Vehicle team control method and system in automatic driving exclusive lane

Abstract

The invention relates to a motorcade control method and system in an automatic driving exclusive lane. The method comprises the following steps: firstly, shooting an entrance ramp traffic flow in real time, establishing a relationship between the position and time of a passing vehicle when the passing vehicle is detected, and obtaining and calibrating a space-time window occupied by the passing vehicle on an automatic driving special lane; then calculating a longitudinal track of the automatic driving vehicle in the automatic driving lane, and taking a space-time part where the longitudinal track intersects with a space-time window occupied by the passing vehicle on the automatic driving lane as a control range limit; solving the track of each automatic driving vehicle in the control range boundary so as to obtain a track control instruction; and finally, the track control instruction is transmitted to the automatic driving vehicle for execution, and then a new round of calculation is performed and the state is updated. Compared with the prior art, the method has the advantages of realizing effective control of the motorcade in the automatic driving special lane when all the vehicles are manually driven on the ramp, improving the efficiency, increasing the universality and the safety and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of motion control of autonomous driving vehicles, and in particular to a method and system for controlling a vehicle fleet in an autonomous driving dedicated lane. Background Art

[0002] Setting up dedicated lanes for autonomous driving on highways can separate autonomous driving vehicles from ordinary manually driven vehicles, reduce the interference of manual driving on autonomous driving, and improve safety and efficiency. Setting up the outer lanes of highways as dedicated lanes is feasible in practical applications, such as the dedicated lanes for autonomous driving trucks on the Donghai Bridge. However, it is inevitable that at the entrance and exit ramps of highways, manual driving needs to cross the dedicated lanes for autonomous driving because of the main line traffic, which will inevitably interfere with the normal passage of autonomous driving.

[0003] Existing autonomous driving following technologies, such as Cooperative Adaptive Cruise Control (CACC), can achieve front and rear following driving of an autonomous driving fleet based on data interaction between multiple vehicles, but its information is limited to the autonomous driving vehicles. When a manually driven vehicle crosses an autonomous driving fleet, CACC cannot respond effectively and in a timely manner. Only when the first vehicle in the autonomous driving fleet detects the crossing behavior of a manually driven vehicle through its own sensors, it will take emergency deceleration measures until it stops and gives way, which greatly reduces traffic efficiency. The root problem is that it cannot perceive the merging and crossing of manually driven vehicles beyond visual range and cannot respond in advance.

[0004] Most of the existing autonomous driving merging control methods, such as the invention patents with publication numbers CN109598950A, CN110930697A, and CN115273501A, solve the fully autonomous driving scenario. However, the reality is that due to cost and market development reasons, autonomous driving will coexist with manual driving for a long time. The invention patents with publication numbers CN114999152A and CN116386385A consider the problem of mixing in manual driving, but the solutions disclosed in these patents require cooperation between smart cars, guide cars, and auxiliary cars. If there is no smart car on the ramp, control cannot be implemented. In addition, the existing technology only controls the merging main line scenario, and does not consider the control of the autonomous driving vehicle in the scenario where the manually driven vehicle crosses the autonomous driving dedicated lane. If the existing merging technology is directly used for the crossing scenario, the distance between the autonomous driving vehicles will be widened, and the subsequent driving efficiency will be reduced. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a method and system for controlling a fleet in an autonomous driving dedicated lane.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] According to one aspect of the present invention, a method for controlling a fleet in an autonomous driving dedicated lane is provided, the method comprising the following steps:

[0008] S1. Real-time video of the entrance ramp traffic flow. When a crossing vehicle is detected, the position, speed and acceleration data of the crossing vehicle are calculated, the relationship between the position and time of the crossing vehicle is established, and the time-space window of the autonomous driving lane occupied by the crossing vehicle is obtained and calibrated;

[0009] S2. Calculate the longitudinal trajectory of the autonomous driving vehicle in the dedicated autonomous driving lane, and use the spatiotemporal portion of the trajectory that intersects with the spatiotemporal window of the dedicated autonomous driving lane occupied by the crossing vehicle as the control range boundary;

[0010] S3, starting from the autonomous driving vehicle closest to the crossing point, in order from far to near, calculate the trajectory of each autonomous driving vehicle within the control range, solve the optimal acceleration value, speed value and position value of the corresponding autonomous driving vehicle at each time step, and thus obtain the trajectory control instruction;

[0011] S4: The trajectory control instructions are transmitted to all autonomous driving vehicles involved in trajectory adjustment for execution. After that, the process returns to step S1 to perform a new round of calculations and update the status.

[0012] As a preferred technical solution, the specific steps of calculating the position, velocity and acceleration data of the crossing vehicle and establishing the position and time relationship of the crossing vehicle in S1 are:

[0013] S11. When a crossing vehicle enters the monitoring range, its instantaneous position, instantaneous speed and instantaneous acceleration are monitored;

[0014] S12, calculating the time and distance of the crossing vehicle accelerating to the maximum speed limit, and establishing a relationship between the position and time of the crossing vehicle;

[0015] S13. Execute S11 and S12 once for the crossing vehicle at a set interval, update the estimated time when the vehicle arrives at the crossing point, and terminate when its instantaneous position exceeds the crossing point. At the end of the period, the vehicle is removed from the monitoring range.

[0016] As a preferred technical solution, the specific formula for the relationship between the position and time of the crossing vehicle is:

[0017]

[0018]

[0019]

[0020]

[0021] Where t′ is the time it takes for the crossing vehicle to accelerate to the maximum speed limit, v M is the maximum speed limit, v mt is the instantaneous speed of the crossing vehicle, a mt is the instantaneous acceleration of the crossing vehicle, s′ is the distance between the crossing vehicle and the monitoring point when the crossing vehicle accelerates to the maximum speed limit, that is, the relative position of the crossing vehicle. is the estimated time when the crossing vehicle arrives at the crossing point, L R is the distance between the crossing point and the detection point, and t is the current time.

[0022] As a preferred technical solution, the specific method of marking the time-space window in S1 is to draw a time-space diagram of the time-space window in the form of a rectangle. In the time-space diagram, the top edge of the rectangle represents the downstream safety distance when the crossing vehicle occupies the automatic driving lane, and the bottom edge represents the upstream safety distance when the crossing vehicle occupies the automatic driving lane. The left side represents the earliest safety time when the crossing vehicle occupies the automatic driving lane, and the right side represents the latest safety time when the crossing vehicle occupies the automatic driving lane. The specific formula is:

[0023] L1=L+ε

[0024] L2=L-ε

[0025]

[0026]

[0027] Where L1 is the downstream safety distance, L2 is the upstream safety distance, t1 is the earliest safety time, t2 is the latest safety time, L is the coordinate of the crossing point, ε is the safety distance between the crossing vehicle and the preceding or following vehicle in the dedicated autonomous driving lane when the crossing vehicle passes through the crossing point, is the moment when the center point of the crossing vehicle passes through the crossing point, and τ is the safe time interval between the crossing vehicle and the preceding or following vehicle in the autonomous driving lane when the crossing vehicle passes through the crossing point.

[0028] As a preferred technical solution, the intersection of the space-time graph and the longitudinal trajectory of the autonomous driving vehicle in S2 is a collision, and the boundary of the intersection is the control range limit.

[0029] As a preferred technical solution, the longitudinal trajectory of the autonomous driving vehicle in the dedicated autonomous driving lane is calculated in S2 by establishing a vehicle position and speed transfer equation between two adjacent unit time steps, and establishing a relationship between the position and time of the autonomous driving vehicle. The specific formula is:

[0030]

[0031] v i+1 =v i +a i ·Δt

[0032]

[0033] Where t0 is the time of the initial state when calculating the vehicle trajectory; t is the position coordinate s to be calculated t moment; s0 is the coordinate of the initial state when calculating the vehicle trajectory; s t is the displacement of the vehicle at time t; v0 is the velocity of the initial state when calculating the vehicle trajectory; v t is the speed of the vehicle at time t; a0 is the acceleration of the initial state when calculating the vehicle trajectory; a t is the acceleration of the vehicle at time t; Δt is the unit time step, i is the sequence number of the time step, that is, the initial moment of the i-th time step, is the floor operator, and % is the remainder operator.

[0034] As a preferred technical solution, the trajectory of each autonomous driving vehicle is calculated in S3 by solving the optimal trajectory of the autonomous driving vehicle based on the constraints. The specific formula is:

[0035]

[0036] Where n is the number of the autonomous driving vehicle, a i，n is the acceleration of vehicle n at the initial moment of the ith time step, t is the current moment, Δt is the unit time step, t0 is the moment of the initial state when calculating the vehicle trajectory, and i is the sequence number of the time step, that is, the initial moment of the ith time step.

[0037] As a preferred technical solution, the constraint conditions for calculating the trajectory of each autonomous driving vehicle in S3 include the first constraint to the seventh constraint, and the specific formula is:

[0038]

[0039] In the formula, the first constraint describes the coordinate transfer equation of any vehicle n within the control range, the second constraint describes the speed transfer equation of n, the third and fourth constraints restrict the trajectory from intersecting any rectangle, and the fifth constraint restricts the speed to not exceed the maximum speed limit v max , and is not a negative number, that is, no reversing occurs. The sixth constraint limits the safe distance μ between the two vehicles in succession, that is, there is no intersection between the two autonomous driving vehicles. The seventh constraint limits the controllable acceleration a of vehicle n in step size i. i,n Can only be selected from a finite number of values, such as a + Indicates the acceleration value specified when the vehicle accelerates, a- The deceleration value specified when the vehicle decelerates. 0 means the vehicle does not accelerate or decelerate. i,n is the coordinate of vehicle n at the initial moment of the i-th time distribution, v i,n is the speed of vehicle n at the initial moment of the i-th time span, s t1 is the coordinate of the autonomous driving vehicle at time t1, s t2 is the coordinate of the autonomous driving vehicle at time t2, l is the length of the autonomous driving vehicle, L is the coordinate of the crossing point, ε is half the distance between the two autonomous driving vehicles in the autonomous driving lane when the crossing vehicle passes the crossing point, and v max is the maximum speed limit, a + Indicates the acceleration value specified when the vehicle accelerates, a - It is the deceleration value specified when the vehicle decelerates. 0 means the vehicle does not accelerate or decelerate. μ is the safe distance between vehicles.

[0040] As a preferred technical solution, when it is determined in S2 that the trajectory of the autonomous driving vehicle intersects the spatiotemporal window of the crossing vehicle occupying the dedicated lane, the situations where the trajectory of the autonomous driving vehicle intersects with the crossing vehicle include the situation where the vehicle is far away from the downstream safety distance and the situation where the vehicle is close to the upstream safety distance, and the specific formula is:

[0041]

[0042]

[0043] In the formula, s t1 is the coordinate of the autonomous driving vehicle at time t1, s 12 is the coordinate of the autonomous driving vehicle at time t2, l is the length of the autonomous driving vehicle, L is the coordinate of the crossing point, and ε is half the distance between the two autonomous driving vehicles in the autonomous driving lane when the crossing vehicle passes the crossing point.

[0044] According to another aspect of the present invention, a control system for a fleet in an autonomous driving dedicated lane is provided, the system is applied to work by a method for controlling a fleet in an autonomous driving dedicated lane as described above, and the system includes a video detector, a roadside controller, and a communicator;

[0045] The video detector uses a general road traffic monitoring camera to completely capture the ramp and acceleration lane area, monitor and identify vehicles on the ramp in real time, and use the image processing technology configured therein to calculate the real-time position coordinates, speed and acceleration data of the vehicle, and send the data to the roadside controller;

[0046] The roadside controller is placed within the control area and is used to receive data from the video detector and the autonomous driving vehicle, establish a relationship between the position and time of the crossing vehicle, calculate the longitudinal trajectory of the autonomous driving vehicle in the autonomous driving lane, calculate the trajectory control command, and communicate it to all autonomous driving vehicles involved in trajectory adjustment;

[0047] The communicator is installed in the video detector, the roadside controller and the autonomous driving vehicle. The video detector sends data of vehicles crossing the ramp to the roadside controller through the communicator. The autonomous driving vehicle sends vehicle trajectory data in the autonomous driving lane to the roadside controller through the communicator. The roadside controller issues trajectory control instructions to the autonomous driving vehicle through the communicator.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. The present invention takes real-time photos of the entrance ramp traffic flow. When a crossing vehicle is detected, the position, speed, and acceleration data of the crossing vehicle are calculated, the relationship between the position and time of the crossing vehicle is established, and the time-space window occupied by the crossing vehicle in the autonomous driving lane is obtained and calibrated; then the longitudinal trajectory of the autonomous driving vehicle in the autonomous driving lane is calculated, and the time-space part intersecting with the time-space window occupied by the crossing vehicle in the autonomous driving lane is used as the control range limit; starting from the autonomous driving vehicle closest to the crossing point, in order from far to near, the trajectory of each autonomous driving vehicle within the control range limit is calculated to obtain the trajectory control instruction; finally, the trajectory control instruction is conveyed to all autonomous driving vehicles involved in trajectory adjustment, and after a sampling time interval, the steps are repeated to perform a new round of calculation and update the status. Mixed flow control can be achieved without the cooperation between the autonomous driving vehicle on the ramp, the guide vehicle on the main line, and the auxiliary vehicle (cooperative vehicle), and effective control of the fleet in the autonomous driving lane can be achieved when all vehicles on the ramp are manually driven.

[0050] 2. The specific method of marking the space-time window in the present invention is to draw a space-time diagram with the space-time window in the form of a rectangle. In the space-time diagram, the top side of the rectangle represents the downstream safety distance when the crossing vehicle occupies the automatic driving lane, and the bottom side represents the upstream safety distance when the crossing vehicle occupies the automatic driving lane. The left side represents the earliest safety moment when the crossing vehicle occupies the automatic driving lane, and the right side represents the latest safety moment when the crossing vehicle occupies the automatic driving lane. By predicting the time when the ramp vehicle passes the vehicle crossing point and establishing a space-time window, the acceleration of the main road automatic driving vehicle is adjusted to avoid the space-time window. This makes the present invention not only applicable to ramp vehicles that are manually driven vehicles, but also applicable to ramp vehicles that are automatic driving vehicles, that is, the automatic driving vehicle is treated as a manually driven vehicle to track and predict its trajectory, which broadens the scope of application of the method and makes the fleet control method universal.

[0051] 3. The present invention calculates the trajectory of each autonomous driving vehicle within the control range in order from far to near, starting from the autonomous driving vehicle closest to the crossing point, and obtains trajectory control instructions, so that the distance between autonomous driving vehicles is increased when there is a crossing vehicle, and is narrowed after the crossing vehicle leaves. This solves the problem of controlling autonomous driving vehicles in a scenario where a manually driven vehicle crosses an autonomous driving lane, and improves driving efficiency in the autonomous driving lane.

[0052] 4. The present invention calculates the trajectory of each autonomous driving vehicle by solving the optimal trajectory of the autonomous driving vehicle by constraints from the first to the seventh constraints. The first constraint describes the coordinate transfer equation of any vehicle n within the control range, the second constraint describes the speed transfer equation of n, the third and fourth constraints restrict the trajectory from intersecting any rectangle, the fifth constraint restricts the speed not to exceed the maximum speed limit and not to be negative, that is, no reversing occurs, the sixth constraint restricts the safety distance between the two vehicles in succession, and the seventh constraint restricts the controllable acceleration of vehicle n within the step length to only be selected from a finite number of values; by using the constraints, the number of times the vehicle acceleration is adjusted is minimized, and the avoidance of crossing vehicles is smoothly achieved with a smaller control cost, and the goal is to minimize the sum of the absolute values ​​of the acceleration of all vehicles within the control range, while avoiding crossing vehicles. Minimizing energy consumption and ensuring transportation efficiency.

[0053] 5. When constructing the time-space window occupied by the crossing vehicle, the present invention introduces the time-space margin, that is, ε and τ are half the distance between the two autonomous driving vehicles in the autonomous driving lane when the ε crossing vehicle passes the crossing point, is the time when the center point of the crossing vehicle passes through the crossing point, and τ is the half safe time interval between the front and rear vehicles in the autonomous driving lane from the time when the rear of the first vehicle passes through the crossing point to the time when the front of the second vehicle passes through the crossing point. By introducing these two parameters, the success rate of manually driven vehicles crossing the autonomous driving lane is improved, and the overall safety is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a schematic diagram of a fleet control system in an autonomous driving dedicated lane in the present invention;

[0055] Figure 2 A schematic diagram of the steps of a method for controlling a vehicle fleet in a dedicated lane for autonomous driving in the present invention;

[0056] Figure 3 This is a flow chart of the control of a fleet in an autonomous driving lane in response to the crossing of a manually driven vehicle in an embodiment;

[0057] Figure 4 It is a schematic diagram showing the spatial-temporal window rectangle of the crossing vehicle occupying the dedicated autonomous driving lane in the embodiment expressed on the spatial-temporal graph;

[0058] Figure 5 Schematic diagram of the temporal and spatial window intersection judgment rule between the trajectory of the autonomous driving vehicle and the crossing vehicle occupying the dedicated lane in the embodiment;

[0059] Figure 6a A flowchart for solving the feasible acceleration operation in the trajectory of the autonomous driving vehicle in the embodiment;

[0060] Figure 6b A flowchart for solving the problem of an acceleration operation being infeasible in the trajectory of an autonomous driving vehicle in an embodiment;

[0061] Figure 7 A schematic diagram of the space-time window of the crossing vehicle and the avoidance trajectory of the first autonomous driving vehicle in the embodiment;

[0062] Figure 8 Schematic diagram of the trajectories of six autonomous driving vehicles avoiding five crossing vehicles in the embodiment. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0064] Example

[0065] Existing autonomous driving following technologies, such as Cooperative Adaptive Cruise Control (CACC), can achieve front and rear following driving of an autonomous driving convoy based on data interaction between multiple vehicles, but its information is limited to the autonomous driving vehicles. When a manually driven vehicle crosses an autonomous driving convoy, CACC cannot respond effectively and in a timely manner. Only when the first vehicle in the autonomous driving convoy detects the crossing behavior of a manually driven vehicle through its own sensors, it will take emergency deceleration measures until it stops to give way, which greatly reduces the traffic efficiency.

[0066] In this embodiment, a fleet control system in an autonomous driving lane is constructed, and a fleet control method in an autonomous driving lane is adopted to control the trajectory of the autonomous driving fleet in the lane at the entrance ramp of a highway with an outer autonomous driving lane, so that the entrance ramp traffic can orderly cross the autonomous driving lane and enter the inner lane of the highway to avoid deceleration, parking and congestion due to crossing difficulties. The crossing vehicle can be either a manually driven vehicle or an autonomous driving vehicle.

[0067] In this embodiment, the fleet control system constructed in the autonomous driving dedicated lane includes a video detector, a roadside controller and a communicator.

[0068] In this embodiment, the video detector uses a general road traffic monitoring camera, which can completely capture the ramp and acceleration lane area, detect and identify vehicle targets on the ramp in real time, and provide the real-time position coordinates, speed, and acceleration of the vehicle. The roadside controller is placed within the control area, and can communicate with the autonomous driving vehicle and the video detector, and execute the fleet control algorithm in the autonomous driving lane of the highway. The video detector, the roadside controller, and the autonomous driving vehicle are all equipped with a communicator. The video detector and the autonomous driving vehicle send the vehicle trajectory data in the ramp and the vehicle trajectory data in the autonomous driving lane to the roadside controller through the communicator. The roadside controller can issue an instruction to adjust the acceleration to the autonomous driving vehicle through the communicator.

[0069] In this embodiment, the method for controlling a fleet in a dedicated autonomous driving lane is as follows: Figure 2 As shown, it mainly includes the following steps:

[0070] S1. Real-time video of the entrance ramp traffic flow. When a crossing vehicle is detected, the position, speed and acceleration data of the crossing vehicle are calculated, the relationship between the position and time of the crossing vehicle is established, and the time-space window of the autonomous driving lane occupied by the crossing vehicle is obtained and calibrated;

[0071] S2. Calculate the longitudinal trajectory of the autonomous driving vehicle in the dedicated autonomous driving lane, and use the spatiotemporal portion of the trajectory that intersects with the spatiotemporal window of the dedicated autonomous driving lane occupied by the crossing vehicle as the control range boundary;

[0072] S3, starting from the autonomous driving vehicle closest to the crossing point, in order from far to near, calculate the trajectory of each autonomous driving vehicle within the control range, solve the optimal acceleration value, speed value and position value of the corresponding autonomous driving vehicle at each time step, and thus obtain the trajectory control instruction;

[0073] S4: Communicate the trajectory control instructions to all autonomous driving vehicles involved in trajectory adjustment. After a sampling time interval, return to step S1 to perform a new round of calculations and update the status.

[0074] In this embodiment, the autonomous driving vehicles form a queue in the autonomous driving lane in a coordinated adaptive cruise mode, and broadcast the longitudinal position coordinates, instantaneous speed, and instantaneous acceleration data of each autonomous driving vehicle along the highway mileage to the outside world through the communicator, and the data can be directly received by the roadside controller. When a manually driven vehicle enters the highway from the entrance ramp, it needs to cross the autonomous driving lane and enter the inner lane. The crossing point of the manually driven vehicle is located on the center line of the autonomous driving lane. After the entrance ramp vehicle accelerates in the acceleration lane, the center point of the vehicle crosses the autonomous driving lane from the crossing point and enters the inner lane. The control range limit is set at the position of L meters from the crossing point of the manually driven vehicle in the autonomous driving lane. When the center point of the autonomous driving vehicle enters the control range limit, the trajectory data of the autonomous driving vehicle can be received by the roadside controller, and the autonomous driving vehicle can also receive the trajectory control command issued by the roadside controller.

[0075] In this embodiment, the trajectory control command is solved by the roadside controller through the proposed convoy control method for dealing with the crossing of manually driven vehicles in the dedicated automatic driving lane of the highway, and the specific information is transmitted to the automatic driving vehicle through the communicator, including the acceleration value, speed value, and position value of the corresponding vehicle at the specific time.

[0076] In this embodiment, the highway merging area scene is set as follows Figure 1 As shown in the figure, the outermost lane of the road is the dedicated lane for autonomous driving. The crossing vehicles on the ramp need to cross the dedicated lane for autonomous driving and then merge into the inner lane of the main line. The control range is L = 200m, and the distance from the crossing point L on the ramp is R = = 100m away, a video detector is set up. A roadside controller is set up on the main road within the control range. Each autonomous driving vehicle is equipped with a communicator, and the video detector and each vehicle can exchange data with the roadside controller.

[0077] This embodiment adopts a method for controlling a fleet in a dedicated lane for autonomous driving. The specific process of the method is as follows: Figure 3 As shown, the specific process is as follows:

[0078] S1. Real-time video of the entrance ramp traffic flow. When a crossing vehicle is detected, the position, speed and acceleration data of the crossing vehicle are calculated, the relationship between the position and time of the crossing vehicle is established, and the time-space window of the autonomous driving lane occupied by the crossing vehicle is obtained and calibrated;

[0079] The specific process of S1 is as follows:

[0080] S11: Manually drive the vehicle through point L using the distance set on the ramp RThe video detector is a general road traffic monitoring camera. The camera has the functions of tracking vehicles in the field of view, measuring the distance between the center point of the vehicle and the camera setting position, and measuring the instantaneous speed and instantaneous acceleration of the vehicle.

[0081] S12: When a vehicle enters the detection range of the video detector, a unique number m=1, 2, ..., M is assigned to the vehicle in the order in which it enters. At time t, the instantaneous position S of the vehicle m is detected once. mt , instantaneous speed v mt , instantaneous acceleration a mt , where the instantaneous position S mt It refers to the distance from the center point of vehicle m to the location of the video detector.

[0082] S13: Predict the time when vehicle m reaches the crossing point of the manually driven vehicle

[0083] S131: Calculate the acceleration of vehicle m to the maximum speed limit v M The time t′=(v M -v mt ) / a mt , and accelerate to v M Distance

[0084] S132: If s′<L R ,but If s′≥L R ,but

[0085] Where t′ is the time it takes for the crossing vehicle to accelerate to the maximum speed limit, v M is the maximum speed limit, v mt is the instantaneous speed of the crossing vehicle, a mt is the instantaneous acceleration of the crossing vehicle, s′ is the distance between the crossing vehicle and the monitoring point when the crossing vehicle accelerates to the maximum speed limit, that is, the relative position of the crossing vehicle. is the estimated time when the crossing vehicle arrives at the crossing point, L R is the distance between the crossing point and the detection point, and t is the current time.

[0086] Through the above calculation, the estimated arrival time of each crossing vehicle at the entrance ramp at this time is shown in Table 1.

[0087] S14: Every Δt, execute S12 and S13 for each vehicle m within the range of the video detector, and update the estimated arrival time of vehicle m at the crossing point. Until S mt ≥L RWhen the vehicle m is closer to the crossing point, the predicted The more accurate it is.

[0088] Table 1 The estimated arrival time of each vehicle crossing the entrance ramp at this time

[0089]

[0090] In this embodiment, the following control parameters are set according to the actual scenario:

[0091] L=200m, Δt=1s, l=18m, v o =12m / s,v max =22m / s, ε=40m, τ=3s, a + =2m / s 2 , a - =-3m / s 2 .

[0092] S2. Calculate the longitudinal trajectory of the autonomous driving vehicle in the dedicated autonomous driving lane, and use the spatiotemporal portion of the trajectory that intersects with the spatiotemporal window of the dedicated autonomous driving lane occupied by the crossing vehicle as the control range boundary;

[0093] In this embodiment, the specific process of S2 is as follows:

[0094] S21: Establish the vehicle position and speed transfer equation between two adjacent time steps Δt:

[0095]

[0096] v i+1 =v i +a i ·Δt

[0097] S22: Export the position coordinates s of the autonomous driving vehicle at time t t The formula is as follows:

[0098]

[0099] Where t0 is the time of the initial state when calculating the vehicle trajectory; t is the position coordinate s to be calculated t moment; s0 is the coordinate of the initial state when calculating the vehicle trajectory; s t is the displacement of the vehicle at time t; v0 is the velocity of the initial state when calculating the vehicle trajectory; v t is the speed of the vehicle at time t; a0 is the acceleration of the initial state when calculating the vehicle trajectory; a tis the acceleration of the vehicle at time t; Δt is the unit time step, and i is the sequence number of the time step, that is, the initial time of the i-th time step. is the floor operator, and % is the remainder operator. The coordinate of point 0 is the coordinate of the control range boundary, and the coordinate of the passing point of the manually driven vehicle is L. This formula can calculate the vehicle coordinate at any time t (t0 < t) under the state at any initial time t0.

[0100] S23: Establish a spatio-temporal diagram. Draw the spatio-temporal window in which the on-ramp passing vehicle occupies the dedicated lane obtained in S1 in the form of a rectangle in the spatio-temporal diagram of this step, as Figure 7 shown. Among them, the top side of the rectangle represents the downstream safety distance when the passing vehicle occupies, the bottom side represents the upstream safety distance when the passing vehicle occupies, the left side represents the earliest safety time when the passing vehicle occupies, and the right side represents the latest safety time when the passing vehicle occupies. Whenever the prediction of the on-ramp passing vehicle trajectory is updated again in step S1, the spatio-temporal window rectangle in the spatio-temporal diagram established in this step is updated in real time.

[0101] In this embodiment, let the time when the center point of the passing vehicle passes through the passing point be Then the spatio-temporal window rectangle is expressed in the spatio-temporal diagram as Figure 4 shown, and the expressions of its four sides are specifically:

[0102] S231 is the downstream safety distance: L1 = L + ε;

[0103] S232 is the upstream safety distance: L2 = L - ε;

[0104] S233 is the earliest safety time:

[0105] S234 is the latest safety time:

[0106] In the formula, ε is the safety distance between the passing vehicle and the vehicle in front or behind in the dedicated lane for autonomous driving when the passing vehicle passes through the passing point, is the time when the center point of the passing vehicle passes through the passing point, and τ is the safety time interval between the passing vehicle and the vehicle in front or behind in the dedicated lane for autonomous driving when the passing vehicle passes through the passing point.

[0107] Table 2 The time when each autonomous vehicle n enters the detection range

[0108]

[0109] S24: For any vehicle, use the s t formula to calculate the coordinate s of the vehicle from any initial state, including position s0, speed v0, and acceleration a0, to time t1 t1 and the coordinate s of the vehicle to time t2t2 , respectively judge the relationship with the above rectangular boundary. If the following relationship is satisfied, the situation where the trajectory of the autonomous driving vehicle does not intersect with the crossing vehicle includes the situation where the vehicle is far away from the downstream safety distance and the situation where the vehicle is close to the upstream safety distance. The judgment rules are as follows: Figure 5 As shown, specifically:

[0110] S241 is the case where the vehicle is farther than the downstream safety distance. The specific formula is:

[0111] S242 is the case where the vehicle is close to the upstream safety distance, and the specific formula is:

[0112] S243 is the situation where the trajectory of the autonomous driving vehicle intersects with the crossing vehicle. t1 and t2 The following relations are satisfied:

[0113] and

[0114] S25: If there are multiple crossing vehicles within the detection range of the entrance ramp, that is, step S1 provides multiple spatiotemporal windows (rectangles) of crossing vehicles occupying the dedicated lane, then the rectangles are given unique numbers m=1, 2, ..., M according to the order of the crossing vehicles, that is, the rectangles closer to the left are considered first, and then the intersection of the vehicle trajectory in any dedicated lane and the rectangles is determined in turn. The rectangle number of the trajectory that intersects with the vehicle is recorded.

[0115] S26: For the autonomous driving vehicles in the dedicated lane that enter the control range, unique vehicle numbers n=1, 2, ..., N are assigned in order of the vehicle that enters the range first, that is, the vehicle closest to the crossing point first, and all rectangles described in S25 are judged whether they intersect according to the judgment method in S24. The numbers of the vehicles that intersect and the numbers of the corresponding intersection rectangles are recorded.

[0116] In this embodiment, Figure 7 As shown, the solid line is the new control trajectory, the rectangle is the time-space diagram of the crossing vehicle, and the dotted line is the original driving trajectory of the first autonomous driving vehicle (n=1), which will obviously intersect with the time-space window of the first manual driving crossing vehicle (m=1). At this time, it is necessary to enter step S3 to calculate the new control trajectory of the autonomous driving vehicle n=1.

[0117] S3, starting from the autonomous driving vehicle closest to the crossing point, in order from far to near, calculate the trajectory of each autonomous driving vehicle within the control range, solve the optimal acceleration value, speed value and position value of the corresponding autonomous driving vehicle at each time step, and thus obtain the trajectory control instruction;

[0118] In this embodiment, the specific process of S3 is as follows:

[0119] S31: In order to adjust the vehicle acceleration as few times as possible and smoothly avoid crossing vehicles with a smaller control cost, the goal is to minimize the sum of the absolute values ​​of the acceleration of all vehicles within the control range. The optimal trajectory satisfies the following formula:

[0120]

[0121]

[0122] Among them, the first constraint describes the coordinate transfer equation of any vehicle n within the control range, the second constraint describes the speed transfer equation of n, the third and fourth constraints restrict the trajectory from intersecting any rectangle, and the fifth constraint restricts the speed to not exceed the maximum speed limit v max , and is not negative (i.e., no reversing occurs). The sixth constraint limits the safe distance μ between the two vehicles in a row (i.e., no intersection occurs between the two autonomous driving vehicles). The seventh constraint limits the controllable acceleration a of vehicle n in step size i. i,n Can only be selected from a finite number of values, such as a + Indicates the acceleration value specified when the vehicle accelerates, a - The deceleration value specified when the vehicle decelerates. 0 means the vehicle does not accelerate or decelerate.

[0123] S32: To satisfy the above inequality constraints, a concise and feasible solution algorithm is provided.

[0124] S321: For vehicles in the dedicated lane within the control range, calculation is performed in the order of n=1, 2, ..., N.

[0125] S322: Use S24 method to verify whether intersection occurs. If n intersects with the corresponding rectangle m, then immediately accelerate at the current moment, starting from the initial moment t0, the vehicle to a i，n Assign a + , recalculate v i，n and i，n , and calculate s t1,m，n and t2,m，n , until satisfied Except for n=1, the position relationship between n and n-1 needs to be considered. Existence k，n +μ>s k，n-1 , then the safety distance between the front and rear vehicles is insufficient, and it should be mandatory Give a - Slow down vehicle n. If v i，n >v max , or if all step length calculations are completed but not satisfied If the acceleration operation is not feasible, the process needs to be transferred to S323 for deceleration operation. Otherwise, the process directly transfers to S325. The flow chart of step S322 is as follows: Figure 6a shown.

[0126] S323: If the intersection occurs but the acceleration operation is not feasible, the deceleration control is immediately adopted at the current moment, and the vehicle starts from the initial moment t0 according to to a i，n Assign a - , recalculate v i，n and i，n , and calculate s t1,m,n and t2,m，n , until satisfied Except for n=1, the position relationship between n and n-1 needs to be considered. Existence k，n +μ>s k，n-1 , then the safety distance between the front and rear vehicles is insufficient, and it should be mandatory Give a - Slow down vehicle n. If v i+1，n ≤0, then a i，n You can no longer assign a - , should be assigned a value of 0. The flow chart of step S323 is as follows Figure 6b shown.

[0127] S324: If the vehicle implemented deceleration control in the previous round of decision-making, and the intersection in S322 does not occur in this round of decision-making, then the acceleration control is preferentially executed in this round of decision-making to compensate for the speed loss caused by the deceleration control, and the process is the same as S322. If not, enter S325.

[0128] S325: If there is no intersection in this round of decision-making, then press to a i，n Assign the value 0.

[0129] S326: After a time step of Δt, the position coordinates s of all vehicles n = 1, 2, ..., N are updated with the latest observed state values 0，n , speed v 0，n , acceleration a 0，n If the vehicles and windows within the control range change, n=1, 2, ..., N are updated with the latest vehicles and corresponding state parameters, and S1 is re-executed to obtain updated m=1, 2, ..., M and its parameters. Return to S322 to execute a new round of decision making.

[0130] In this embodiment, Figure 7 As shown, n = 1 intersects with a rectangle m = 1, but does not intersect with other rectangles m. At the current moment, the vehicle is immediately accelerated. Starting from the initial moment t0, the vehicle is accelerated. to a i，n Assign a + , recalculate v i，n and i，n , and calculate s t1,m,n and t2,m，n , until satisfied The condition ends. v does not appear in the current calculation object i，n >v max , then the acceleration operation is feasible and directly transfer to S325.

[0131] In S325, if there is a crossover in this round of decision making, a cannot be given i，n Assign 0. At this time, the calculation for n=1 is completed. Then perform S2 and S3 for n=2, 3, 4, 5, 6. When executing step S322 for n=2, the acceleration operation is not feasible, so go to S323 and execute S323. After S323 is executed, S324 is executed sequentially. Since the deceleration control is implemented in this round of decision-making (time t), the acceleration control is preferentially performed on n=2 in the decision-making process of the next round of decision-making (time t+Δt) to compensate for the speed loss caused by the deceleration control. The process is the same as S322.

[0132] Through the above calculation, a deceleration control trajectory sequence of n=2 is obtained, and after the deceleration is completed, acceleration control is implemented to reduce the interval with n=1.

[0133] In this embodiment, the S2 and S3 operations for n=3, 4, 5, 6 are similar to the above process and can be deduced in sequence. The trajectory of the n=1, 2, 3, 4, 5, 6 autonomous driving vehicles avoiding the m=1, 2, 3, 4, 5 crossing vehicles in this embodiment is as follows Figure 8 As shown, the rectangle is the space-time window and the solid line is the trajectory of the autonomous driving vehicle.

[0134] S4: Communicate the trajectory control instructions to all autonomous driving vehicles involved in trajectory adjustment. After a sampling time interval, return to step S1 to perform a new round of calculations and update the status.

[0135] In this embodiment, the specific process of S4 is:

[0136] S41: The roadside controller obtains the acceleration instruction (acceleration, deceleration, no acceleration or deceleration) of each autonomous driving vehicle within the control range through the operation of step S3, and transmits the instruction to all autonomous driving vehicles involved in trajectory adjustment through the communicator.

[0137] S42: Starting from the moment when the autonomous driving vehicle starts to adjust the acceleration, within a sampling time interval Δt, repeat the steps S1, S2, and S3 to calculate the optimal acceleration, speed, and position parameters for each time step in a new round. The calculated acceleration command (acceleration, deceleration, no acceleration or deceleration) is again transmitted to all autonomous driving vehicles involved in trajectory adjustment through the communicator. The time interval between the moment when each autonomous driving vehicle starts to adjust the acceleration and the last adjustment is a sampling time interval.

[0138] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for controlling a fleet in an autonomous driving lane, characterized in that: The method comprises the following steps: S1. Real-time video of the entrance ramp traffic flow. When a crossing vehicle is detected, the position, speed and acceleration data of the crossing vehicle are calculated, the relationship between the position and time of the crossing vehicle is established, and the time-space window of the autonomous driving lane occupied by the crossing vehicle is obtained and calibrated; S2. Calculate the longitudinal trajectory of the autonomous driving vehicle in the dedicated autonomous driving lane, and use the spatiotemporal portion of the trajectory that intersects with the spatiotemporal window of the dedicated autonomous driving lane occupied by the crossing vehicle as the control range boundary; S3, starting from the autonomous driving vehicle closest to the crossing point, in order from far to near, calculate the trajectory of each autonomous driving vehicle within the control range, solve the optimal acceleration value, speed value and position value of the corresponding autonomous driving vehicle at each time step, and thus obtain the trajectory control instruction; S4: The trajectory control instructions are transmitted to all autonomous driving vehicles involved in trajectory adjustment for execution. After that, the process returns to step S1 to perform a new round of calculations and update the status.

2. The method for controlling a fleet in an autonomous driving lane according to claim 1, characterized in that: The specific steps of calculating the position, velocity and acceleration data of the crossing vehicle and establishing the position and time relationship of the crossing vehicle in S1 are: S11. When a crossing vehicle enters the monitoring range, its instantaneous position, instantaneous speed and instantaneous acceleration are monitored; S12, calculating the time and distance of the crossing vehicle accelerating to the maximum speed limit, and establishing a relationship between the position and time of the crossing vehicle; S13. Execute S11 and S12 once for the crossing vehicle at a set interval, update the estimated time when the vehicle arrives at the crossing point, and terminate when its instantaneous position exceeds the crossing point. At the end of the period, the vehicle is removed from the monitoring range.

3. The method for controlling a fleet in an autonomous driving lane according to claim 2, characterized in that: The specific formula for the relationship between the position and time of the crossing vehicle is: Where, t ′ is the time it takes for the crossing vehicle to accelerate to the maximum speed limit, v M is the maximum speed limit, v mt is the instantaneous speed of the crossing vehicle, a mt is the instantaneous acceleration of the crossing vehicle, s′ is the distance between the crossing vehicle and the monitoring point when the crossing vehicle accelerates to the maximum speed limit, that is, the relative position of the crossing vehicle. is the estimated time when the crossing vehicle arrives at the crossing point, L R is the distance between the crossing point and the detection point, and t is the current time.

4. The method for controlling a fleet in an autonomous driving lane according to claim 1, characterized in that: The specific method of marking the space-time window in S1 is to draw a space-time diagram of the space-time window in the form of a rectangle. In the space-time diagram, the top edge of the rectangle represents the downstream safety distance when the crossing vehicle occupies the automatic driving lane, the bottom edge represents the upstream safety distance when the crossing vehicle occupies the automatic driving lane, the left side represents the earliest safety time when the crossing vehicle occupies the automatic driving lane, and the right side represents the latest safety time when the crossing vehicle occupies the automatic driving lane. The specific formula is: L1=L+ε L2=L-ε Where L1 is the downstream safety distance, L2 is the upstream safety distance, t1 is the earliest safety time, t2 is the latest safety time, L is the coordinate of the crossing point, ε is the safety distance between the crossing vehicle and the preceding or following vehicle in the dedicated autonomous driving lane when the crossing vehicle passes through the crossing point, is the moment when the center point of the crossing vehicle passes through the crossing point, and τ is the safe time interval between the crossing vehicle and the preceding or following vehicle in the autonomous driving lane when the crossing vehicle passes through the crossing point.

5. The method for controlling a fleet in an autonomous driving lane according to claim 4, characterized in that: The intersection of the space-time diagram and the longitudinal trajectory of the autonomous driving vehicle in S2 is a collision, and the boundary of the intersection is the control range boundary.

6. The method for controlling a fleet in an autonomous driving lane according to claim 1, characterized in that: The calculation of the longitudinal trajectory of the autonomous driving vehicle in the autonomous driving lane in S2 is specifically to establish a vehicle position and speed transfer equation between two adjacent unit time steps, and to establish a relationship between the position and time of the autonomous driving vehicle, and the specific formula is: v i+1 =v i +a i ·Δt Where t0 is the time of the initial state when calculating the vehicle trajectory; t is the position coordinate s to be calculated t moment; s0 is the coordinate of the initial state when calculating the vehicle trajectory; s t is the displacement of the vehicle at time t; v0 is the velocity of the initial state when calculating the vehicle trajectory; v t is the speed of the vehicle at time t; a0 is the acceleration of the initial state when calculating the vehicle trajectory; a t is the acceleration of the vehicle at time t; Δt is the unit time step, i is the serial number of the time step, that is, the initial moment of the i-th time step, is the floor operator, and % is the remainder operator.

7. The method for controlling a fleet in an autonomous driving lane according to claim 1, characterized in that: The calculation of the trajectory of each autonomous driving vehicle in S3 is specifically to solve the optimal trajectory of the autonomous driving vehicle based on the constraints, and the specific formula is: Where n is the number of the autonomous driving vehicle, a i,n is the acceleration of vehicle n at the initial moment of the ith time step, t is the current moment, Δt is the unit time step, t0 is the moment of the initial state when calculating the vehicle trajectory, and i is the sequence number of the time step, that is, the initial moment of the ith time step.

8. The method for controlling a fleet in a dedicated lane for automatic driving according to claim 7, characterized in that: The constraint conditions for calculating the trajectory of each autonomous driving vehicle in S3 include the first constraint to the seventh constraint, and the specific formula is: In the formula, the first constraint describes the coordinate transfer equation of any vehicle n within the control range, the second constraint describes the speed transfer equation of n, the third and fourth constraints restrict the trajectory from intersecting any rectangle, and the fifth constraint restricts the speed to not exceed the maximum speed limit v max , and is not a negative number, that is, no reversing occurs. The sixth constraint limits the safe distance μ between the two vehicles in succession, that is, there is no intersection between the two autonomous driving vehicles. The seventh constraint limits the controllable acceleration a of vehicle n in step size i. i,n Can only be selected from a finite number of values, such as a + Indicates the acceleration value specified when the vehicle accelerates, a - The deceleration value specified when the vehicle decelerates. 0 means the vehicle does not accelerate or decelerate. i,n is the coordinate of vehicle n at the initial moment of the i-th time distribution, v i,n is the speed of vehicle n at the initial moment of the i-th time span, s t1 is the coordinate of the autonomous driving vehicle at time t1, s t2 is the coordinate of the autonomous driving vehicle at time t2, l is the length of the autonomous driving vehicle, L is the coordinate of the crossing point, ε is half the distance between the two autonomous driving vehicles in the autonomous driving lane when the crossing vehicle passes the crossing point, and v max is the maximum speed limit, a + Indicates the acceleration value specified when the vehicle accelerates, a - It is the deceleration value specified when the vehicle decelerates. 0 means the vehicle does not accelerate or decelerate. μ is the safe distance between vehicles.

9. The method for controlling a fleet in an autonomous driving lane according to claim 1, characterized in that: When it is determined in S2 that the autonomous driving vehicle trajectory intersects with the spatiotemporal window of the crossing vehicle occupying the dedicated lane, the situation where the autonomous driving vehicle trajectory intersects with the crossing vehicle includes the situation where the vehicle is far away from the downstream safety distance and the situation where the vehicle is close to the upstream safety distance, and the specific formula is: In the formula, s t1 is the coordinate of the autonomous driving vehicle at time t1, s t2 is the coordinate of the autonomous driving vehicle at time t2, l is the length of the autonomous driving vehicle, L is the coordinate of the crossing point, and ε is half the distance between the two autonomous driving vehicles in the autonomous driving lane when the crossing vehicle passes the crossing point.

10. A fleet control system in an autonomous driving dedicated lane, characterized in that: The system is applied to work according to a method for controlling a fleet in an autonomous driving lane as described in any one of claims 1 to 9, and the system includes a video detector, a roadside controller and a communicator; The video detector uses a general road traffic monitoring camera to completely capture the ramp and acceleration lane area, monitor and identify vehicles on the ramp in real time, and use the image processing technology configured therein to calculate the real-time position coordinates, speed and acceleration data of the vehicle, and send the data to the roadside controller; The roadside controller is placed within the control area and is used to receive data from the video detector and the autonomous driving vehicle, establish a relationship between the position and time of the crossing vehicle, calculate the longitudinal trajectory of the autonomous driving vehicle in the autonomous driving lane, calculate the trajectory control command, and communicate it to all autonomous driving vehicles involved in trajectory adjustment; The communicator is installed in the video detector, the roadside controller and the autonomous driving vehicle. The video detector sends data of vehicles crossing the ramp to the roadside controller through the communicator. The autonomous driving vehicle sends vehicle trajectory data in the autonomous driving lane to the roadside controller through the communicator. The roadside controller issues trajectory control instructions to the autonomous driving vehicle through the communicator.

Images