Virtual signal lamp intersection scheduling system based on fuzzy controller

Abstract

The invention provides a virtual signal lamp intersection scheduling system based on a fuzzy controller, and belongs to the technical field of automatic driving specific scenes, the intersection scheduling system is realized based on an intelligent network connection collaborative automatic driving system, and vehicle scheduling is carried out on an intersection without a traffic light; according to the invention, through the V2X vehicle and road cloud cooperation system, a traffic flow control effect similar to a signal lamp effect is realized in a scene without signal lamps, such as a factory, so that the vehicle passing efficiency and safety of a signal lamp-free intersection are improved; the lighting time of the traffic lights can be adjusted in real time according to the traffic state information, and the intelligent degree and the response speed of the traffic control system are improved; by establishing the virtual signal lamp intersection scheduling system based on the fuzzy controller, the intersection management control cost and human input are reduced, and the traffic management efficiency and benefit are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of specific scenarios of autonomous driving, and in particular relates to a virtual traffic light intersection dispatching system based on a fuzzy controller. Background Art

[0002] With the development of economy, the number of cars is also increasing, and traffic accidents or traffic jams are coming along with it. Such incidents most often occur at intersections without traffic lights. Since there are no traffic lights to guide such intersections, the order of traffic depends entirely on the game between drivers. Some aggressive drivers constantly cut in to make themselves pass quickly, which poses a great safety threat to other vehicles and pedestrians, making intersections prone to collision accidents. Accidents often cause traffic jams, greatly reducing road traffic efficiency, and even causing secondary accidents, leading to further escalation of the incident. Summary of the invention

[0003] To solve the above problems, this paper proposes a virtual traffic light intersection dispatching system based on a fuzzy controller, which aims to achieve a traffic light-like control of the traffic time of each intersection through V2X (Vehicle to Everything, wireless communication technology for vehicles), thereby improving the traffic efficiency and safety of vehicles at intersections without traffic lights. The technical solutions provided by the present invention are as follows:

[0004] A virtual traffic light intersection dispatching system based on a fuzzy controller, the intersection dispatching system is implemented based on an intelligent network-connected collaborative automatic driving system, and performs vehicle dispatching at intersections without traffic lights; specifically, the intersection dispatching system includes a vehicle side, a road side, and a cloud side;

[0005] The vehicle side includes intelligent networked autonomous driving vehicles and manual vehicles; the intelligent networked autonomous driving vehicles realize autonomous driving through the intelligent networked collaborative autonomous driving system, and both the intelligent networked autonomous driving vehicles and manual vehicles are equipped with vehicle wireless communication technology V2X on-board unit OBU communication equipment to communicate with the road side and the cloud; the manual vehicle is also equipped with display alarm equipment;

[0006] The roadside device is a roadside device installed on the roadside, including roadside sensors, edge computing units MEC, and roadside units RSU; roadside sensors are used to detect vehicles and road conditions in the area; MEC performs structured processing on the data monitored by the sensors in real time, and RSU communicates with the cloud and vehicle side;

[0007] The cloud server includes a cloud server, which analyzes and calculates the vehicle and road condition information obtained in real time from the vehicle and road ends, judges the vehicle and road condition information through a fuzzy controller, outputs the travel time in each direction of the intersection, and sends it to the roadside equipment and then to the vehicle end via the RSU;

[0008] Specifically, the implementation process of the vehicle receiving the travel time of each direction at the intersection is as follows:

[0009] For autonomous driving vehicles, based on the intelligent network-connected collaborative autonomous driving system, the vehicle reports its own positioning and other information to the cloud, and the cloud directly sends the travel time in each direction of the intersection and the vehicle trajectory to the vehicle through the fuzzy controller;

[0010] For manual vehicles, the V2X OBU device on the vehicle side communicates with the RSU on the road side. The sensors on the road side obtain vehicle information in real time and perform structured processing. The RSU on the road side uploads the real-time roadside information and the information uploaded to the road side by the manual vehicle OBU to the cloud. The cloud sends the travel time in each direction of the intersection to the road side through the fuzzy controller. The RSU on the road side forwards the travel time in each direction of the intersection sent by the cloud to the vehicle side. After receiving the information, the OBU on the vehicle side prompts the driver of the traffic conditions of the intersection ahead by displaying the alarm device.

[0011] Furthermore, the specific construction process of the fuzzy controller is as follows:

[0012] Step 1: Traverse the traffic phases according to the road conditions at the intersection to obtain the traffic phase sequence;

[0013] Step 2: Select the deviation e between the actual traffic flow and the target traffic flow as the observed quantity, and select the traffic light duration u as the controlled quantity; for one of the traffic phases, the expected traffic flow within the duration u when the traffic light is green is C, but the actual measured traffic flow of this phase in the cycle is X, and then the deviation e = ΔX = XC is obtained;

[0014] Step 3: Fuzzify the observed and controlled variables using the triangular membership function;

[0015] Step 4: Divide the deviation e into at least 5 fuzzy sets, including negative large, negative small, zero, positive small, and positive large; a negative e indicates that the current traffic flow is lower than the target traffic flow, and a positive e indicates that the current traffic flow is higher than the target traffic flow. Set the value range of the deviation e to [-E, E], where E is a positive integer;

[0016] The control quantity u is divided into at least 5 fuzzy sets, including negative large, negative small, zero, positive small, and positive large; u is negative, which means increasing the duration of the phase signal light, and u is positive, which means decreasing the duration of the phase signal light. The value range of u is set to [-D, D];

[0017] Step 5: According to the traffic flow control rules, formulate fuzzy rules and find the fuzzy relationship R;

[0018] Step 6: The control quantity is output by the fuzzy controller and defuzzified according to the "maximum membership principle" to obtain the travel time under this phase. The cloud sends the travel time output by the fuzzy controller to the vehicle and road ends. The design process of the fuzzification rules for the remaining phases is the same, and the rules between each phase are mutually exclusive. Only one phase is passable at the same time point.

[0019] Furthermore, the intelligent connected autonomous driving vehicle is equipped with intelligent sensors, intelligent driving controllers, positioning equipment, and V2X OBU communication equipment; wherein the intelligent sensors are responsible for sensing the driving status and driving environment of the vehicle; the positioning equipment includes the global navigation satellite system GNSS and the inertial navigation system INS, the GNSS uses real-time dynamic technology to provide high-precision positioning for the autonomous driving vehicle, the INS monitors the vehicle's operating status and calculates the path, and provides redundancy for vehicle positioning, the intelligent driving controller receives vehicle control instructions from the cloud, makes decisions based on the data from the intelligent sensors, and sends the vehicle control instructions to the local area network of the vehicle controller; the V2X OBU communication equipment is responsible for connecting the cloud and the intelligent driving controller, the communication equipment receives vehicle control instructions from the cloud server, transmits the instructions to the intelligent driving controller, and uploads the data from the vehicle-side intelligent sensors to the cloud server.

[0020] Furthermore, the roadside sensor includes a low-latency camera and a solid-state lidar, which is used to detect all targets in the area and output the target's location, speed and other vehicle and road condition information.

[0021] Furthermore, the cloud server communicates with the vehicle side and the road side through the TCP / IP protocol.

[0022] Furthermore, the road conditions at the intersection include a one-way two-lane cross intersection, and the lane vehicle selection phase for the intersection includes a four-phase sequence: east-west execution, east-west left turn, north-south execution, and north-south left turn.

[0023] Furthermore, the formula for fuzzification using the triangle membership function is as follows:

[0024]

[0025] Among them, x is the independent variable, which is the change level of the deviation e, u(x) is the dependent variable, which is the membership value corresponding to the change level, a, b, c are the left endpoint, vertex and right endpoint of the isosceles triangle, and the lengths of ab and bc in each fuzzy set are set to 2 unit lengths.

[0026] Furthermore, the fuzzy rules are formulated as follows:

[0027] If e is negative and large, then u is negative and large;

[0028] If e is negatively small, then u is negatively small;

[0029] If e is zero, then u is zero;

[0030] If e is positively small, then u is positively small;

[0031] If e is positive, then u is positive;

[0032] When u is negative, the phase green light time is reduced, and when u is positive, the phase green light time is extended.

[0033] The beneficial effects of this method are as follows:

[0034] 1) Through the V2X vehicle-road-cloud collaborative system, a traffic light-like effect is achieved to control traffic flow in scenes without traffic lights such as factories, thereby improving the efficiency and safety of vehicle traffic at intersections without traffic lights.

[0035] 2) By introducing a fuzzy controller to establish an intersection dispatching system, the lighting time of traffic lights can be adjusted in real time according to traffic status information, thereby improving the intelligence and response speed of the traffic control system.

[0036] 3) By establishing a virtual traffic light intersection dispatching system based on fuzzy controller, the cost and manpower input of intersection management and control are reduced, and the efficiency and benefits of traffic management are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is the architecture diagram of the intelligent networked collaborative autonomous driving system;

[0038] Figure 2 This is a schematic diagram of roadside equipment;

[0039] Figure 3 This is a diagram of the virtual traffic light architecture;

[0040] Figure 4 It is the phase diagram of the classic intersection;

[0041] Figure 5 This is the structure diagram of the fuzzy controller based on the prediction results. DETAILED DESCRIPTION

[0042] In order to better understand the purpose, structure and function of the present invention, the following is a further detailed description of a virtual traffic light intersection scheduling method and system based on a fuzzy controller of the present invention in conjunction with the accompanying drawings.

[0043] This embodiment provides a virtual traffic light intersection dispatching system based on a fuzzy controller. The system is based on an intelligent networked collaborative automatic driving system. Its architecture is as follows: Figure 1 As shown in the figure, the system is divided into three parts: vehicle side, field side and cloud side.

[0044] The vehicle side includes intelligent networked autonomous driving vehicles and manual vehicles; specifically, the intelligent networked autonomous driving vehicles are equipped with intelligent sensors, intelligent driving controllers (including industrial computers and cloud controllers), positioning equipment (GNSS and INS), and V2X OBU (On Board Uni) communication equipment, etc.; the intelligent sensor can be a laser radar, which is responsible for sensing the driving status and driving environment of the vehicle; GNSS (Global Navigation Satellite System) uses RTK technology (Real-time Kinematic) to provide centimeter-level and meter-level high-precision positioning for autonomous driving vehicles; INS (Inertial Navigation System) can monitor the vehicle's operating status and calculate the path, providing redundancy for vehicle positioning. The intelligent driving controller includes an industrial computer and a cloud controller. The cloud controller is responsible for receiving vehicle control commands from the cloud. The industrial computer makes decisions based on the data of the intelligent sensor and controls the cloud controller to send the vehicle control command to the vehicle CAN (Controller Area Network) network. The V2X OBU communication device is responsible for connecting the cloud and the intelligent driving controller. The communication device receives vehicle control commands from the cloud server and transmits the commands to the cloud controller. At the same time, it uploads data from the vehicle-side intelligent sensors to the cloud server. The intelligent driving controller includes the vehicle controller VCU, the electronic power steering system EPS, the electronic stability system ESC, etc. Manual vehicles are equipped with V2X OBU devices, display alarm devices, etc. The V2X OBU device is used to communicate with the roadside RSU (Road Side Unit) to indirectly upload its own information to the system cloud platform or receive warning and travel time information. The display alarm device is used to provide drivers with intersection travel time and warning information issued by the intersection dispatch system. The field end is a roadside device installed on the roadside, such as Figure 2 As shown in the figure, the roadside sensors on the roadside equipment include two types of sensors, low-latency cameras and solid-state laser radars, which are used to detect all targets in the area and output information such as the location and speed of the targets to provide redundancy for vehicle safety. At the same time, they are equipped with MEC (Multi-access Edge Computing), which can perform structured processing on the data from low-latency cameras and solid-state laser radars. Finally, they are also equipped with roadside units RSU, which can communicate with the cloud and the vehicle.

[0045] The cloud mainly refers to the cloud server, which analyzes and calculates the vehicle and road condition information uploaded in real time by the vehicle and road ends through the TCP / IP protocol, and sends the vehicle control commands to the vehicle end.

[0046] At present, traffic signal dispatching at intersections mainly changes the phase and color of traffic lights to reduce traffic congestion, avoid traffic accidents, and optimize traffic efficiency. However, in scenarios such as factories, the construction of physical traffic lights often has some problems. In order to solve this technical problem, such as Figure 3 As shown, this embodiment proposes a virtual traffic light solution based on a cloud platform and V2X devices, which is intended to solve the problem of vehicle traffic at intersections without traffic lights or at intersections where it is inconvenient to install traffic lights, such as factories. The intersection is equipped with a V2X roadside device RSU, which can communicate with the cloud or vehicles with V2X on-board devices OBU, and the roadside is also equipped with roadside sensors, which can monitor the traffic flow in all directions of the intersection in real time (traffic flow is defined as the number of vehicles passing within a specified time period) and upload it to the cloud after being processed by MEC. It is required that all vehicles passing through the intersection are intelligent networked self-driving cars that can communicate with the cloud and the roadside and be controlled by them, or vehicles equipped with at least a V2X OBU device that can communicate with the roadside and a corresponding display that can provide traffic suggestions for drivers.

[0047] The cloud server in the cloud uses a fuzzy controller to judge the traffic flow and traffic priority in each direction of the intersection, outputs the travel time in each direction of the intersection, and sends it to the roadside equipment, and then sends it to all vehicles via the RSU.

[0048] like Figure 3 As shown in the figure, at an intersection without traffic lights, vehicles passing through pass according to the instructions of the intersection dispatching system; specifically, the vehicles passing through can be intelligent connected autonomous driving vehicles equipped with V2X OBU communication equipment, or manually driven vehicles equipped with V2X OBU equipment and display alarm equipment; the roadside equipment located on the road side is equipped with sensors, MEC and RSU, and MEC performs a certain degree of fusion processing on the information sensed by the sensors, and communicates with the cloud and the vehicle side through RSU. The cloud server inside the cloud receives data from the vehicle side and the road side, processes and makes decisions, and uses a fuzzy controller to determine the travel time in each direction according to the traffic volume; specifically, for autonomous driving vehicles, the vehicle side reports its own positioning and other information to the cloud, and the cloud directly sends the travel time in each direction of the intersection and the vehicle trajectory to the vehicle side. For manual vehicles, the V2X OBU device on the vehicle side communicates with the RSU on the road side. The sensors on the road side acquire vehicle information and perform structured processing. The RSU on the road side uploads the acquired real-time roadside information and the information uploaded by the manual vehicle OBU to the cloud. The cloud sends the travel time in each direction at the intersection to the road side. The RSU on the road side forwards the travel time in each direction at the intersection sent by the cloud to the vehicle side. After receiving the information, the OBU on the vehicle side prompts the driver of the traffic conditions at the intersection ahead by displaying an alarm device.

[0049] The construction process of the fuzzy controller is as follows:

[0050] Fuzzy control is an intelligent control method based on fuzzy set theory, fuzzy language variables and fuzzy logic reasoning. It is an intelligent control method that imitates the fuzzy reasoning and decision-making process of human behavior. This method first compiles the operator or expert experience into fuzzy rules, then fuzzifies the real-time signal from the sensor, uses the fuzzified signal as the input of the fuzzy rule, and adds the output to the actuator after completing the fuzzy reasoning. Its characteristics are: simple structure and strong robustness; no need to know the precise mathematical model of the controlled object; fuzzy control is highly consistent with the characteristics of human brain activity, and fuzziness and experience are both based on long-term experience accumulation.

[0051] This embodiment takes the classic one-way two-lane crossroad as a reference, such as Figure 4 As shown, the lane vehicles choose the passage phase in the classic four-phase sequence: east-west execution, east-west left turn, north-south execution, and north-south left turn.

[0052] From the perspective of actual traffic control, when the queue time at the intersection is long, the signal cycle should be lengthened, but generally not more than 200 seconds, to prevent drivers from getting bored; when the queue length is short, the signal cycle should be shortened, but generally not less than 15 seconds, to avoid the phenomenon that vehicles are too late to pass the intersection or speed up to pass the intersection. Figure 5 As shown, the specific process is as follows:

[0053] First, we need to select the observed quantity and the controlled quantity. The observed quantity is the deviation e between the actual traffic flow and the expected traffic flow, and the controlled quantity is the duration u of the traffic light. Assume that in one phase of the traffic light control, the expected traffic flow in the time period u when the traffic light is green is C, but the actual measured traffic flow in this phase in the cycle is X, and then we get the deviation e=ΔX=XC.

[0054] After that, the observed quantity (output quantity) and the controlled quantity (input quantity) are fuzzified. Here, the classic triangular membership function is used (the shape of the triangular membership function is an isosceles triangle, hence the name. It is usually defined by three parameters, namely: the left endpoint (a), the vertex (b) and the right endpoint (c). These three parameters together determine the shape and range of the function) to fuzzify the traffic flow deviation value and the virtual signal light control travel time. The formula is as follows:

[0055]

[0056] Among them, x is the independent variable, which is the change level e of the deviation value, and u(x) is the dependent variable, which is the corresponding membership value under this change level. a, b, c are the left endpoint, vertex, and right endpoint of the isosceles triangle. In each fuzzy set in this paper, the lengths of ab and bc are set to 2 unit lengths.

[0057] The deviation e is divided into 5 fuzzy sets, namely Negative Big (NB e ), Negative Small (NS e ), Zero (O e ), Positive Small (PS e ), and Positive Big (PB e ). When e is negative, it means the current traffic flow is lower than the target traffic flow, and when e is positive, it means the current traffic flow is higher than the target traffic flow. The value range of e is set as [-3, 3] (the value range is determined according to experience), and the corresponding fuzzy table is as follows:

[0058]

[0059] Similarly, the control variable u is also divided into 5 fuzzy sets, namely Negative Big (NB u ), Negative Small (NS u ), Zero (O u ), Positive Small (PS u ), and Positive Big (PB u ). When u is negative, it means to increase the duration of the signal light for this phase, and when u is positive, it means to decrease the duration of the signal light for this phase. The value range of u is set as [-D, D] (both the input and output of the fuzzy controller are selected according to experience and can be modified at any time), and the corresponding fuzzy table is as follows:

[0060]

[0061] Then, according to the traffic flow control rules, fuzzy rules are formulated as follows:

[0062] If e is Negative Big, then u is Negative Big;

[0063] If e is Negative Small, then u is Negative Small;

[0064] If e is Zero, then u is Zero;

[0065] If e is Positive Small, then u is Positive Small;

[0066] If e is Positive Big, then u is Positive Big;

[0067] When u is negative, the green light time of the phase should be reduced, and when u is positive, the green light time of the phase should be extended. The rules are as shown in the following table:

[0068]

[0069] Then, the fuzzy relation R can be obtained. It can be expressed as the Cartesian product of the input and output variables. The calculation method is similar to that of a matrix, but in the fuzzy relation calculation, multiplication takes the AND operation, and addition takes the OR operation. According to the aforementioned fuzzy rules, we have:

[0070] R = (NB e × NBu )∪(NS e ×NS u )∪(0 e ×0 u )∪(PS e ×PS u )∪(PB e ×PB u )

[0071] Easy to get:

[0072]

[0073] Finally, the output can be controlled by the fuzzy controller. The phase control quantity transit time u=e°R. For example, when e=NB,

[0074] u=e°R=[1.0 0.5 0.5 0.5 0.0 0.0 0.0 0.0 0.0]

[0075] The controller output is a simulation vector, then:

[0076]

[0077] According to the "maximum membership principle", defuzzification is performed, and the control quantity is selected as u = -D, that is, the green light timing of this phase is shortened, and the time D is shortened. For intelligent connected autonomous driving vehicles that can communicate directly with the cloud, the cloud sends the command to the vehicle; for external manual vehicles that cannot communicate directly with the cloud but are equipped with V2X OBU that can communicate with the roadside V2X device RSU, the cloud sends the command to the roadside RSU, and then broadcasts it to the vehicle through the RSU. The design process of the fuzzification rules for the remaining phases is the same, and the rules between each phase are mutually exclusive. At the same time point, only one phase has a green light for passage.

[0078] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

Claims

1. A virtual traffic light intersection dispatching system based on fuzzy controller, characterized in that: The intersection dispatching system is implemented based on an intelligent network-connected collaborative automatic driving system to dispatch vehicles at intersections without traffic lights; specifically, the intersection dispatching system includes a vehicle side, a road side, and a cloud side; The vehicle side includes intelligent networked autonomous driving vehicles and manual vehicles; the intelligent networked autonomous driving vehicles realize autonomous driving through the intelligent networked collaborative autonomous driving system, and both the intelligent networked autonomous driving vehicles and manual vehicles are equipped with vehicle wireless communication technology V2X on-board unit OBU communication equipment to communicate with the road side and the cloud; the manual vehicle is also equipped with display alarm equipment; The roadside device is a roadside device installed on the roadside, including roadside sensors, edge computing units MEC, and roadside units RSU; roadside sensors are used to detect vehicles and road conditions in the area; MEC performs structured processing on the data monitored by the sensors in real time, and RSU communicates with the cloud and vehicle side; The cloud server includes a cloud server, which analyzes and calculates the vehicle and road condition information obtained in real time from the vehicle and road ends, judges the vehicle and road condition information through a fuzzy controller, outputs the travel time in each direction of the intersection, and sends it to the roadside equipment and then to the vehicle end via the RSU; Specifically, the implementation process of the vehicle receiving the travel time of each direction at the intersection is as follows: For autonomous driving vehicles, based on the intelligent network-connected collaborative autonomous driving system, the vehicle reports its own positioning and other information to the cloud, and the cloud directly sends the travel time in each direction of the intersection and the vehicle trajectory to the vehicle through the fuzzy controller; For manual vehicles, the V2X OBU device on the vehicle side communicates with the RSU on the road side. The sensors on the road side obtain vehicle information in real time and perform structured processing. The RSU at the road end uploads the acquired real-time roadside information and the information uploaded by the manual vehicle OBU to the cloud. The cloud sends the travel time in each direction of the intersection to the road end through the fuzzy controller. The RSU at the road end forwards the travel time in each direction of the intersection sent by the cloud to the vehicle end. After receiving the information, the OBU at the vehicle end prompts the driver of the traffic conditions at the intersection ahead by displaying an alarm device.

2. A virtual traffic light intersection dispatching system based on fuzzy controller according to claim 1, characterized in that: The specific construction process of the fuzzy controller is as follows: Step 1: Traverse the traffic phases according to the road conditions at the intersection to obtain the traffic phase sequence; Step 2: Select the deviation e between the actual traffic flow and the target traffic flow as the observed quantity, and select the traffic light duration u as the controlled quantity; for one of the traffic phases, the expected traffic flow within the duration u when the traffic light is green is C, but the actual measured traffic flow of this phase in the cycle is X, and then the deviation e = ΔX = XC is obtained; Step 3: Fuzzify the observed and controlled variables using the triangular membership function; Step 4: Divide the deviation e into at least 5 fuzzy sets, including negative large, negative small, zero, positive small, and positive large; a negative e indicates that the current traffic flow is lower than the target traffic flow, and a positive e indicates that the current traffic flow is higher than the target traffic flow. Set the value range of the deviation e to [-E, E], where E is a positive integer; Divide the control quantity u into at least 5 fuzzy sets, including negative large, negative small, zero, positive small, and positive large; u is negative, which means increasing the duration of the phase signal light; u is positive, which means decreasing the duration of the phase signal light. The value range of u is set to [-D, D]; Step 5: According to the traffic flow control rules, formulate fuzzy rules and find the fuzzy relationship R; Step 6: The control quantity output by the fuzzy controller is defuzzified according to the "maximum membership principle" to obtain the travel time under this phase. The cloud sends the travel time output by the fuzzy controller to the vehicle and road ends; the design process of the fuzzification rules for the remaining phases is the same, and the rules between each phase are mutually exclusive. Only one phase is passable at the same time point.

3. A virtual traffic light intersection dispatching system based on fuzzy controller according to claim 2, characterized in that: The intelligent networked autonomous driving vehicle is equipped with intelligent sensors, intelligent driving controllers, positioning equipment, and V2X OBU communication equipment; wherein the intelligent sensors are responsible for sensing the driving status and driving environment of the vehicle; the positioning equipment includes the global navigation satellite system GNSS and the inertial navigation system INS. The GNSS uses real-time dynamic technology to provide high-precision positioning for the autonomous driving vehicle. The INS monitors the vehicle's operating status and calculates the path, providing redundancy for vehicle positioning. The intelligent driving controller receives vehicle control instructions from the cloud, makes decisions based on the data from the intelligent sensors, and sends the vehicle control instructions to the vehicle controller local area network. The V2XOBU communication device is responsible for connecting the cloud and the intelligent driving controller. The communication device receives vehicle control commands from the cloud server and transmits the commands to the intelligent driving controller. At the same time, it uploads data from the vehicle-side intelligent sensors to the cloud server.

4. A virtual traffic light intersection dispatching system based on fuzzy controller according to claim 3, characterized in that: The roadside sensors include low-latency cameras and solid-state laser radars, which are used to detect all targets in the area and output the target's location, speed, and other vehicle and road condition information.

5. A virtual traffic light intersection dispatching system based on fuzzy controller according to claim 4, characterized in that: The cloud server communicates with the vehicle side and the road side via TCP / IP protocol.

6. A virtual traffic light intersection dispatching system based on fuzzy controller according to claim 4, characterized in that: The road conditions at the intersection include a one-way two-lane cross intersection, and the lane vehicle selection phases at the intersection include a four-phase sequence: east-west execution, east-west left turn, north-south execution, and north-south left turn.

7. A virtual traffic light intersection dispatching system based on fuzzy controller according to claim 6, characterized in that: The formula for fuzzification using the triangle membership function is as follows: Among them, x is the independent variable, which is the change level of the deviation e, u(x) is the dependent variable, which is the membership value corresponding to the change level, a, b, c are the left endpoint, vertex and right endpoint of the isosceles triangle, and the lengths of ab and bc in each fuzzy set are set to 2 unit lengths.

8. A virtual traffic light intersection dispatching system based on fuzzy controller according to claim 7, characterized in that: The fuzzy rules are formulated as follows: If e is negative and large, then u is negative and large; If e is negatively small, then u is negatively small; If e is zero, then u is zero; If e is positively small, then u is positively small; If e is positive, then u is positive; When u is negative, the phase green light time is reduced, and when u is positive, the phase green light time is extended.

Images