platooning arrangements and expansion, responding to the requirements of different scenarios.

[0003] Existing research has proposed several methods for vehicle merging into a platoon. Chinese invention patent application CN202310019950.8, entitled "Method and Apparatus for Dynamic Reorganization of Vehicle Placing into a Placing of Non-Connected Vehicles," proposes a method for vehicle merging into a platoon. This method predicts the merging trajectory of merging vehicles based on vehicle information and generates target control actions based on the merging trajectory to control the vehicle platoon. However, this method mainly considers merging vehicles that are not intelligent connected vehicles. Chinese invention patent application CN202211348153.6, entitled "A Control Method and System for Vehicles Merging into an Intelligent Connected Placing Team," proposes a control method for vehicles merging into an intelligent connected platoon. This method, considering different levels of vehicle intelligence, designs a longitudinal speed guidance scheme and longitudinal control based on model predictive control through vehicle-to-infrastructure communication technology to achieve precise control of vehicles safely merging into the intelligent connected platoon. However, this method does not consider the situation where multiple vehicles merge into the platoon simultaneously.

[0004] However, in real-world traffic situations, queue control primarily targets intelligent connected vehicles, and the computational complexity and complex coupling relationships arise when controlling multiple vehicles to merge into a queue simultaneously. Therefore, improving the efficiency of multiple vehicles merging into a queue simultaneously and reducing computational complexity are of great significance. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, the present invention aims to provide a method for controlling the self-organized merging of intelligent connected vehicles into a queue, thereby solving the problems of high computational complexity and low efficiency in the control of multiple vehicles merging into a queue in the existing technologies. The method of the present invention can iteratively organize merging vehicles into a local merging queue based on the speed and spacing of the merging vehicles, and control the local merging queue to merge into the target queue as a whole, thus reducing the computational complexity of the problem of multiple vehicles merging into a queue.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] The present invention provides a method for controlling the self-organized merging into a queue for intelligent connected vehicles, comprising the following steps:

[0008] Step 1: Obtain the speeds and positions of all target queue vehicles and merging vehicles within the communication range;

[0009] Step 2: Calculate the headway between each merging vehicle j and the merging vehicle j - 1 in front of it, where j = 2, 3, …, n, and n is the number of merging vehicles within the communication range;

[0010] Step 3: According to the set sub - queue threshold and the headway in Step 2, determine whether each merging vehicle j and the merging vehicle j - 1 in front of it belong to the same local merging queue; when the headway ht j between the merging vehicle j and the merging vehicle j - 1 is less than the sub - queue threshold, the two vehicles belong to the same local merging queue, otherwise the two vehicles belong to different local merging queues respectively; count the number m of different local merging queues;

[0011] Step 4: Judge whether the number of local merging queues calculated in Step 3 is greater than the set maximum number of local merging queues m max (m <00​​​​​​​​​​​​​​​​​​​​​​​​​​This indicates the speed of the merging vehicle j.

[0018] Furthermore, the specific calculation process in step 5 is as follows:

[0019] Step 51: Solve for the candidate inflow location set as follows:

[0020]

[0021] In the formula, s c Let x be the c-th candidate merge position. c Let x be the position of the c-th vehicle in the target queue. c+1 Let p be the position of the (c+1)th vehicle in the target queue, and p be the number of vehicles in the target queue.

[0022] Step 52: Calculate the distance between the first vehicle in the local merging queue and the candidate merging positions of all local merging queues:

[0023] Δs 1c =x M1 -s c

[0024] In the formula, Δs 1c Let x be the distance between the first vehicle in the local merging queue and the c-th candidate merging position. M1 This indicates the position of the first vehicle to join the queue locally.

[0025] Step 53: Select a value that makes Δs 1c The candidate inflow position with the minimum value is taken as the target inflow position. In the target queue, the target merge location The first vehicle in front is c * .

[0026] Furthermore, the design method for the distributed model prediction controller in step 6 is specifically as follows:

[0027] Step 61: Define the control input of vehicle o as u o (t), which is the vehicle acceleration; the state of vehicle o is d. o (t)=[x o (t),v o (t),u o (t)] T The control output is z. o (t)=[x o (t),v o (t)] T In the formula, x o (t) represents the position of vehicle o at time t, v o (t) represents the velocity of vehicle o at time t, u o(t) represents the control input for vehicle o at time t; when controlling the target queue, vehicle o is a vehicle in the target queue; when controlling the local merging queue, vehicle o is a vehicle in the local merging queue.

[0028] Step 62: Define the discrete kinematic equations:

[0029]

[0030] In the formula, d o (t) and d o (t+1) are the state vectors of vehicle o at time t and t-1, respectively; u o (t) represents the control input of vehicle o at time t; τ is the sampling time interval;

[0031] Step 63: The cost function is defined as follows:

[0032]

[0033] In the formula, J o Let k be the cost function; k = 1, ..., N p N p To predict the step size, N c To control the step size; k|t is the k-th prediction performed at time t; when N c ≤k≤N p At that time, u o (k|t)=u o (N c |t);P o Q o M o N o All are symmetric positive definite weight matrices; Let L be the desired position of vehicle o, and d be the length of the vehicle. des Desired vehicle spacing; The desired vehicle speed is set; DV is the target queue vehicle set, DV = {1, 2, ..., p}, where p is the total number of vehicles in the target queue; MV is the local merging queue vehicle set, MV = {1, 2, ..., q}, where q is the total number of vehicles in the local merging queue; v o (k|t) and x o (k|t) is obtained from the following formula:

[0034]

[0035] In the formula, d o (kt)=[x o (k|t),v o (k|t),u o (k|t)] T;d o (t)=[x o (t),v o (t),u o (t)] T This represents the state of vehicle o at the current time t.

[0036] Step 64: Determine the desired positions of vehicles in the target queue and the local merging queue:

[0037] The desired position of vehicle i in the target queue is:

[0038]

[0039] In the formula, This represents the position of the first vehicle in the local merging queue after the k-th step prediction at time t. and These are the position and velocity of the first vehicle in the local merging queue at time t; The position of the last vehicle in the local merging queue after the k-th step prediction at time t. and These are the position and velocity of the last vehicle in the local merging queue at time t;

[0040] The expected position of vehicle j in the locally merged queue is:

[0041]

[0042] In the formula, Merge the target in the target queue at the target position. The first vehicle ahead, C * The position after the k-th prediction at time t. and Vehicle c in the target queue * Position and velocity at time t;

[0043] Step 65: The constraint optimization control problem for vehicles in the target queue and the local merging queue is expressed as:

[0044]

[0045] u min ≤u o (k|t)≤u max

[0046] Δu min ≤u o (k|t)-u o(k|t-1)≤Δu max

[0047] In the formula, u min u max For the maximum and minimum control input constraints, Δu min , Δu max The above formula is used to constrain the difference between two adjacent control inputs. When vehicle o is a target queue vehicle, the above formula optimizes the acceleration of the target queue vehicle. When vehicle o is a locally merging queue vehicle, the above formula optimizes the acceleration of the locally merging queue vehicle.

[0048] Step 66: Use the constraint programming solver to solve the constraint optimization control problem of the vehicle platoon, obtain the optimal acceleration command, and the vehicles merging into the platoon and the target platoon drive according to their respective optimal acceleration commands.

[0049] The beneficial effects of this invention are:

[0050] This invention considers the speed and spacing of multiple merging vehicles and self-organizes them into a local merging queue. Considering the safety space and acceleration constraints of the two adjacent queues, the local merging queue and the target queue, a distributed model prediction controller is designed to control the local merging queue to merge into the target queue as a whole, thereby reducing the computational complexity when multiple vehicles merge into the queue at the same time and improving the merging efficiency. Attached Figure Description

[0051] Figure 1 This is a flowchart of the method of the present invention.

[0052] Figure 2 This is a schematic diagram of a local inbound queue for self-organized vehicles.

[0053] Figure 3 This is a diagram illustrating the process of merging a local inbound queue into a target queue. Detailed Implementation

[0054] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.

[0055] Reference Figure 1 As shown, the intelligent connected vehicle self-organizing merging queue control method of the present invention includes the following steps:

[0056] Step 1: Obtain the speed and location of all target queuing vehicles and merging vehicles within the communication range; specifically, this includes: using V2I technology to send the speed and location information of target queuing vehicles and merging vehicles to the roadside unit and store it.

[0057] Step 2: Calculate the headway between each incoming vehicle j and the incoming vehicle j-1 in front of it, where j = 2, 3, …, n, and n is the number of incoming vehicles within the communication range;

[0058] The calculation formula for the headway between each incoming vehicle and the incoming vehicle in front of it is as follows:

[0059]

[0060] In the formula, ht j represents the headway between incoming vehicle j and the incoming vehicle j-1 in front of it; x j and x j-1 respectively represent the positions of incoming vehicle j and the incoming vehicle j-1 in front of it; v j represents the speed of incoming vehicle j.

[0061] Step 3: Refer to Figure 2 As shown, based on the set division threshold and the headway in Step 2, determine whether each incoming vehicle j and the incoming vehicle j-1 in front of it belong to the same local incoming queue; when the headway ht j between incoming vehicle j and incoming vehicle j-1 is less than the division threshold, the two vehicles belong to the same local incoming queue, otherwise the two vehicles belong to different local incoming queues respectively; count the number m of different local incoming queues;

[0062] Step 4: Determine whether the number of local incoming queues calculated in Step 3 is greater than the set maximum number of local incoming queues m max (m max < n), if so, increase the division threshold by 0.1 s and return to Step 3, otherwise execute Step 5;

[0063] Step 5: Calculate the distance between the first vehicle in the local incoming queue and the candidate incoming positions of all local incoming queues, and select the candidate incoming position with the smallest distance as the target incoming position; the specific process is as follows:

[0064] Step 51: Solve the set of candidate incoming positions as follows:

[0065]

[0066] In the formula, s c is the c-th candidate incoming position, x c is the position of the c-th vehicle in the target queue, x c+1 is the position of the (c + 1)-th vehicle in the target queue, and p is the number of vehicles in the target queue;

[0067] Step 52: Calculate the distance between the first vehicle in the local incoming queue and the candidate incoming positions of all local incoming queues:

[0068] Δs 1c =x M1 -s c

[0069] In the formula, Δs 1c Let x be the distance between the first vehicle in the local merging queue and the c-th candidate merging position. M1 This indicates the position of the first vehicle to join the queue locally.

[0070] Step 53: Select a value that makes Δs 1c The candidate inflow position with the minimum value is taken as the target inflow position. In the target queue, the target merge location The first vehicle in front is c * .

[0071] Step 6: Refer to Figure 3 As shown, a distributed model predictive controller is designed, considering the safety space and acceleration constraints of the local merging queue and the target queue. The controller controls the vehicles in the target queue that are in front of the target merging position to accelerate and the vehicles that are behind the target merging position to decelerate. At the target merging position, a space is formed that meets the driving requirements of the local merging queue. The controller also controls the vehicles in the local merging queue to merge into the target merging position of the target queue as a whole.

[0072] The specific design method for distributed model predictive controllers is as follows:

[0073] Step 61: Define the control input of vehicle o as u o (t), which is the vehicle acceleration; the state of vehicle o is d. o (t)=[x o (t),v o (t),u o (t)] T The control output is z. o (t)=[x o (t),v o (t)] T In the formula, x o (t) represents the position of vehicle o at time t, v o (t) represents the velocity of vehicle o at time t, u o (t) represents the control input for vehicle o at time t; when controlling the target queue, vehicle o is a vehicle in the target queue; when controlling the local merging queue, vehicle o is a vehicle in the local merging queue.

[0074] Step 62: Define the discrete kinematic equations:

[0075]

[0076] In the formula, do (t) and d o (t+1) are the state vectors of vehicle o at time t and t-1, respectively; u o (t) represents the control input of vehicle o at time t; τ is the sampling time interval;

[0077] Step 63: The cost function is defined as follows:

[0078]

[0079] In the formula, J o Let k be the cost function; k = 1, ..., N p N p To predict the step size, N c To control the step size; k|t is the k-th prediction performed at time t; when N c ≤k≤N p At that time, u o (k|t)=u o (N c |t);P o Q o M o N o All are symmetric positive definite weight matrices; Let L be the desired position of vehicle o, and d be the length of the vehicle. des Desired vehicle spacing; The desired vehicle speed is set; DV is the target queue vehicle set, DV = {1, 2, ..., p}, where p is the total number of vehicles in the target queue; MV is the local merging queue vehicle set, MV = {1, 2, ..., q}, where q is the total number of vehicles in the local merging queue; v o (k|t) and x o (k|t) is obtained from the following formula:

[0080]

[0081] In the formula, d o (kt)=[x o (k|t),v o (k|t),u o (k|t)] T ;d o (t)=[x o (t),v o (t),u o (t)] T This represents the state of vehicle o at the current time t.

[0082] Step 64: Determine the desired positions of vehicles in the target queue and the local merging queue:

[0083] The desired position of vehicle i in the target queue is:

[0084]

[0085] In the formula, This represents the position of the first vehicle in the local merging queue after the k-th step prediction at time t. and These are the position and velocity of the first vehicle in the local merging queue at time t; The position of the last vehicle in the local merging queue after the k-th step prediction at time t. and These are the position and velocity of the last vehicle in the local merging queue at time t;

[0086] The expected position of vehicle j in the locally merged queue is:

[0087]

[0088] In the formula, Merge the target in the target queue at the target position. The first vehicle ahead, C * The position after the k-th prediction at time t. and Vehicle c in the target queue * Position and velocity at time t;

[0089] Step 65: The constraint optimization control problem for vehicles in the target queue and the local merging queue is expressed as:

[0090]

[0091] u min ≤u o (k|t)≤u max

[0092] Δu min ≤u o (k|t)-u o (k|t-1)≤Δu max

[0093] In the formula, u min u max For the maximum and minimum control input constraints, Δu min , Δu maxThe above formula is used to constrain the difference between two adjacent control inputs. When vehicle o is a target queue vehicle, the above formula optimizes the acceleration of the target queue vehicle. When vehicle o is a locally merging queue vehicle, the above formula optimizes the acceleration of the locally merging queue vehicle.

[0094] Step 66: Use the constraint programming solver to solve the constraint optimization control problem of the vehicle platoon, obtain the optimal acceleration command, and the vehicles merging into the platoon and the target platoon drive according to their respective optimal acceleration commands.

[0095] This invention has many specific applications. The above description is only a preferred embodiment of this invention. It should be noted that for those skilled in the art, several improvements can be made without departing from the principle of this invention, and these improvements should also be considered within the scope of protection of this invention.

Claims

1. A method for controlling the self-organized merging into a queue of intelligent connected vehicles, characterized in that, The steps are as follows: Step 1: Obtain the speed and position of all target queue vehicles and merging vehicles within the communication range; Step 2: Calculate the headway between each merging vehicle j and the merging vehicle j-1 ahead of it, j = 2, 3, ..., n, where n is the number of merging vehicles within the communication distance; Step 3: Based on the set platooning threshold and the headway in Step 2, determine whether each merging vehicle j and its preceding merging vehicle j-1 belong to the same local merging queue; when the headway ht between merging vehicle j and merging vehicle j-1 is... j When the number of vehicles is less than the platoon threshold, the two vehicles belong to the same local merging queue; otherwise, the two vehicles belong to different local merging queues. Count the number of different local merging queues, m. Step 4: Determine whether the number of local inbound queues calculated in Step 3 is greater than the set maximum number of local inbound queues m. max If yes, increase the squad threshold by 0.1s and return to step 3; otherwise, proceed to step 5. Step 5: Calculate the distance between the first vehicle in the local merging queue and the candidate merging positions of all local merging queues, and select the candidate merging position with the smallest distance as the target merging position; Step 6: Design a distributed model predictive controller, considering the safety space and acceleration constraints of the local merging queue and the target queue. Control the vehicles in the target queue that are in front of the target merging position to accelerate and the vehicles that are behind the target merging position to decelerate. Create a space at the target merging position that meets the driving requirements of the local merging queue, and control the vehicles in the local merging queue to merge into the target merging position of the target queue as a whole. The design method of the distributed model prediction controller in step 6 is as follows: Step 61: Define the control input of vehicle o as u o (t), which is the vehicle acceleration; the state of vehicle o is d. o (t)=[x o (t),v o (t),u o (t)] T The control output is z. o (t)=[x o (t),v o (t)] T In the formula, x o (t) represents the position of vehicle o at time t, v o (t) represents the velocity of vehicle o at time t, u o (t) represents the control input for vehicle o at time t; when controlling the target queue, vehicle o is the target queue vehicle; when controlling the local merging queue, vehicle o is the local merging queue vehicle. Step 62: Define the discrete kinematic equations: In the formula, d o (t) and d o (t+1) represents the state vectors of vehicle o at time t and t-1, respectively; u o (t) represents the control input of vehicle o at time t; τ is the sampling time interval; Step 63: The cost function is defined as follows: In the formula, J o Let k be the cost function; k = 1, ..., N p N p To predict the step size, N c To control the step size; k|t is the k-th prediction performed at time t; when N c ≤k≤N p At that time, u o (k|t)=u o (N c |t);P o Q o M o N o All are symmetric positive definite weight matrices; Let L be the desired position of vehicle o, L be the vehicle length, and d be the length of vehicle o. des Desired vehicle spacing; The desired vehicle speed is set; DV is the target queue vehicle set, DV = {1, 2, ..., p}, where p is the total number of vehicles in the target queue; MV is the local merging queue vehicle set, MV = {1, 2, ..., q}, where q is the total number of vehicles in the local merging queue; v o (k|t) and x o (k|t) is obtained from the following formula: In the formula, d o (k|t)=[x o (k|t),v o (k|t),u o (k|t)] T ;d o (t)=[x o (t),v o (t),u o (t)] T This represents the state of vehicle o at the current time t. Step 64: Determine the desired positions of vehicles in the target queue and the local merging queue: The desired position of vehicle i in the target queue is: In the formula, This represents the position of the first vehicle in the local merging queue after the k-th step prediction at time t. and These are the position and velocity of the first vehicle in the local merging queue at time t; The position of the last vehicle in the local merging queue after the k-th step prediction at time t. and These are the position and velocity of the last vehicle in the local merging queue at time t; The expected position of vehicle j in the locally merged queue is: In the formula, Merge the target in the target queue at the target position. The first vehicle ahead, C * The position after the k-th prediction at time t. and Vehicle c in the target queue * Position and velocity at time t; Step 65: The constraint optimization control problem for vehicles in the target queue and the local merging queue is expressed as: u min ≤u o (k|t)≤u max Δu min ≤u o (k|t)-u o (k|t-1)≤Δu max In the formula, u min u max For the maximum and minimum control input constraints, Δu min , Δu max The difference constraint between two adjacent control inputs is used; when vehicle o is a target queue vehicle, the acceleration of the target queue vehicle is optimized; when vehicle o is a locally merging queue vehicle, the acceleration of the locally merging queue vehicle is optimized. Step 66: Use the constraint programming solver to solve the constraint optimization control problem of the vehicle platoon, obtain the optimal acceleration command, and the vehicles merging into the platoon and the target platoon drive according to their respective optimal acceleration commands.

2. The intelligent connected vehicle self-organizing merging queue control method according to claim 1, characterized in that, Step 1 specifically includes: using V2I technology to send the speed and location information of the target queue vehicles and merging vehicles to the roadside unit and store it.

3. The intelligent connected vehicle self-organizing merging queue control method according to claim 1, characterized in that, The formula for calculating the headway between each merging vehicle and the vehicle merging in front of it in step 2 is as follows: In the formula, ht j This represents the headway between merging vehicle j and the merging vehicle j-1 ahead of it; x j and x j-1 These represent the positions of merging vehicle j and the merging vehicle j-1 ahead of it, respectively; v j This indicates the speed of the merging vehicle j.

4. The intelligent connected vehicle self-organizing merging queue control method according to claim 1, characterized in that, The specific calculation process in step 5 is as follows: Step 51: Solve for the candidate inflow location set as follows: In the formula, s c Let x be the c-th candidate merge position. c Let x be the position of the c-th vehicle in the target queue. c+1 Let p be the position of the (c+1)th vehicle in the target queue, and p be the number of vehicles in the target queue. Step 52: Calculate the distance between the first vehicle in the local merging queue and the candidate merging positions of all local merging queues: Δs 1c =x M1 -s c In the formula, Δs 1c Let x be the distance between the first vehicle in the local merging queue and the c-th candidate merging position. M1 This indicates the position of the first vehicle to join the queue locally. Step 53: Select a value that makes Δs 1c The candidate inflow position with the minimum value is taken as the target inflow position. In the target queue, the target merge location The first vehicle in front is c * .

Images