Long queue control method under multi-head vehicle structure based on cellular automaton

Abstract

The invention discloses a cellular automaton-based long queue control method under a multi-head vehicle structure. The method comprises the following steps: S1, establishing a long queue control model; s2, acquiring driving data of long-queue vehicles and vehicles around the long-queue vehicles through roadside facilities, and obtaining an optimal sub-queue head vehicle sequence by using a long-queue control model; s3, the vehicles in the long queue output the speeds and positions of the long queue head vehicle, the sub-queue head vehicle and the sub-queue following vehicles through a long queue control model according to the optimal sub-queue head vehicle sequence information issued by the roadside facility in real time and in combination with the state of multiple front vehicles in the queue and the mutual influence among the vehicles in the queue. Therefore, the longitudinal driving of the long queue is controlled. According to the method, the long queue is effectively divided into a plurality of dynamically changing sub-queues, so that the speed disturbance influence caused by the speed change of the vehicles in front of the long queue is effectively reduced, the speed fluctuation time and the speed fluctuation range of the vehicles in the long queue are reduced, and the method has relatively high stability and safety.

Description

Technical Field

[0001] The present invention relates to the field of intelligent vehicle application technology, and in particular to a long queue control method under a multi-vehicle structure based on cellular automata. Background Art

[0002] Autonomous driving vehicle platoons can effectively improve road capacity, enhance traffic safety, and reduce fuel consumption. In recent years, with the development of vehicle networking and vehicle-road collaboration technology, the connection range of vehicles has been greatly expanded, making "long platoons" an important development trend of platoon technology.

[0003] The number of vehicles in a long queue can reach dozens or even hundreds. By forming a long queue, the advantages of the queue in improving traffic efficiency and reducing energy consumption can be fully utilized. Long queues of vehicles have received more and more attention. At present, there are few studies on long queues. The current research mainly divides the long queue into multiple sub-queues and assigns a fixed sub-queue head vehicle to each sub-queue to strengthen the control of the long queue. However, these studies have relatively fixed selection of sub-queue head vehicles and do not dynamically optimize the sub-queue head vehicle.

[0004] However, as external traffic conditions change, speed disturbances will quickly spread to the entire long queue, causing speed changes in the vehicles in the long queue. In this case, the strategy of fixing the head vehicle of the sub-queue cannot adjust the number of following vehicles in the sub-queue in time, resulting in the inability to reduce the adverse effects of speed disturbances in time. Long queues are in an unstable state for a long time, which will affect the stability, safety and efficiency of the long queue. In contrast, the dynamic optimization of the head vehicle of the sub-queue can adjust the order of the head vehicle of the sub-queue and the number of its member vehicles in time according to changes in the surrounding traffic. This can further reduce the speed difference between vehicles in the long queue and stabilize the long queue quickly. Moreover, this can also keep the vehicles in the long queue at an ideal distance, thereby improving the stability and safety of the long queue.

[0005] In addition, in the existing research on the control of the queue, only the influence of one vehicle in front of the queue is considered. However, the speed synchronization of the queue is crucial to the control of the queue. In a convoy, speed synchronization allows vehicles to travel safely at a higher speed. Due to the use of vehicle networking technology, more leading vehicles can be considered in the queue control. Considering multiple leading vehicles and synchronizing the speed with these vehicles in advance can reduce the speed fluctuations of vehicles in the queue caused by the disturbance of the leading vehicles. This will make the traffic flow wave dissipate as quickly as possible during the propagation of the queue, thereby improving the stability and safety of the queue. Therefore, it is necessary to consider the influence of multiple leading vehicles in the control of long queues.

[0006] At present, the research on long queues is mainly from the perspective of reducing communication delay or improving communication reliability. The long queue is divided into multiple sub-queues through the head vehicle of the sub-queue or some intermediate vehicles in the long queue, and the consistency of vehicle behavior in the long queue is improved through the information transmission of the head vehicle of the sub-queue or the relay vehicle. However, these studies have relatively fixed the selection of the head vehicle or relay vehicle of the sub-queue, and have not dynamically found the optimal distribution of the head vehicle of the sub-queue in the long queue according to the changes in traffic conditions. They have also not considered the impact of multiple leading vehicles in a longer range on the entire long queue. There are also few studies on the control of the head vehicle of the sub-queue and the control of the following vehicles under the leadership of the head vehicle of the sub-queue. Summary of the invention

[0007] The purpose of the present invention is to provide a long queue control method under a multi-car structure based on cellular automata, which dynamically optimizes the sequence of head cars in the sub-queue according to changes in the external traffic environment, and updates the speed and position of the vehicles in the long queue according to the rules of cellular automata, thereby improving the stability, safety and efficiency of the long queue vehicles.

[0008] To achieve the above object, the present invention provides the following technical solution: a long queue control method under a multi-car structure based on cellular automata, comprising the following steps:

[0009] S1. Establish a long queue control model, including a multi-car optimization model, a queue head car control model, a sub-queue head car control model and a sub-queue following car control model;

[0010] S2, collecting driving data of long queue vehicles and surrounding vehicles through roadside facilities, and inputting the driving data into the multi-vehicle optimization model to output the optimal sub-queue head vehicle sequence;

[0011] S3. The vehicles in the long queue are based on the optimal sub-queue head vehicle sequence information released in real time by the roadside facilities and combined with the status of multiple front vehicles in the queue and the mutual influence between vehicles in the queue. The queue head vehicle control model, the sub-queue head vehicle control model and the sub-queue following vehicle control model output the speed and position of the long queue head vehicle, the sub-queue head vehicle and the sub-queue following vehicle, so as to control the longitudinal travel of the long queue.

[0012] In some embodiments, the present invention starts from the stability of the long queue and the comfort of passengers, and establishes a dynamic multi-car optimization model based on optimization theory. The stability of the long queue is represented by the degree of deviation between the speed of the head car of the sub-queue and the head car of the queue at a specified time step. The closer the speed of the head car of the sub-queue is to the speed of the head car, the better the stability of the long queue. Passenger comfort is represented by the acceleration change rate of the head car of the sub-queue. The smaller the acceleration change rate of the head car of the sub-queue, the higher the passenger comfort. The multi-car optimization model is:

[0013]

[0014] In the formula, Z virtual is the objective function; is the stability evaluation index of the head vehicle of the i-th sub-queue; is the passenger comfort evaluation index of the head car of the i-th sub-queue; γ i Whether the vehicle in the queue is the head vehicle of the sub-queue; v max is the maximum speed of the platoon vehicles; a min is the minimum deceleration of the platoon vehicles; a max is the maximum acceleration of the vehicles in the queue; n is the total number of head vehicles in the sub-queue; N is the total number of vehicles in the queue; is the acceleration of the head vehicle of the i-th sub-queue at time t; is the acceleration change of the head vehicle of the i-th sub-queue at time t and time t-1; is the acceleration of the head vehicle of the i-th sub-queue at time t-1; is the speed difference between the head car and the head car of the i-th sub-queue of the long queue at time t; is the speed of the first vehicle in the queue at time t; is the speed of the first vehicle in the i-th sub-queue in the long queue at time t.

[0015] In some embodiments, establishing a queue head vehicle control model includes the following steps:

[0016] Based on the fact that the head car in the queue is affected by the safety constraints of the car in front of it, as well as the speed and acceleration of multiple cars in front of the queue, an acceleration model of the head car in the queue is established.

[0017] In the acceleration model of the first vehicle in the queue, it is considered that the first vehicle in the queue needs to maintain the desired safe speed with the vehicle in front of it;

[0018] In the acceleration model of the leading vehicle in a queue, different weights are assigned to the leading vehicles according to their different influences on the leading vehicle at different positions at the same time.

[0019] Based on the maximum speed limit of the vehicle and the output of the acceleration model of the first vehicle in the queue, the speed evolution rule of the first vehicle in the queue is established;

[0020] The position of the head vehicle in a long queue is updated based on the speed evolution update rule of the head vehicle in the long queue.

[0021] In some embodiments, from a safety perspective, the first vehicle in the queue needs to maintain a desired safe speed with the vehicle in front of it. At the same time, based on the long communication perception range of the autonomous driving vehicle, the impact of the speed and acceleration of the vehicle in front of the queue on the first vehicle is considered. The acceleration model of the first vehicle in the queue is:

[0022]

[0023] The safe speed in the queue head vehicle control model is:

[0024]

[0025] Since the influence weight of multiple leading vehicles on the leading vehicle is mainly related to the speed difference and spacing difference between the multiple leading vehicles and the leading vehicle, the influence weight of multiple leading vehicles on the leading vehicle is:

[0026]

[0027] In the formula, is the acceleration of the first vehicle in the queue at time t; is the required safe speed of the leading vehicle at time t; v l (t) is the speed of the leading vehicle at time t; M is the number of leading vehicles; k m is the influence weight of the mth front vehicle on the leading vehicle; a em (t) is the acceleration of the mth vehicle in front at time t; v em (t) is the speed of the mth vehicle in front at time t; k0, k1, k2 are the weight coefficients of each item; b l is the maximum deceleration of the head car in a long queue; τ is the reaction time of the head car; x e (t) is the position of the vehicle immediately preceding the head vehicle in the queue at time t; x l (t) is the position of the lead vehicle at time t; l e is the vehicle length of the vehicle in front; v e (t) is the speed of the vehicle immediately preceding the head vehicle in the queue at time t.

[0028] In some embodiments, establishing a sub-queue head vehicle control model includes the following steps:

[0029] Based on the fact that the head vehicle of a sub-queue is affected by the safety constraints of its preceding vehicle, as well as by the preceding head vehicle of the sub-queue and the preceding vehicles in front of the queue, an acceleration model of the head vehicle of the sub-queue is established.

[0030] In the acceleration model of the head vehicle of a sub-queue, it is considered that the head vehicle of a sub-queue needs to maintain the desired safe speed with the vehicle in front of it;

[0031] In the acceleration model of the head vehicle of a sub-queue, different weights are assigned to the head vehicles according to their different influences on the head vehicle of the sub-queue.

[0032] Based on the maximum speed limit of the vehicle and the output of the acceleration model of the head vehicle in the sub-queue, the speed evolution rule of the head vehicle in the sub-queue is established;

[0033] The position of the head vehicle in the sub-queue is updated based on the speed evolution update rule of the head vehicle in the sub-queue.

[0034] In some embodiments, the safety of the head vehicle of a sub-queue and the preceding vehicle is considered while the influence of the preceding head vehicle of the sub-queue is considered, so that the speed of the vehicle in the sub-queue is synchronized with the speed of the preceding vehicle as much as possible. The acceleration model of the head vehicle of the sub-queue is:

[0035]

[0036] The safe speed of the sub-queue head vehicle control model is:

[0037]

[0038] The influence weight of multiple leading vehicles on the head vehicle of a sub-queue is derived from the positive and negative correlation between the speed difference and spacing difference between multiple leading vehicles and the head vehicle of a sub-queue. The influence weight of multiple leading vehicles on the head vehicle of a sub-queue is:

[0039]

[0040] Where j is the head vehicle of the jth sub-queue in the long queue; is the required safe speed of the jth head vehicle in the sub-queue at time t; v j (t) is the speed of the first vehicle in the j-th sub-queue at time t; v j-1 (t) is the speed of the j-1th head vehicle in the sub-queue at time t; is the influence weight of the mth front vehicle on the jth head vehicle of the sub-queue; k3, k4, k5 are the weight coefficients of each item; b j is the maximum deceleration of the jth head vehicle in the sub-queue; x j+1 (t) is the position of the vehicle in front of the j-th sub-queue at time t; x j (t) is the position of the head vehicle of the j-th sub-queue at time t; l p is the length of the vehicle in front of the j-th sub-queue head vehicle; v j+1 (t) is the speed of the vehicle in front of the j-th sub-queue at time t.

[0041] In some embodiments, establishing a sub-queue following vehicle control model includes the following steps:

[0042] Based on the fact that the following vehicle in a sub-queue is affected by its preceding vehicle and the head vehicle of the sub-queue, an acceleration model of the following vehicle in the sub-queue is established.

[0043] In the sub-queue following vehicle acceleration model, it is considered that the sub-queue following vehicle needs to maintain the desired safety speed with its leading vehicle;

[0044] Based on the maximum speed limit of the vehicle and the output of the acceleration model of the sub-queued following vehicle, the speed evolution rule of the sub-queued following vehicle is established;

[0045] The position of the following vehicle in the sub-queue is updated based on the speed evolution update rule of the following vehicle in the sub-queue.

[0046] In some embodiments, the queue head vehicle and the sub-queue head vehicle divide the queue into multiple sub-queues. The following vehicles in the sub-queue are affected by the sub-queue head vehicle and its preceding vehicle. The acceleration model of the sub-queue following vehicle is:

[0047]

[0048] The safe speed of the sub-queue following vehicle control model is:

[0049]

[0050] In the formula, is the acceleration of the i-th following vehicle under the leadership of the j-th sub-platoon leader at time t; is the required safe speed of the i-th following vehicle under the leadership of the j-th sub-platoon leader at time t; is the speed of the i-th following vehicle under the j-th sub-queue head vehicle at time t; v j (t) is the speed of the first vehicle in the j-th sub-queue at time t; k0 and k6 are weight coefficients; b i is the maximum deceleration of the following vehicle; is the position of the front vehicle of the i-th following vehicle under the leadership of the j-th sub-queue head vehicle at time t; is the position of the i-th following vehicle under the j-th sub-queue head vehicle at time t; l i+1 The length of the vehicle in front of the following vehicle; It is the speed of the front vehicle of the i-th following vehicle under the leadership of the j-th sub-queue head vehicle at time t.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] (1) The present invention can timely adjust the sequence of the head vehicles of the sub-queues according to changes in the external traffic environment, effectively dividing a long queue into multiple sub-queues, thereby effectively reducing the speed disturbance caused by the speed change of the vehicle in front of the queue. As the size of the queue increases, the number of head vehicles in the sub-queues also increases, which is conducive to the control of long queues.

[0053] (2) Compared with the existing control model, in various acceleration and deceleration disturbance scenarios, the model proposed by the present invention reacts faster to the acceleration and deceleration of the vehicle in front of the long queue, the increase or decrease of the speed of the vehicles in the queue is also faster, the speed fluctuation time of the entire queue is short, and the stability is better. In addition, as the size of the queue increases, the stability advantage of the queue controlled by the model of the present invention becomes more obvious, and the speed fluctuation time of the entire queue vehicles can be reduced by up to 33%.

[0054] (3) Compared with the existing control model, under various acceleration and deceleration disturbance scenarios, the speed fluctuation of the platoon vehicles under the control of the model proposed by the present invention is smaller, reducing the speed fluctuation range by 40%. This shows that the platoon safety under the control of the model of the present invention is better. In particular, when the platoon is large, the existing control model is prone to lose control of the vehicles at the end of the platoon, which shows that the model proposed by the present invention has better control over longer platoons and better safety.

[0055] (4) Compared with the existing control model, the queue length under the control of the model proposed by the present invention is smaller than that under the existing control model, indicating that the model proposed by the present invention is more effective in controlling the queue length. In addition, the larger the queue size, the better the control effect of the model proposed by the present invention on the queue length, indicating that the queue travel efficiency under the control of the model of the present invention is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of a long queue of the present invention.

[0057] Figure 2 FIG. 5 is a schematic diagram of the dynamic change of the head vehicle sequence of a sub-queue under different queue sizes when the first vehicle among multiple leading vehicles decelerates at the maximum deceleration for 3 seconds or accelerates at the maximum acceleration for 3 seconds at the 500th time step in an embodiment of the present invention. During the process of queue speed change, FIG. (a) and (c) correspond to the dynamic change diagrams of the head vehicle sequence of a sub-queue with queue sizes of 30 vehicles and 80 vehicles when the first vehicle among multiple leading vehicles decelerates at the maximum deceleration, and FIG. (b) and (d) correspond to the dynamic change diagrams of the head vehicle sequence of a sub-queue with queue sizes of 30 vehicles and 80 vehicles when the first vehicle among multiple leading vehicles accelerates at the maximum acceleration, respectively.

[0058] Figure 3 The figure is a speed heat map corresponding to different queue sizes when the speed of the first vehicle among the multiple front vehicles in the embodiment of the present invention drops from 30 m / s to 18 m / s at the 500th time step; wherein, Figures (a) and (b) are speed heat maps when the queue size is 30 vehicles, (c) and (d) are speed heat maps when the queue size is 50 vehicles, and (e) and (f) are speed heat maps when the queue size is 80 vehicles, wherein (a), (c) and (e) are speed heat maps corresponding to the model proposed in the present invention, and (b), (d) and (f) are speed heat maps corresponding to the existing control model.

[0059] Figure 4Figure 1 is a box plot of queue speed changes at different time steps under different queue sizes, where the first vehicle among multiple front vehicles in an embodiment of the present invention decelerates at the maximum deceleration for 3s at the 500th time step, and the vehicle speed drops from 30m / s to 18m / s; wherein, Figures (a) and (b) are box plots of queue speed changes when the queue size is 30 vehicles, Figures (c) and (d) are box plots of queue speed changes when the queue size is 50 vehicles, and Figures (e) and (f) are box plots of queue speed changes when the queue size is 80 vehicles, wherein Figures (a), (c) and (e) are box plots of queue speed changes corresponding to the model proposed by the present invention, and Figures (b), (d) and (f) are box plots of queue speed changes corresponding to the existing control model.

[0060] Figure 5 Schematic diagram of queue lengths corresponding to the control model proposed by the present invention and the existing control model when the first vehicle among multiple leading vehicles in an embodiment of the present invention decelerates at the maximum deceleration for 3 seconds and accelerates at the maximum acceleration for 3 seconds at the 500th time step; wherein Figure (a) is a queue length diagram under the control of the two models when the first vehicle among multiple leading vehicles decelerates at the maximum deceleration for 3 seconds and the long queue size is 30 vehicles, 50 vehicles and 80 vehicles; Figure (b) is a queue length diagram under the control of the two models when the first vehicle among multiple leading vehicles accelerates at the maximum acceleration for 3 seconds and the long queue size is 30 vehicles, 50 vehicles and 80 vehicles. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0062] See also Figure 1-Figure 5 A long queue control method under a multi-car structure based on cellular automata comprises the following steps:

[0063] 1. Establish a multi-car optimization model;

[0064] This embodiment establishes a dynamic multi-car optimization model based on the stability of a long queue and the comfort of passengers. The stability of the long queue is represented by the degree of deviation between the speed of the head car of the sub-queue and the head car of the queue at a specified time step. The closer the speed of the head car of the sub-queue is to the speed of the head car, the better the stability of the long queue. Passenger comfort is represented by the acceleration change rate of the head car of the sub-queue. The smaller the acceleration change rate of the head car of the sub-queue, the higher the passenger comfort.

[0065] The stability of the long queue is expressed by the ratio of the square of the speed difference between the head car and the head car of the subqueue at a specified time step to the speed of the head car:

[0066]

[0067] The passenger comfort is expressed by the ratio of the square of the acceleration difference between the previous time step and the current time step of the head car of the sub-queue to the acceleration of the previous time step:

[0068]

[0069] The objective function is determined by two evaluation indicators: queue stability and passenger comfort:

[0070]

[0071] In the formula, Z virtual is the objective function; is the stability evaluation index of the head vehicle of the i-th sub-queue; is the passenger comfort evaluation index of the head car of the i-th sub-queue; γ i Whether the vehicle in the queue is the head vehicle of the sub-queue; is the acceleration of the head vehicle of the i-th sub-queue at time t; is the acceleration change of the head vehicle of the i-th sub-queue at time t and time t-1; is the acceleration of the head vehicle of the ith sub-queue at time t-1; is the speed difference between the head car and the head car of the i-th sub-queue of the long queue at time t; is the speed of the first vehicle in the queue at time t; is the speed of the first vehicle in the i-th sub-queue in the long queue at time t.

[0072] Since the number of vehicles in a long queue can reach dozens or even hundreds, this embodiment sets the total number of vehicles in the long queue to be greater than or equal to 30 vehicles. Assuming that the number of vehicles in a sub-queue is greater than or equal to 3 vehicles, the total number of vehicles in the long queue and the total number of head vehicles in the sub-queue should satisfy:

[0073]

[0074] Where n is the total number of head vehicles in the sub-queue; N is the total number of vehicles in the queue.

[0075] In this embodiment, the head vehicle of the long queue sub-queue is defined by a 0-1 variable, that is, γ i =1 means that the i-th vehicle in the long queue is the head vehicle of the sub-queue, γ i = 0 means that the i-th vehicle in the long queue is the following vehicle, from which we can conclude:

[0076]

[0077] The speed and acceleration of vehicles in a long queue are also limited by the maximum speed, minimum acceleration and maximum acceleration of the vehicles, so:

[0078]

[0079] In the formula, v max is the maximum speed of the platoon vehicles; a min is the minimum deceleration of the platoon vehicles; a max is the maximum acceleration of the vehicles in the platoon.

[0080] In summary, the dynamic multi-car optimization model is:

[0081]

[0082] 2. After establishing the multi-vehicle optimization model, based on the cellular automaton theory and combined with the impact of multiple leading vehicles in front of the queue on the queue, as well as the mutual influence between vehicles within the queue, the queue head vehicle control model, the sub-queue head vehicle control model and the sub-queue following vehicle control model are established.

[0083] 2.1、Queue head vehicle control model.

[0084] (1) Speed ​​evolution rules;

[0085] From a safety perspective, the head vehicle in the queue needs to maintain the desired safe speed with the vehicle in front of it. At the same time, since the communication perception range of the autonomous driving vehicle is relatively long, this embodiment considers the impact of the speed and acceleration of multiple vehicles in front of the queue on the head vehicle. In addition, different weights are given to multiple vehicles in front of them according to their different impacts on the head vehicle at the same time. The acceleration model of the head vehicle in the queue is:

[0086]

[0087] If v em (t) = v l (t), that is, at time t, the speed of the first car is equal to the speeds of the multiple cars in front of the queue. Then the acceleration model of the first car in the queue is:

[0088]

[0089] In the acceleration model of the leading vehicle in a long queue, the safe speed is obtained using the Gipps safety distance model:

[0090] The safe speed in the queue head vehicle control model is:

[0091]

[0092] In the formula, is the acceleration of the first vehicle in the queue at time t; is the required safe speed of the leading vehicle at time t; v l (t) is the speed of the leading vehicle at time t; M is the number of leading vehicles; k m is the influence weight of the mth front vehicle on the leading vehicle; a em (t) is the acceleration of the mth vehicle in front at time t; v em (t) is the speed of the mth vehicle in front at time t; k0, k1, k2 are the weight coefficients of each item, where k0 = 1.33, k1 = 0.18, k2 = 0.67; b l is the maximum deceleration of the head car in a long queue; τ is the reaction time of the head car; x e (t) is the position of the vehicle immediately preceding the head vehicle in the queue at time t; x l (t) is the position of the lead vehicle at time t; l e is the vehicle length of the preceding vehicle; v e (t) is the speed of the vehicle immediately preceding the head vehicle in the queue at time t.

[0093] In the acceleration model of the leading vehicle in a queue, the weight of the influence of multiple leading vehicles on the leading vehicle is mainly related to the speed difference and spacing difference between the multiple leading vehicles and the leading vehicle. The greater the speed difference between the multiple leading vehicles and the leading vehicle, the greater the influence of the multiple leading vehicles on the leading vehicle, indicating that the weight is positively correlated with the speed difference. The greater the spacing difference between the multiple leading vehicles and the leading vehicle, the smaller the influence of the multiple leading vehicles on the leading vehicle, indicating that the weight is negatively correlated with the spacing difference. Therefore, the weight of multiple leading vehicles on the leading vehicle can be expressed by its positive and negative correlation with the speed difference and spacing difference. The formula for the influence weight of the mth leading vehicle on the leading vehicle is as follows:

[0094]

[0095] The acceleration of the first car in a long queue is obtained based on the acceleration model of the first car in the queue Based on this, considering the maximum speed limit of vehicles, the following evolution rule of the speed of the first vehicle in the queue is obtained:

[0096]

[0097] Where δ is the headway time.

[0098] (2) Location update rules;

[0099] v l (t+1)→x l (t)+δv l (t+1) (14);

[0100] Based on the speed evolution update rule of the head car in a long queue, the position of the head car in the queue is updated.

[0101] 2.2. Sub-queue head vehicle control model.

[0102] (1) Speed ​​evolution rules;

[0103] In the control of the head vehicle of a sub-queue, the safety of the head vehicle of the sub-queue and the preceding vehicle is considered first. Secondly, the influence of the preceding head vehicle of the sub-queue is considered, so that the speed of the vehicle in the sub-queue can be synchronized with the speed of the preceding vehicle as much as possible. In addition, the influence of multiple preceding vehicles on the head vehicle of the sub-queue is also considered. Different weights are given to the multiple preceding vehicles according to their different influences on the head vehicle of the sub-queue. Based on the above factors, the acceleration model of the head vehicle of the sub-queue is:

[0104]

[0105] If v em (t) = v j (t), that is, at time t, the speed of the head car in the sub-queue is equal to the speed of the multiple front cars in front of the queue. Then the acceleration model of the head car in the queue is:

[0106]

[0107] In the acceleration model of the head vehicle of a sub-queue, the safe speed between the head vehicle of a sub-queue and the preceding vehicle is derived from the Gipps safety distance model:

[0108]

[0109] In the acceleration model of the head vehicle of a sub-queue, the weight of the influence of multiple leading vehicles on the head vehicle of a sub-queue is derived from the positive and negative correlation between the speed difference and the spacing difference between the multiple leading vehicles and the head vehicle of the sub-queue. The weight of the influence of the mth leading vehicle on the jth head vehicle of a sub-queue is:

[0110]

[0111] Where j is the head vehicle of the jth sub-queue in the long queue; is the required safe speed of the jth head vehicle in the sub-queue at time t; v j (t) is the speed of the first vehicle in the j-th sub-queue at time t; v j-1 (t) is the speed of the j-1th head vehicle in the sub-queue at time t; is the influence weight of the mth front vehicle on the jth head vehicle of the sub-queue; k3, k4, k5 are the weight coefficients of each item, among which k3 = 0.73, k4 = 0.14, k5 = 0.82; b j is the maximum deceleration of the jth head vehicle in the sub-queue; x j+1 (t) is the position of the vehicle in front of the j-th sub-queue at time t; x j (t) is the position of the head vehicle of the j-th sub-queue at time t; l p is the length of the vehicle in front of the j-th sub-queue head vehicle; v j+1(t) is the speed of the vehicle in front of the j-th sub-queue at time t.

[0112] Based on the acceleration model of the head vehicle in the sub-queue, the acceleration of the j-th head vehicle in the sub-queue is obtained. On the basis of , considering the maximum speed limit of vehicles, the speed evolution rule of the j-th sub-queue head vehicle can be obtained:

[0113]

[0114] Where δ is the headway time.

[0115] (2) Location update rules;

[0116] v j (t+1)→x j (t)+δv j (t+1) (20);

[0117] Based on the speed evolution update rule of the head vehicle in the sub-queue, the position of the head vehicle in the sub-queue is updated.

[0118] 2.3. Sub-queue following vehicle control model.

[0119] (1) Speed ​​evolution rules;

[0120] The queue head car and the sub-queue head car divide the queue into multiple sub-queues. The following cars in the sub-queue are affected by the sub-queue head car. In addition, the following car is also affected by the safety constraints from its front car, so the following car acceleration model is:

[0121]

[0122] The safe speed of the following vehicle and the preceding vehicle in the sub-queue is still derived using the Gipps safety distance model:

[0123]

[0124] In the formula, is the acceleration of the i-th following vehicle under the leadership of the j-th sub-platoon leader at time t; is the required safe speed of the i-th following vehicle under the leadership of the j-th sub-platoon leader at time t; is the speed of the i-th following vehicle under the j-th sub-queue head vehicle at time t; v j (t) is the speed of the jth vehicle at the head of the sub-queue at time t; k0 and k6 are weight coefficients, where k6 = 0.74; b i is the maximum deceleration of the following vehicle; is the position of the front vehicle of the i-th following vehicle under the leadership of the j-th sub-queue head vehicle at time t; is the position of the i-th following vehicle under the j-th sub-queue head vehicle at time t; l i+1 The length of the vehicle in front of the following vehicle; It is the speed of the front vehicle of the i-th following vehicle under the leadership of the j-th sub-queue head vehicle at time t.

[0125] The acceleration model of the sub-queue following vehicle is used to obtain On this basis, considering the maximum speed limit of the vehicle, the speed evolution rule of the following vehicle is obtained:

[0126]

[0127] Where δ is the headway time.

[0128] (2) Location update rules;

[0129]

[0130] Based on the speed evolution update rule of the sub-queue following vehicle, the position of the sub-queue following vehicle is updated.

[0131] 3. Collect driving data (including speed, acceleration, lateral position, longitudinal position, etc.) of long-queue vehicles and surrounding vehicles through roadside facilities (including edge computing boxes, video detection equipment, radar equipment, etc.), input the driving data into the multi-vehicle optimization model, and use the genetic algorithm to solve the model to output the optimal sub-queue head vehicle sequence.

[0132] 4. Roadside facilities publish real-time optimal sub-queue head vehicle sequence information to vehicles in the long queue. Vehicles in the long queue obtain the optimal sub-queue head vehicle sequence information published in real time by roadside facilities through on-board equipment (including edge computing boxes, V2X equipment, etc.), and obtain the driving data of vehicles within a certain range before and after them through on-board equipment.

[0133] The queue head vehicle control model, sub-queue head vehicle control model and sub-queue following vehicle control model are based on the optimal sub-queue head vehicle sequence. The status of multiple vehicles in front of the queue, the safety speed limit of the vehicle and its front vehicle, the speed difference between the adjacent sub-queue head vehicles, the speed difference between the sub-queue following vehicle and the sub-queue head vehicle, and the vehicle maximum speed limit are introduced into the model, and the speed and position of the long queue head vehicle, sub-queue head vehicle and sub-queue following vehicle are output to control the longitudinal travel of the long queue.

[0134] In a specific embodiment, a simulation platform is built using Python, the length of the single-lane road is set to be infinite, three vehicles are set in front of the long queue, the cell length is set to 0.5m, the maximum speed of the vehicle is set to 37.5m / s (75 cells / s), the maximum deceleration of the vehicle is set to -3.5m / s2, and the maximum acceleration of the vehicle is set to 3m / s 2. In the simulation, the queue is controlled by the model proposed in the present invention and the model in the prior art (V. Milanés, SE Shladover, Modeling cooperative and autonomous adaptive cruise control dynamic responses using experimental data, Transportation Research Part C-Emerging Technologies, 48 ​​(2014): 285-300). Multiple front vehicles are controlled by the model in the prior art (D. Yang, XP Qiu, LN Ma, DH Wu, LL Zhu, HB Liang, Cellular Automata-Based Modeling and Simulation of a Mixed Traffic Flow of Manual and Automated Vehicles, Transportation Research Record, 2622 (2017): 105-116), which includes acceleration rules, deceleration rules and position rules. At the beginning of the simulation, the first front vehicle (the frontmost one among the three front vehicles) is driven at a speed of 2.5 m / s, and the speed of other vehicles is 0 m / s. After the speeds of all vehicles are stable, the first front vehicle is subjected to acceleration or deceleration disturbance at a specified time step. Based on the state changes of the multiple front vehicles, all vehicles in the long queue will make corresponding acceleration and deceleration reactions, and update their speed and position according to the proposed speed evolution rules and position rules.

[0135] like Figure 2 As shown in the figure, no matter what the queue size is, during the queue speed fluctuation, the sub-queue head car sequence has been changing dynamically. When the queue speed tends to be stable, the sub-queue head car sequence also tends to be stable. This shows that the multi-car optimization model of the present invention can dynamically adjust the sub-queue head car sequence in a timely manner according to changes in the external traffic environment. It can also be seen from the figure that as the queue size increases, the number of sub-queue head cars also increases, indicating that the larger the queue size, the more sub-queues there are. Therefore, the sub-queue head car can effectively divide the long queue into multiple sub-queues, which is conducive to the control of the long queue.

[0136] like Figure 3 As shown, Figure 3It can be seen from (a) and (b) that after the deceleration disturbance of multiple leading vehicles, the speed of the long queue fluctuates greatly for a certain period of time and then tends to be stable. The speed fluctuation of the queue controlled by the present invention is mainly concentrated in the range of 500-540 steps, while the speed fluctuation of the queue controlled by the CACC model is mainly concentrated in the range of 500-560 steps. This shows that the speed fluctuation time of the queue controlled by the present invention is shorter and the stability is higher. Figure 3 The same trend can be seen in (c), (d), (e), and (f), except that as the queue size increases, the queue controlled by the existing control model takes longer to stabilize its speed. Figure 3 It can also be seen that when the model proposed by the present invention is used, the color of the speed of the vehicles in the queue is relatively light, and there is almost no dark blue. This shows that the queue controlled by the present invention reacts faster to the deceleration of multiple leading vehicles and reduces the speed faster, so the stability is better.

[0137] like Figure 4 As shown in the figure, it can be seen that no matter what the size of the queue is, the speed box of the queue under the control of the model of the present invention is smaller than the speed box of the queue under the control of the existing model, so the control method proposed by the present invention has higher safety. In particular, when the size of the queue is 80 vehicles, the average speed of the queue under the control of the model of the present invention is reduced to 24m / s at 540 steps, and the average speed of the queue fluctuates less in the subsequent 40 steps, only fluctuating between 2m / s-4m / s. However, the average speed of the queue under the control of the existing model is only reduced to 27m / s at 540 steps, and the speed fluctuation range is as high as 10m / s in the subsequent 40 steps. This shows that when the model proposed by the present invention responds to the deceleration disturbance, it first decelerates quickly, and then the speed changes relatively smoothly. This also further illustrates that the larger the size of the queue, the better the control method of the present invention performs in terms of safety.

[0138] like Figure 5As shown in Figure (a), it can be seen that in the deceleration disturbance scenario, when the long queue size is 30 vehicles, 50 vehicles and 80 vehicles, the queue length under the control of the present invention is stable at 759m, 1276.5m and 2095m, while the queue length under the control of the existing control model is stable at 786m, 1344m and 2171m, which shows that in this scenario, the queue length under the control of the present invention is shorter and more efficient. As shown in Figure (b), in the acceleration disturbance scenario, when the long queue size is 30 vehicles, 50 vehicles and 80 vehicles, the queue length under the control of the present invention is stable at 1258.5m, 2065m and 3295.5m, while the queue length under the control of the existing control model is stable at 1408.5, 2373m and 3817m, which also shows that in the acceleration scenario, the queue length controlled by the present invention is smaller than the queue length controlled by the existing control model, which is consistent with the results obtained in the deceleration scenario. In addition, it can be seen from the figure that as the queue size increases, the present invention has better control over the queue length of the longer queue, which means that the present invention has higher control efficiency over the longer queue.

Claims

1. A long queue control method under a multi-car structure based on cellular automata, characterized in that: The following steps are involved: S1. Establish a long queue control model, including a multi-car optimization model, a queue head car control model, a sub-queue head car control model and a sub-queue following car control model; S2, collecting driving data of long queue vehicles and their surrounding vehicles through roadside facilities, and inputting the driving data into the multi-vehicle optimization model to output the optimal sub-queue head vehicle sequence; S3. The vehicles in the long queue are based on the optimal sub-queue head vehicle sequence information released in real time by the roadside facilities, combined with the status of multiple front vehicles in the queue and the mutual influence between vehicles in the queue. The long queue control model outputs the driving status of the long queue vehicles, thereby controlling the driving of the long queue.

2. The long queue control method under a multi-car structure based on cellular automata according to claim 1 is characterized in that: The multi-car optimization model is: In the formula, Z virtual is the objective function; is the stability evaluation index of the head vehicle of the i-th sub-queue; is the passenger comfort evaluation index of the head car of the i-th sub-queue; γ i Whether the vehicle in the queue is the head vehicle of the sub-queue; v max is the maximum speed of the platoon vehicles; a min is the minimum deceleration of the platoon vehicles; a max is the maximum acceleration of the vehicles in the queue; n is the total number of head vehicles in the sub-queue; N is the total number of vehicles in the queue; is the acceleration of the head vehicle of the i-th sub-queue at time t; is the acceleration change of the head vehicle of the i-th sub-queue at time t and time t-1; is the acceleration of the head vehicle of the i-th sub-queue at time t-1; is the speed difference between the head car and the head car of the i-th sub-queue of the long queue at time t; is the speed of the first vehicle in the queue at time t; is the speed of the first vehicle in the i-th sub-queue in the long queue at time t.

3. The long queue control method under a multi-car structure based on cellular automata according to claim 1 is characterized in that: The establishment of the queue head vehicle control model includes the following steps: Based on the fact that the head car in the queue is affected by the safety constraints of the car in front of it, as well as the speed and acceleration of multiple cars in front of the queue, an acceleration model of the head car in the queue is established. In the acceleration model of the first vehicle in the queue, it is considered that the first vehicle in the queue needs to maintain the desired safe speed with the vehicle in front of it; In the acceleration model of the leading vehicle in a queue, different weights are assigned to the leading vehicles according to their different influences on the leading vehicle at different positions at the same time. Based on the maximum speed limit of the vehicle and the output of the acceleration model of the first vehicle in the queue, the speed evolution rule of the first vehicle in the queue is established; The position of the head vehicle in a long queue is updated based on the speed evolution update rule of the head vehicle in the long queue.

4. The long queue control method under a multi-car structure based on cellular automata according to claim 3 is characterized in that: The acceleration model of the queue head vehicle is: The safe speed in the queue head vehicle control model is: The weight of the impact of multiple leading vehicles on the leading vehicle is: In the formula, is the acceleration of the first vehicle in the queue at time t; is the required safe speed of the leading vehicle at time t; v l (t) is the speed of the leading vehicle at time t; M is the number of leading vehicles; k m is the influence weight of the mth front vehicle on the leading vehicle; a em (t) is the acceleration of the mth vehicle in front at time t; v em (t) is the speed of the mth vehicle in front at time t; k0, k1, k2 are the weight coefficients of each item; b l is the maximum deceleration of the head car in a long queue; τ is the reaction time of the head car; x e (t) is the position of the vehicle immediately preceding the head vehicle in the queue at time t; x l (t) is the position of the lead vehicle at time t; l e is the vehicle length of the vehicle in front; v e (t) is the speed of the vehicle immediately preceding the head vehicle in the queue at time t.

5. The long queue control method under a multi-car structure based on cellular automation according to claim 1 is characterized in that: The establishment of the sub-queue head vehicle control model includes the following steps: Based on the fact that the head vehicle of a sub-queue is affected by the safety constraints of its preceding vehicle, as well as by the preceding head vehicle of the sub-queue and the preceding vehicles in front of the queue, an acceleration model of the head vehicle of the sub-queue is established. In the acceleration model of the head vehicle of a sub-queue, it is considered that the head vehicle of a sub-queue needs to maintain the desired safe speed with the vehicle in front of it; In the acceleration model of the head vehicle of a sub-queue, different weights are assigned to the head vehicles according to their different influences on the head vehicle of the sub-queue. Based on the maximum speed limit of the vehicle and the output of the acceleration model of the head vehicle in the sub-queue, the speed evolution rule of the head vehicle in the sub-queue is established; The position of the head vehicle in the sub-queue is updated based on the speed evolution update rule of the head vehicle in the sub-queue.

6. The long queue control method under a multi-car structure based on cellular automation according to claim 5 is characterized in that: The acceleration model of the head vehicle of the sub-queue is: The safe speed of the sub-queue head vehicle control model is: The weight of the impact of multiple leading vehicles on the leading vehicle of a sub-queue is: Where j is the head vehicle of the jth sub-queue in the long queue; is the required safe speed of the jth head vehicle in the sub-queue at time t; v j (t) is the speed of the first vehicle in the j-th sub-queue at time t; v j-1 (t) is the speed of the j-1th head vehicle in the sub-queue at time t; is the influence weight of the mth front vehicle on the jth head vehicle of the sub-queue; k3, k4, k5 are the weight coefficients of each item; b j is the maximum deceleration of the jth head vehicle in the sub-queue; x j+1 (t) is the position of the vehicle in front of the j-th sub-queue at time t; x j (t) is the position of the head vehicle of the j-th sub-queue at time t; l p is the length of the vehicle in front of the j-th sub-queue head vehicle; v j+1 (t) is the speed of the vehicle in front of the j-th sub-queue at time t.

7. The long queue control method under a multi-car structure based on cellular automata according to claim 1 is characterized in that: The establishment of the sub-queuing following vehicle control model includes the following steps: Based on the fact that the following vehicle in a sub-queue is affected by its preceding vehicle and the head vehicle of the sub-queue, an acceleration model of the following vehicle in the sub-queue is established. In the sub-queue following vehicle acceleration model, it is considered that the sub-queue following vehicle needs to maintain the desired safety speed with its leading vehicle; Based on the maximum speed limit of the vehicle and the output of the acceleration model of the sub-queued following vehicle, the speed evolution rule of the sub-queued following vehicle is established; The position of the following vehicle in the sub-queue is updated based on the speed evolution update rule of the following vehicle in the sub-queue.

8. The long queue control method under a multi-car structure based on cellular automata according to claim 7 is characterized in that: The acceleration model of the following vehicle in the sub-queue is: The safe speed of the sub-queue following vehicle control model is: In the formula, is the acceleration of the i-th following vehicle under the leadership of the j-th sub-platoon leader at time t; is the required safe speed of the i-th following vehicle under the leadership of the j-th sub-platoon leader at time t; is the speed of the i-th following vehicle under the j-th sub-queue head vehicle at time t; v j (t) is the speed of the first vehicle in the j-th sub-queue at time t; k0 and k6 are weight coefficients; b i is the maximum deceleration of the following vehicle; is the position of the front vehicle of the i-th following vehicle under the leadership of the j-th sub-queue head vehicle at time t; is the position of the i-th following vehicle under the j-th sub-queue head vehicle at time t; l i+1 The length of the vehicle in front of the following vehicle; It is the speed of the front vehicle of the i-th following vehicle under the leadership of the j-th sub-queue head vehicle at time t.

Images