Cloud-based hierarchical vehicle platoon predictive cruise control method and apparatus

Abstract

The application relates to a cloud-based layered vehicle queue predictive cruise control method and device. The method comprises the following steps: receiving a predictive cruise control request of a vehicle queue, obtaining real-time position and pilot vehicle state information of the vehicle queue based on the request, determining an optimal reference speed curve of the vehicle queue according to the real-time position, the pilot vehicle state information, a preset fuel consumption model and a preset dynamics model, and sending the optimal reference speed curve to a vehicle-mounted platform, so that the optimal driving speed obtained by the optimal reference curve, preset queue system constraints and a preset queue control error model is sent to the vehicle queue through the vehicle-mounted platform to perform cruise control on the vehicle queue. Therefore, the real-time position and state information of the vehicle are combined through the cloud platform to solve the problems of limited prediction range, insufficient information acquisition capability and queue predictive speed planning of the existing queue predictive cruise, and the driving speed of the vehicle queue is planned more forwardly and intelligently.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a cloud-based, hierarchical vehicle platooning predictive cruise control method and apparatus. Background Technology

[0002] The rapid development of road transportation has also brought a series of severe challenges, such as traffic safety and environmental pollution. Commercial vehicles are the mainstay of road transportation, and their energy consumption and safety issues are becoming increasingly serious. In highway scenarios, predictive cruise control of queuing is primarily achieved by considering road gradient, curvature, and traffic flow conditions. For predictive cruise control of queuing, the focus is on combining static road gradient information and the current queuing status to plan queuing speed.

[0003] Among related technologies, platooning predictive cruise control methods focus on predictive cruise control combined with road gradient. This method can achieve economical driving of vehicles, and the formation of vehicles can improve road traffic efficiency and throughput.

[0004] However, the following drawbacks exist in this queuing predictive cruise control method: (1) The range of prediction of vehicle-mounted queuing predictive cruise control is limited, and the ability to acquire and perceive information is limited, which cannot further expand the advantages of queuing predictive cruise control; (2) There is a lack of research and architecture research on queuing predictive cruise control combined with cloud, and the cloud cannot be taken into account in the existing queuing predictive cruise control; (3) The existing technology considers global queuing predictive cruise control algorithms, but rarely considers predictive cruise algorithms in the rolling distance domain, so it cannot offset the uncertainty of future queuing operation, which needs to be solved urgently. Summary of the Invention

[0005] This application provides a cloud-based hierarchical vehicle platoon predictive cruise control method and device to solve the problems of limited prediction range, insufficient information acquisition capability and platoon predictive speed planning in existing platoon predictive cruise, so as to realize more advanced and intelligent planning of vehicle platoon driving speed.

[0006] To achieve the above objectives, the first aspect of this application proposes a cloud-based, hierarchical vehicle platooning predictive cruise control method, comprising:

[0007] Receive the predictive cruise control request from the vehicle convoy;

[0008] Based on the predictive cruise control request, the real-time location of the vehicle convoy and the status information of the lead vehicle are obtained; and

[0009] The optimal reference speed curve of the vehicle convoy is determined based on the real-time location, the status information of the lead vehicle, the preset fuel consumption model, and the preset dynamics model. The optimal reference speed curve is then sent to the vehicle platform. The vehicle platform then sends the optimal driving speed obtained from the optimal reference curve, the preset convoy system constraints, and the preset convoy control error model to the vehicle convoy. The vehicle convoy is then cruise-controlled using the optimal driving speed.

[0010] According to one embodiment of this application, determining the optimal reference speed curve of the vehicle convoy based on the real-time location, the navigator vehicle status information, a preset fuel consumption model, and a preset dynamics model includes:

[0011] The road gradient information ahead of the vehicle convoy is determined based on the real-time location.

[0012] Based on the preset dynamic model, the longitudinal force on each vehicle in the vehicle convoy is calculated according to the road slope information ahead of the vehicle convoy and the status information of the lead vehicle.

[0013] The lead vehicle of the vehicle platoon is divided into states within the planning period. Based on the division results and the preset speed planning cost function, the optimal reference speed curve of the vehicle platoon is determined according to the preset fuel consumption model and the longitudinal force on each vehicle in the vehicle platoon.

[0014] According to one embodiment of this application, the preset fuel consumption model is:

[0015]

[0016] Where, ξ i,j The parameters T are fitted to the preset fuel consumption model. tq n is the vehicle engine torque, and n is the vehicle engine speed.

[0017] According to one embodiment of this application, the preset dynamic model is:

[0018]

[0019] Where, m i For the quality of the vehicle, For the vehicle's acceleration, F e,i For engine traction, F g,i For slope resistance, F r,i For rolling resistance, F air,i This refers to air resistance during vehicle operation.

[0020] According to one embodiment of this application, the preset queue system constraints

[0021]

[0022] Among them, e s (t) represents the vehicle spacing error in the queue, e v(i,0 (t) represents the initial speed error of the vehicles in the queue. For the speed of the vehicle, For the impact force of the vehicle, u i (t) represents the control quantity of the vehicle.

[0023] According to one embodiment of this application, the preset speed planning cost function is:

[0024]

[0025] in, Let J(j,k+1) be the cost function from the current state to the next state, and let J(j,k+1) be the cost from the next state to the final state.

[0026] According to one embodiment of this application, the vehicle queue communicates with each other via a first communication topology, and the vehicle queue communicates with the cloud via a second communication topology, wherein...

[0027] The first communication topology is:

[0028] Q i ={j|α i,j =1,j∈N};

[0029] The second communication topology is:

[0030]

[0031] Where, α i,j The elements in the adjacent matrix are N, which is a natural number, and E represents the relationship between the connected cloud nodes.

[0032] The cloud-based hierarchical vehicle platoon predictive cruise control method proposed in this application receives predictive cruise control requests from vehicle platoons and obtains the real-time location and lead vehicle status information of the platoon based on these requests. It then determines the optimal reference speed curve for the platoon based on the real-time location, lead vehicle status information, a preset fuel consumption model, and a preset dynamics model, and sends this curve to the onboard platform. The onboard platform then sends the optimal driving speed, obtained from the optimal reference curve, preset platoon system constraints, and a preset platoon control error model, to the vehicle platoon for cruise control. Therefore, by combining the real-time location and status information of vehicles with a cloud platform, this method can solve the problems of limited prediction range, insufficient information acquisition capabilities, and platoon predictive speed planning in existing platoon predictive cruise systems, enabling more advanced and intelligent planning of vehicle platoon speeds.

[0033] To achieve the above objectives, a second aspect of this application provides a cloud-based, hierarchical vehicle platooning predictive cruise control device, comprising:

[0034] A receiving module is used to receive predictive cruise control requests from the vehicle convoy;

[0035] The acquisition module is used to acquire the real-time location of the vehicle convoy and the status information of the lead vehicle based on the predictive cruise control request; and

[0036] The control module is used to determine the optimal reference speed curve of the vehicle convoy based on the real-time location, the status information of the lead vehicle, a preset fuel consumption model, and a preset dynamics model, and to send the optimal reference speed curve to the vehicle platform. The vehicle platform then sends the optimal driving speed obtained from the optimal reference speed, the preset convoy system constraints, and the preset convoy control error model to the vehicle convoy, and performs cruise control on the vehicle convoy using the optimal driving speed.

[0037] According to one embodiment of this application, the control module is specifically used for:

[0038] The road gradient information ahead of the vehicle convoy is determined based on the real-time location.

[0039] Based on the preset dynamic model, the longitudinal force on each vehicle in the vehicle convoy is calculated according to the road slope information ahead of the vehicle convoy and the status information of the lead vehicle.

[0040] The lead vehicle of the vehicle platoon is divided into states within the planning period. Based on the division results and the preset speed planning cost function, the optimal reference speed curve of the vehicle platoon is determined according to the preset fuel consumption model and the longitudinal force on each vehicle in the vehicle platoon.

[0041] According to one embodiment of this application, the preset fuel consumption model is:

[0042]

[0043] Where, ξ i,j The parameters T are fitted to the preset fuel consumption model. tq n is the vehicle engine torque, and n is the vehicle engine speed.

[0044] According to one embodiment of this application, the preset dynamic model is:

[0045]

[0046] Where, m i For the quality of the vehicle, For the vehicle's acceleration, F e,i For engine traction, F g,i For slope resistance, F r,i For rolling resistance, F air,i This refers to air resistance during vehicle operation.

[0047] According to one embodiment of this application, the preset queue system constraints

[0048]

[0049] Among them, e s (t) represents the vehicle spacing error in the queue, e v(i,0) (t) represents the initial speed error of the vehicles in the queue. For the speed of the vehicle, For the impact force of the vehicle, u i (t) represents the control quantity of the vehicle.

[0050] According to one embodiment of this application, the preset speed planning cost function is:

[0051]

[0052] in, Let J(j,k+1) be the cost function from the current state to the next state, and let J(j,k+1) be the cost from the next state to the final state.

[0053] According to one embodiment of this application, the vehicle queue communicates with each other via a first communication topology, and the vehicle queue communicates with the cloud via a second communication topology, wherein...

[0054] The first communication topology is:

[0055] Q i ={j|αi,j =1,j∈N};

[0056] The second communication topology is:

[0057]

[0058] Where, α i,j Let N be the element in the adjacent matrix, N be a natural number, and E be the connected cloud nodes.

[0059] The cloud-based hierarchical vehicle platoon predictive cruise control device proposed in this application receives predictive cruise control requests from vehicle platoons and obtains the real-time location of the platoon and the status information of the lead vehicle based on the requests. It then determines the optimal reference speed curve for the vehicle platoon based on the real-time location, the lead vehicle status information, a preset fuel consumption model, and a preset dynamics model, and sends this curve to the onboard platform. The onboard platform then sends the optimal driving speed, obtained from the optimal reference curve, preset platoon system constraints, and a preset platoon control error model, to the vehicle platoon for cruise control. Therefore, by combining the real-time location and status information of the vehicles with a cloud platform, the device can solve the problems of limited prediction range, insufficient information acquisition capabilities, and predictive speed planning in existing platoon predictive cruise systems, achieving more advanced and intelligent planning of vehicle platoon speeds.

[0060] To achieve the above objectives, a third aspect of this application provides a server comprising: one or more processors; a storage device for storing one or more programs; and, when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the cloud-based hierarchical vehicle platooning predictive cruise control method as described in the above embodiments.

[0061] To achieve the above objectives, a fourth aspect of this application provides a computer storage medium storing a computer program that is executed by a processor to implement the cloud-based hierarchical vehicle platoon predictive cruise control method as described in the above embodiments.

[0062] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0063] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0064] Figure 1 A flowchart of a cloud-based hierarchical vehicle platoon predictive cruise control method according to an embodiment of this application;

[0065] Figure 2 This is a schematic diagram illustrating the complete system composition and working principle according to an embodiment of this application;

[0066] Figure 3 This is a schematic diagram of a vehicle-cloud layered architecture for a cloud-based layered vehicle platooning predictive cruise control system according to an embodiment of this application.

[0067] Figure 4 This is a schematic diagram of a submodule fuel consumption model of a cloud-based hierarchical vehicle platooning predictive cruise control system according to an embodiment of this application.

[0068] Figure 5 This is a schematic diagram of the dynamics model of a submodule vehicle of a cloud-based hierarchical vehicle platooning predictive cruise control system according to an embodiment of this application.

[0069] Figure 6 This is a schematic diagram of the state partitioning of a state partitioning module in a cloud-based speed planning algorithm, a submodule of a cloud-based hierarchical vehicle platooning predictive cruise control system according to an embodiment of this application.

[0070] Figure 7 This is a schematic diagram of a submodule queue control error model of a cloud-based hierarchical vehicle queue predictive cruise control system according to an embodiment of this application.

[0071] Figure 8 This is a control flowchart of a cloud-based hierarchical vehicle platooning predictive cruise control system according to an embodiment of this application;

[0072] Figure 9 This is a block diagram of a cloud-based hierarchical vehicle platoon predictive cruise control device according to an embodiment of this application;

[0073] Figure 10 This is a schematic diagram of the server structure according to an embodiment of this application. Detailed Implementation

[0074] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0075] The predictive cruise control method and apparatus for a cloud-based hierarchical vehicle platoon, based on embodiments of this application, will now be described with reference to the accompanying drawings. First, the predictive cruise control method for a cloud-based hierarchical vehicle platoon, based on embodiments of this application, will be described with reference to the accompanying drawings.

[0076] Figure 1 This is a flowchart of a cloud-based hierarchical vehicle platoon predictive cruise control method according to an embodiment of this application.

[0077] Before introducing the cloud-based hierarchical vehicle platoon predictive cruise control method proposed in the embodiments of this application, let's briefly introduce the cloud-based hierarchical vehicle platoon predictive cruise control system involved in this method and its working principle, such as... Figure 2 As shown, the system includes: cloud control platform 1, vehicle platform 2, and vehicle queue 3.

[0078] The cloud control platform 1 includes: a fuel consumption model 11, a positioning module 12, a road slope information module 13, a vehicle dynamics model 14, and a cloud-based speed planning algorithm module 15. The vehicle fuel consumption model 11 is the fuel consumption model for vehicles in the vehicle platoon; the positioning module 12 is used to parse the location information uploaded by the vehicle and locate the real-time position of the vehicle on the map in the cloud support platform, thereby obtaining the road slope information from the road slope information module 13; the vehicle dynamics model 14, after obtaining the vehicle's status information uploaded to the cloud, calculates the longitudinal force on the highway segment where the vehicle is currently located, and injects this information into the cloud-based speed planning algorithm module 15 to solve for the optimal speed for the vehicle platoon.

[0079] In addition, the cloud-based speed planning algorithm module 15 also includes: a state partitioning module 151, a speed planning cost function module 152, and a speed solution module 153. The state partitioning module 151 is used to partition the space for speed solution within the planning period, with the speed interval serving as the upper and lower bounds of the planning and the planning stage as the planning steps. The speed planning cost function module 152 is used to set the control objective of the speed planning module to achieve the economy, smoothness, and efficiency of queue driving. The speed solution module 153 indexes the value of the cost function solved by the speed planning cost function module 152 and solves for an optimal reference speed curve.

[0080] The vehicle platform 2 includes a queue control error model 21 and a queue stability controller 22. The queue control error model 21 is based on the state information of the vehicle queue uploaded in the vehicle queue 3, and calculates the spacing error and speed error of the vehicle queue. The queue stability controller 22 is used to calculate the control quantity that satisfies the queue stability control after receiving the speed command from the cloud.

[0081] In addition, the queue stabilization controller 22 also includes: a system prediction model 221, a system constraint model 222, and a cost function module 223 for solving the system. The system prediction model 221 describes the predictive model of the system in actual control; the system constraint model 222 requires the actual dynamic constraints of the vehicle to ensure that the design of the queue stabilization controller 22 conforms to actual vehicle control; the cost function module 223 sets the control objective of the queue stabilization controller to satisfy the smoothness of vehicle operation and the accuracy of control.

[0082] Vehicle queue 3 is the actual vehicle queue, which has the ability to communicate with the cloud and communicate between vehicles (V2V) within the queue. The communication and information processing functions between the vehicle platform 2 and the cloud control platform 1 are realized by the T-BOX (Telematics-BOX, vehicle-mounted intelligent remote information processing terminal), which is equipped with GNSS (Global Navigation Satellite System) and RTK (Real Time Kinematic) positioning modules, which can obtain the real-time location of the vehicles and upload it to the cloud to complete the determination of the vehicle's location.

[0083] It is understood that the cloud-based hierarchical predictive cruise control system for vehicle platooning provided in this application adopts a hierarchical system architecture design. The upper layer, the cloud, is used to build the speed planning algorithm for the vehicle platooning, and the lower layer is used to build the vehicle platooning stability controller. Specifically, the speed planning algorithm for the vehicle platooning in the cloud is a dynamic programming algorithm in the rolling distance domain, and the vehicle platooning stability controller is a Distributed Model Predictive Control (DMPC). The cloud provides strategic guidance for vehicle platooning operation to achieve the goal of economical driving; the vehicle-side provides tactical guidance for vehicle platooning operation to achieve the goal of stable vehicle platooning control.

[0084] Furthermore, such as Figure 3 As shown, Figure 3 This is a schematic diagram of a vehicle-cloud layered architecture for a cloud-based layered vehicle platooning predictive cruise control system according to one embodiment of this application. Its specific working principle is as follows:

[0085] The vehicle convoy requests predictive cruise control services from the cloud. The edge cloud acquires the real-time location of the convoy and the status information of the lead vehicle. It also obtains static traffic information (road gradient) ahead of the convoy through a positioning module. Taking into account vehicle energy consumption, the optimal driving speed for the convoy is calculated using a cloud-based speed planning algorithm and sent to the vehicle-side controllers. Upon receiving the speed command from the cloud, the vehicle-side controllers, considering the current vehicle status, output the optimal control parameters to control the convoy. In the next planning cycle, the cloud again receives the service request from the vehicle convoy, the vehicles re-upload their status, and the predictive cruise control planning and control of the convoy are restarted, thus forming a rolling closed-loop control system.

[0086] It should be noted that predictive cruise control for vehicle platooning is a real-time application of the cloud control system, requiring deployment on an edge cloud application platform. Information exchange between the edge cloud and the vehicle platoon is achieved wirelessly via a vehicle-to-cloud gateway. Uploaded information includes the real-time location of the vehicle platoon and the status of the lead vehicle; downloaded information is the speed planned in the cloud. The cloud-based platform provides road gradient information, and the cloud application platform deploys vehicle fuel consumption models, vehicle dynamics models, and cloud-based speed planning algorithm modules. On the vehicle side, the platoon control error model and distributed model predictive controller are deployed. Predictive cruise control for the platoon is achieved through vehicle-cloud collaborative control.

[0087] Specifically, such as Figure 1 As shown, this cloud-based hierarchical vehicle platoon predictive cruise control method includes the following steps:

[0088] In step S101, a predictive cruise control request from the vehicle queue is received.

[0089] Understandably, a vehicle queue can request predictive cruise control services from the cloud, and the cloud can execute subsequent control operations after receiving the predictive cruise control request from the vehicle queue.

[0090] In step S102, based on the predictive cruise control request, the real-time location of the vehicle convoy and the status information of the lead vehicle are obtained.

[0091] Specifically, based on the predictive cruise control requests issued by the vehicle convoy, the edge cloud can obtain the real-time location of the vehicle convoy and the status information of the lead vehicle, including its position, speed, and acceleration.

[0092] In step S103, the optimal reference speed curve of the vehicle platoon is determined based on the real-time location, the status information of the navigator vehicle, the preset fuel consumption model, and the preset dynamics model. The optimal reference speed curve is then sent to the vehicle platform. The vehicle platform then sends the optimal driving speed obtained from the optimal reference curve, the preset platoon system constraints, and the preset platoon control error model to the vehicle platoon. The optimal driving speed is used to perform cruise control on the vehicle platoon.

[0093] It is understood that the embodiments of this application may be based on a preset fuel consumption model (i.e. Figure 2 The fuel consumption model 11 in the model can comprehensively consider the energy consumption of vehicles, and combine the real-time position of the vehicle convoy, the status information of the lead vehicle, and the preset dynamics model (i.e., Figure 2 The vehicle dynamics model 14 in the cloud, after being solved by the speed planning algorithm, can determine the optimal reference speed curve of the vehicle queue, and send the optimal reference speed curve to the vehicle platform (i.e., the vehicle-side controller) of the queue. After receiving the speed command (i.e., the optimal reference speed curve) from the cloud, the vehicle platform combines it with the preset queue system constraints (i.e., Figure 2 The queue stability controller 22) and the preset queue control error model (i.e. Figure 2 The queue control error model 21 can output the optimal driving speed of the vehicles (i.e., the optimal control quantity) to perform cruise control on the vehicle queue.

[0094] Furthermore, in some embodiments, determining the optimal reference speed curve of the vehicle convoy based on real-time location, navigator vehicle status information, a preset fuel consumption model, and a preset dynamics model includes: determining the road slope information ahead of the vehicle convoy based on the real-time location; calculating the longitudinal force on each vehicle in the convoy based on the road slope information ahead of the convoy and the navigator vehicle status information, according to the preset dynamics model; dividing the navigator vehicle of the convoy into states within the planning period, and determining the optimal reference speed curve of the vehicle convoy based on the division results and a preset speed planning cost function, according to the preset fuel consumption model and the longitudinal force on each vehicle in the convoy.

[0095] Specifically, the cloud-based system determines the static road information ahead of the vehicle convoy based on its real-time location using a map positioning module. The road gradient and lead vehicle status information from this static road information are then used as input to the cloud-based speed planning module. Combined with a pre-defined dynamics model, the module calculates the longitudinal force acting on each vehicle in the convoy. The state partitioning module partitions the lead vehicle's state within the planning cycle and, based on the partitioning results and a pre-defined speed planning cost function (i.e., ... Figure 2The speed planning cost function module 152 determines the optimal reference speed curve of the vehicle queue based on the preset fuel consumption model and the longitudinal force on each vehicle in the vehicle queue.

[0096] Understandably, the optimal driving speed planning algorithm also needs to comprehensively consider the preset dynamics model, preset fuel consumption model, and preset queuing system constraints. The cloud-based speed planning algorithm module calculates the optimal reference speed curve through the predictive cruise algorithm and then sends it to all vehicles in the vehicle queuing. After receiving the optimal reference speed curve sent from the cloud, the lead vehicle and subsequent following vehicles compare their current speed with the optimal reference speed curve sent from the cloud, using the speed error as input to the queuing stability controller. They receive the position of the vehicle in front and compare the current position of the vehicle with the desired vehicle spacing to obtain the spacing error of the vehicles at the current moment. The vehicle-side queuing stability controller uses the acceleration of the vehicle in front as the disturbance of the system and integrates the vehicle's own vehicle state information (including speed, position, and acceleration) as input to calculate the desired acceleration of the vehicles in the queuing, thereby achieving cloud-based speed tracking and maintaining queuing stability, realizing cloud-supported predictive cruise control.

[0097] Furthermore, in some embodiments, the preset fuel consumption model is:

[0098]

[0099] Where, ξ i,j T is the parameter fitted to the preset fuel consumption model. tq n is the vehicle engine torque, and n is the vehicle engine speed.

[0100] Furthermore, in some embodiments, the preset dynamic model is:

[0101]

[0102] Where, m i For the quality of the vehicle, For the vehicle's acceleration, F e,i For engine traction, F g,i For slope resistance, F r,i For rolling resistance, F air,i This refers to air resistance during vehicle operation.

[0103] Furthermore, in some embodiments, preset queue system constraints

[0104]

[0105] Among them, e s (t) represents the vehicle spacing error in the queue, e v(i,0)(t) represents the initial speed error of the vehicles in the queue. For the speed of the vehicle, For the impact force of the vehicle, u i (t) represents the control quantity of the vehicle.

[0106] Furthermore, in some embodiments, the preset speed planning cost function is:

[0107]

[0108] in, Let J(j,k+1) be the cost function from the current state to the next state, and let J(j,k+1) be the cost from the next state to the final state.

[0109] Furthermore, in some embodiments, the vehicles within the same queue communicate via a first communication topology, and the vehicle queue communicates with the cloud via a second communication topology, wherein...

[0110] The first communication topology is:

[0111] Q i ={j|α i,j =1,j∈N}; (5)

[0112] The second communication topology is:

[0113]

[0114] Where, α i,j Let N be the element in the adjacent matrix, N be a natural number, and E be the connected cloud nodes.

[0115] To facilitate those skilled in the art to further understand the cloud-based hierarchical vehicle platoon predictive cruise control method proposed in the embodiments of this application, the following detailed description of the specific implementation method is provided.

[0116] First, it is necessary to determine the vehicle-to-cloud communication and the intra-queue communication between the vehicle control platform and the vehicle platform. After combining with the cloud platform, the communication topology of the queue needs to take into account the actual needs and applications of vehicle queue and cloud communication. On the basis of the traditional vehicle queue, cloud nodes are added to change the traditional communication topology. The communication topology of cloud-supported vehicle queue predictive cruise control can be modeled by a directed graph C = {V, E}, where V = {0, 1, 2, ..., N} and E represents the connected nodes. In this embodiment, the cloud is defined as node 0.

[0117] Adjacency matrix A is used to describe the information flow between vehicles in a vehicle queue, where α i,jLet {j,i}∈E be an element in A, indicating that vehicle i (i≥0) can obtain the state information of vehicle j; otherwise, vehicle i cannot obtain the state information of vehicle j.

[0118]

[0119] Vehicle i can obtain the status of the remaining vehicles in the vehicle queue as shown in equation (5). Matrix P is used to represent the connection between cloud communication and vehicles in the vehicle queue. Compared with traditional vehicle queue communication, it emphasizes the role of the cloud as a communication node in the vehicle queue. The cloud and vehicle queue matrix is ​​defined as follows:

[0120] P = diag{p1, p2, ..., p N}; (8)

[0121] The communicability setting between the cloud and vehicle i in the vehicle queue can be characterized by equation (6), where if {1, i} ∈ E, it indicates that the vehicle communicates with the cloud; otherwise, there is no communication.

[0122] The communication topology of the entire cloud-supported vehicle queue can be represented by the following formula:

[0123] Θ i =Q i ∪P i (9)

[0124] Furthermore, the cloud control platform includes a cloud control infrastructure platform and a cloud control application platform. The cloud control infrastructure platform provides basic hardware and software platforms, standard components for perception fusion and decision-making, and can provide basic hardware, software and information support for the cloud control application platform.

[0125] The fuel consumption model is based on the general characteristic data of a certain commercial vehicle model. A high-fidelity polynomial engine fuel consumption model is derived by using linear interpolation and polynomial fitting. Figure 4 As shown in the figure. The polynomial fuel consumption model fits the fuel consumption model into a quadratic polynomial function of vehicle speed and engine torque, as shown in equation (1).

[0126] Wherein, the fitting function ξ i,j Subsequently, the polynomial fuel consumption model can be updated and modified based on the different engine fuel consumption model data of different vehicle models, and then updated and deployed on the cloud control application platform.

[0127] The cloud positioning module and road slope information module locate the vehicle's position after obtaining the position information of the lead vehicle in the vehicle convoy. The cloud data support platform provides the road slope information ahead of the vehicle's position, which is then sent to the cloud speed planning algorithm module in the cloud control application platform as static traffic information for the predictive cruise algorithm.

[0128] The vehicle's dynamics model can be used to comprehensively analyze the longitudinal forces acting on the vehicle, taking into account its engine characteristics, vehicle power, air resistance (i.e., wind resistance), rolling resistance, and gradient resistance. Due to the mass and power characteristics of commercial vehicles, the impact of road gradient on longitudinal dynamics also needs to be considered, thus incorporating speed optimization into the optimization process.

[0129] Vehicle dynamics model analysis, such as Figure 5 As shown, the longitudinal force analysis of the vehicle is performed, and the dynamic model of vehicle i is shown in equation (2).

[0130] Wherein, when vehicle i travels on a road with an inclination angle of θ, the gradient resistance of vehicle i is:

[0131] F g,i =m i gsin(θ); (10)

[0132] Where, m i Let g be the mass of the i-th vehicle, and g be the acceleration due to gravity.

[0133] Rolling resistance is shown in the following formula:

[0134] F r,i =c r m i gcos(θ); (11)

[0135] Among them, c r This is the rolling resistance coefficient.

[0136] The air resistance during vehicle i's operation is:

[0137]

[0138] Where, ρ a C is the density of air. d A is the air drag coefficient. F V is the frontal windward area of ​​a vehicle. wind For wind speed, V x This represents the longitudinal speed of the vehicle.

[0139] At this point, the driving force of vehicle i can be expressed as:

[0140]

[0141] The tire torque required to produce the desired acceleration is:

[0142]

[0143] Combining equations (1), (2), (5) to (14), the net torque T of the vehicle can be obtained. e The relationship with the desired acceleration is as follows:

[0144]

[0145] Among them, i e I is the rotational inertia of the engine. w R is the moment of inertia of the tire. p For the gear ratio, I t Let r be the rotational inertia of the turbine. eff This is the effective radius of the tire.

[0146] The above describes the process of establishing a vehicle dynamics model. After obtaining the state feedback of the vehicle convoy, the dynamics model can calculate the longitudinal forces acting on the vehicles in the cloud, and then consider the road gradient, i.e., the impact of gradient resistance on vehicle operation.

[0147] Furthermore, the cloud-based speed planning algorithm module is used to comprehensively solve the economic speed of vehicle operation based on the fuel consumption model, road slope information, and vehicle dynamics model. This algorithm module is a dynamic programming algorithm in the rolling distance domain, which can solve the problem of queuing economic driving in cloud-supported queuing predictive cruise control systems.

[0148] The state partitioning module partitions the state of the lead vehicle in the vehicle platoon within the planning cycle, such as... Figure 6 As shown, the acquired road slope information of X km is divided into sections, where the planning period is divided into N sections. p The process involves several stages, including velocity planning within the planning period. This includes defining the state points within the planning period using the dynamic programming algorithm, and optimizing the prediction region of the planning period. The prediction region is divided into N parts based on the same spacing. p The optimization problem is decomposed into several sub-problems for solution. Spatially, the velocity optimization problem is divided into stages for solution, with the interval of each planning point being ΔS. RDP At each planning point, the algorithm plans at its maximum speed. and minimum speed For the interval, the speed interval is set to... The dynamic programming process is divided into states based on the speed range, with the speed range serving as the upper and lower bounds of the planning process. The planning stages are used as the steps in the planning process to divide the state space. The entire planning path is also divided. The planning algorithm is designed as a rolling distance domain dynamic programming algorithm, which executes only the first step of the planning process at a time and then restarts the planning of the speed for the next stage. Figure 6The stages shown are stage1, stage2...stageN. This completes the state space partitioning of the navigator vehicle within the planning cycle using the dynamic programming algorithm.

[0149] The velocity planning cost function module sets the transition cost between two state points, as shown in formula (16) for position point p. (j,h) To location point p (j,h+1) The formula for the transition of vehicle speed state between state points is:

[0150]

[0151] Where ΔS is the interval between two state points. Indicates in p (j,h) The speed. Indicates in p (j,h+1) The speed at which the state transition function is determined.

[0152] Based on the state space defined by dynamic programming, the cost function of velocity programming is set as shown in equation (4). This ensures that the state transition from the current state to the final state is also optimal, thus guaranteeing the global optimality of the state transition.

[0153] The cost function from one state point to the next is β, where the cost function is shown in the following equation:

[0154]

[0155] Among them, W cost_fuel W is a penalty factor for the fuel consumption optimization term during operation, which ensures fuel economy during vehicle operation. cost_ref W is a penalty factor for the optimization term that accounts for the deviation between the planned speed and the reference speed, limiting the planned speed from deviating excessively from the set reference speed. cost_△v W is a penalty factor for velocity changes between states, used to avoid large velocity fluctuations. cost_△a This is a penalty factor for acceleration changes, used to avoid excessive acceleration fluctuations.

[0156] Based on the solution of state transition cost, the speed solution module indexes the value of the cost function of the last state, determines whether it is the minimum global cost, solves for a minimum speed sequence (i.e. the optimal reference speed curve), and sends the speed curve to the vehicle platform.

[0157] The onboard platform receives the optimal reference speed curve from the cloud control platform, parses the speed commands, and combines them with the queue control error model to achieve stable control of the vehicle queue. Based on the vehicle speeds sent by the cloud control platform, the speed error and spacing error of the vehicles in the queue are calculated. The positional relationship of the vehicles in the queue, i.e., the relationship between vehicle i-1 and vehicle i, is as follows:Figure 7 As shown, where and These represent the positions of vehicle i-1 and vehicle i respectively during stage j of the planning period. and These represent the speeds of vehicle i-1 and vehicle i respectively during stage j of the planning period. This represents the expected distance between vehicles at the current stage, while This indicates the actual positional relationship between the vehicles.

[0158] The desired queue spacing adopts a constant time headway (CTH) strategy, which is a commonly used spacing strategy in existing queue studies and can improve the stability of queue movement. Based on this spacing strategy, the desired queue spacing strategy is as follows:

[0159]

[0160] Where h represents the time interval coefficient and r represents the stationary distance of the vehicle. Since the queue is a homogeneous queue, this parameter applies to all vehicles in the queue.

[0161] The actual vehicle spacing is:

[0162]

[0163] At this point, the spacing error in the queue It can be represented as:

[0164]

[0165] In this embodiment, the spacing error between vehicles in the queue tends to be 0. When designing the queue system, the stationary distance r of the vehicles is set to 0m, and the speed error in stage j of the queue—that is, the deviation between the speed transmitted from the cloud and the actual speed of vehicle i in the queue—is set as follows:

[0166]

[0167] in, The queue velocity at time j is dynamically planned for the rolling distance domain.

[0168] Furthermore, a dynamic model of the vehicle considering the time lag of the drive is established. Based on reality, the time lag of the drive needs to be considered in vehicle dynamics, rather than assuming the acceleration process is instantaneous, because the vehicle system requires a certain amount of time to reach the required acceleration.

[0169]

[0170] Among them, ai (t) represents the vehicle's acceleration, u i (t) represents the vehicle control input. The desired acceleration of the vehicle; This refers to the vehicle's drive time lag.

[0171] At this point, the system state is defined. for:

[0172]

[0173]

[0174] Where x(t)∈R n Represents the state of the system, u(t)∈R n Indicates control input, a i-1 (t)∈R n This indicates a interference input.

[0175] Furthermore, the state space of the prediction model is discretized based on the vehicle's dynamics model. Although vehicle dynamics is a continuous state, it needs to be discretized during the solution process. By assuming that the control input is a zero-order hold, a discrete version of the state space can be obtained.

[0176] x i,k+1 =A′ i x i,k +B′ i u i,k +D′ i a i-1,k (25)

[0177]

[0178]

[0179]

[0180] Based on the system's current state and the predicted future state, a system dynamics model is used within the prediction range k. p The optimal control problem is solved at each time step k. The controller implements only the first control input (i.e., the optimal solution at time step k) and recalculates the optimal control within the prediction range starting from the next time step k+1. Here, the optimal control sequence (i.e., the acceleration specified by the controller), the predicted future state of the optimal control, and the actual implementation state with unmodeled and unknown disturbances are:

[0181] This represents the distance from time k to the prediction range k. p The optimal control sequence for the vehicle;

[0182] This represents the distance from time k to the prediction range k. p The desired state of the vehicle;

[0183] This represents the distance from time k to the prediction range k. p The actual condition of the vehicle, among which... This represents the initial state of vehicle control.

[0184] Setting N to the prediction time domain of model predictive control, the prediction model can be derived as follows:

[0185]

[0186] Where x(k+j|k) is the system state prediction at time k+j. Since the system input u(k) at time k is unknown, model predictive control is needed to solve for u(k). Control is achieved through incremental control Δu, which means that the predictive control input at time k to time k+j can be transformed into the sum of the control quantity at time k+j-1 and the incremental predictive control input at time k+j, i.e.:

[0187] u(k+j|k)=u(k+j-1|k)+Δu(k+j|k); (30)

[0188] Since u(k-1|k) is known at time k, it can be obtained recursively:

[0189]

[0190] Where, N c To control the time domain, the relationship between the control time domain and the prediction time domain satisfies N c Since the value is less than or equal to N, the model predictions can be rewritten.

[0191] To minimize computational overhead, the control time domain of the model prediction is shorter than the prediction time domain. Therefore, the model prediction needs to be divided into two segments, where 1 ≤ j ≤ N. c At that time, the prediction model is:

[0192]

[0193] When N c When ≤j≤N, the prediction model is:

[0194]

[0195] From the above equations, we can obtain the state transition equations for model predictive control:

[0196] X(k)=Ex(k)+Φ△U(k)+Γu(k-1)+Hw(k); (34)

[0197] X(k)=[x(k+1|k) x(k+2|k) … x(k+N c |k) x(k+N c +1|k) … x(k+N|k)] T (34-a)

[0198] △U=[△u(k|k) … △u(k+N c -1|k)] T (34-b)

[0199]

[0200]

[0201]

[0202]

[0203] The above content completes the state description of the entire system in the prediction model. The following will determine the prediction model of the embodiment of this application based on the state equation of the prediction model derived above.

[0204] The system constraint model sets the constraints for the optimal control problem of queue stability control. It sets the constraints that the queue stability controller needs to meet during the design process, including safety constraints, queue stability requirements, speed range constraints, ride smoothness constraints, and vehicle actuator constraints, as shown in Equation (3) for queue system constraints.

[0205] The cost function module of the queue stability controller is the objective function of the queue stability control optimization problem. This optimization problem is constructed under collision-free and vehicle acceleration constraints, comprehensively considering the vehicle tracking control accuracy and driving comfort within the queue. Based on the objective of optimal control, the cost function is defined as follows:

[0206]

[0207] Where, k p For the prediction range, The costs of each stage of the system before reaching equilibrium. The terminal cost from the end state to the desired point. The initial condition constraint is equal to the measured state at time k. The state cost for each stage, such as the cost of adjusting the maximum deviation from the desired output or acceleration constraint at each time point, u i,t+k Acceleration constraints are used to ensure that the control quantity is within U i ( i )=[u i,mun ,u i,max Within a reasonable range, where u i,min and u i,max These represent the upper and lower limits of acceleration, respectively. These are terminal state constraints used to adjust the terminal state to tend towards the desired terminal state.

[0208] To derive optimal control, stage costs and terminal costs can be specified, thereby deriving local stability and global string stability. The stage cost function is:

[0209]

[0210] Among them, Q i Let R be the positive definite state matrix of the desired output. i To account for the weight of comfort, it is always greater than 0.

[0211]

[0212] The terminal state cost is: Among them, P d,i As a solution to the discrete algebraic regression equation to ensure local stability, P d,i The expression is:

[0213]

[0214] Based on the objective function and constraints of the queue stability controller, the optimization control problem is transformed into a quadratic programming problem for solution.

[0215] The vehicle queue receives control commands from the onboard platform to achieve economic and stability control of the queue. At the start of the planning cycle, it feeds back its status to the cloud control platform to re-plan the queue speed. In addition, the vehicle queue also includes the specific actuator systems of the vehicles.

[0216] In summary, the control flowchart of a cloud-based hierarchical vehicle platooning predictive cruise control system according to one embodiment of this application is as follows: Figure 8 As shown, it includes the following steps:

[0217] Step S801, determine Is the condition met? If it is met, proceed to step S802; otherwise, proceed to step S815.

[0218] Step S802: Obtain vehicle location from the cloud.

[0219] Step S803: Obtain the slope information Xkm ahead of the vehicle queue.

[0220] Step S804: Determine if X < S is true. If true, proceed to step S805; otherwise, end the process.

[0221] Step S805: Divide the state of the lead vehicle in the vehicle platoon within the planning period.

[0222] Step S806: Generate the DP state space.

[0223] Step S807: Solve for the state transition cost function.

[0224] Step S808: Solve for the optimal velocity sequence in DP.

[0225] Step S809: Generate the optimal reference velocity curve.

[0226] Step S810: Determine the stability error of the vehicle-side controller queue.

[0227] Step S811: Perform model prediction.

[0228] Step S812: Generate the optimal control sequence for the vehicle (i.e., the optimal driving speed).

[0229] Step S813: Send to the vehicle's underlying control.

[0230] Step S814, proceed to the next rolling planning cycle (return to step S801).

[0231] Step S815, determine Is the condition met? If it is met, proceed to step S816; otherwise, proceed to step S817.

[0232] Step S816,

[0233] Step S817,

[0234] The cloud-based hierarchical vehicle platoon predictive cruise control method proposed in this application (1) combines real-time and historical road traffic data available in the cloud, which can not only achieve wide-area and long-term perception, but also perform fast real-time planning and decision calculations. At the same time, it can greatly alleviate the computing pressure on the vehicle end, and has great potential to save energy and improve the safety and stability boundary of the platoon system; (2) the vehicle-cloud communication method and specific implementation algorithm, compared with the traditional platoon predictive cruise control, reduce the related vehicle-cloud architecture design and vehicle-cloud communication structure design combined with the cloud. At the same time, it utilizes the long-term static traffic flow information in the cloud, making the planning more forward-looking. The design of the planning control algorithm under the rolling distance domain can offset the uncertainty of future platoon operation; (3) vehicle-cloud hierarchical control, the speed planning algorithm is deployed in the cloud, and the economic speed of platoon driving is solved by the dynamic planning algorithm under the rolling distance domain considering the road slope. After the optimal reference speed curve is issued in the cloud, the platoon stability control algorithm is deployed on the vehicle end to realize the stable control of the vehicle platoon, thereby realizing the design of a cloud-supported vehicle platoon predictive cruise control system.

[0235] The cloud-based hierarchical vehicle platoon predictive cruise control method proposed in this application receives predictive cruise control requests from vehicle platoons and obtains the real-time location and lead vehicle status information of the platoon based on these requests. It then determines the optimal reference speed curve for the platoon based on the real-time location, lead vehicle status information, a preset fuel consumption model, and a preset dynamics model, and sends this curve to the onboard platform. The onboard platform then sends the optimal driving speed, obtained from the optimal reference curve, preset platoon system constraints, and a preset platoon control error model, to the vehicle platoon for cruise control. Therefore, by combining the real-time location and status information of vehicles with a cloud platform, this method can solve the problems of limited prediction range, insufficient information acquisition capabilities, and platoon predictive speed planning in existing platoon predictive cruise systems, enabling more advanced and intelligent planning of vehicle platoon speeds.

[0236] Next, referring to the accompanying drawings, a cloud-based hierarchical vehicle platoon predictive cruise control device according to an embodiment of this application is described.

[0237] Figure 9 This is a block diagram of a cloud-based hierarchical vehicle platoon predictive cruise control device according to an embodiment of this application.

[0238] like Figure 9 As shown, the cloud-based hierarchical vehicle platoon predictive cruise control device 10 includes: a receiving module 100, an acquisition module 200, and a control module 300.

[0239] The receiving module 100 is used to receive predictive cruise control requests from the vehicle convoy.

[0240] The acquisition module 200 is used to acquire the real-time location of the vehicle convoy and the status information of the lead vehicle based on predictive cruise control requests; and

[0241] The control module 300 is used to determine the optimal reference speed curve of the vehicle convoy based on the real-time location, the status information of the navigator vehicle, the preset fuel consumption model and the preset dynamics model, and send the optimal reference speed curve to the vehicle platform. The vehicle platform then sends the optimal driving speed obtained from the optimal reference curve, the preset convoy system constraints and the preset convoy control error model to the vehicle convoy, and performs cruise control on the vehicle convoy based on the optimal driving speed.

[0242] Furthermore, in some embodiments, the control module 300 is specifically used for:

[0243] Determine the road gradient information ahead of the vehicle convoy based on real-time location;

[0244] Based on the preset dynamic model, the longitudinal force on each vehicle in the vehicle platoon is calculated according to the road slope information ahead of the vehicle platoon and the status information of the lead vehicle.

[0245] The lead vehicle in the vehicle platoon is divided into states within the planning period. Based on the division results and the preset speed planning cost function, the optimal reference speed curve of the vehicle platoon is determined according to the preset fuel consumption model and the longitudinal force on each vehicle in the platoon.

[0246] Furthermore, in some embodiments, the preset fuel consumption model is:

[0247]

[0248] Where, ξ i,j T is the parameter fitted to the preset fuel consumption model. tq n is the vehicle engine torque, and n is the vehicle engine speed.

[0249] Furthermore, in some embodiments, the preset dynamic model is:

[0250]

[0251] Where, m i For the quality of the vehicle, For the vehicle's acceleration, F e,i For engine traction, F g,i For slope resistance, F r,i For rolling resistance, F air,i This refers to air resistance during vehicle operation.

[0252] Furthermore, in some embodiments, preset queue system constraints

[0253]

[0254] Among them, e s (t) represents the vehicle spacing error in the queue, e v(i,0) (t) represents the initial speed error of the vehicles in the queue. For the speed of the vehicle, For the impact force of the vehicle, u i (t) represents the control quantity of the vehicle.

[0255] Furthermore, in some embodiments, the preset speed planning cost function is:

[0256]

[0257] in, Let J(j,k+1) be the cost function from the current state to the next state, and let J(j,k+1) be the cost from the next state to the final state.

[0258] Furthermore, in some embodiments, the vehicles within the same queue communicate via a first communication topology, and the vehicle queue communicates with the cloud via a second communication topology, wherein...

[0259] The first communication topology is:

[0260] Q i ={j|α i,j =1,j∈N};

[0261] The second communication topology is:

[0262]

[0263] Where, α i,j Let N be the element in the adjacent matrix, N be a natural number, and E be the connected cloud nodes.

[0264] It should be noted that the foregoing explanation of the embodiment of the predictive cruise control method for cloud-based hierarchical vehicle platoons also applies to the predictive cruise control device for cloud-based hierarchical vehicle platoons in this embodiment, and will not be repeated here.

[0265] The cloud-based hierarchical vehicle platoon predictive cruise control device proposed in this application receives predictive cruise control requests from vehicle platoons and obtains the real-time location of the platoon and the status information of the lead vehicle based on the requests. It then determines the optimal reference speed curve for the vehicle platoon based on the real-time location, the lead vehicle status information, a preset fuel consumption model, and a preset dynamics model, and sends this curve to the onboard platform. The onboard platform then sends the optimal driving speed, obtained from the optimal reference curve, preset platoon system constraints, and a preset platoon control error model, to the vehicle platoon for cruise control. Therefore, by combining the real-time location and status information of the vehicles with a cloud platform, the device can solve the problems of limited prediction range, insufficient information acquisition capabilities, and predictive speed planning in existing platoon predictive cruise systems, achieving more advanced and intelligent planning of vehicle platoon speeds.

[0266] Figure 10 A schematic diagram of the structure of a server provided in an embodiment of this application. The server may include:

[0267] The server includes a processor 1001, a storage device 1002, and a communication device 1003; the number of processors 1001 in the server can be one or more. Figure 10 Taking a processor 1001 as an example; the processor 1001, storage device 1002, and communication device 1003 in the server can be connected via a bus or other means. Figure 10 Taking the bus connection method between China and Israel as an example.

[0268] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described cloud-based hierarchical vehicle platoon predictive cruise control method.

[0269] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0270] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0271] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A cloud-based, hierarchical vehicle platooning predictive cruise control method, characterized in that, The method is applied to a cloud-based hierarchical vehicle platoon predictive cruise control system. The system includes a cloud control platform, which comprises a preset fuel consumption model, a preset dynamics model, and a cloud-based speed planning algorithm module. The cloud-based speed planning algorithm module includes a state partitioning module, a speed planning cost function module, and a speed calculation module. The method includes the following steps: Receive the predictive cruise control request from the vehicle convoy; Based on the predictive cruise control request, the real-time location of the vehicle convoy and the status information of the lead vehicle are obtained; and The optimal reference speed curve of the vehicle platoon is determined based on the real-time location, the navigator status information, the preset fuel consumption model, and the preset dynamics model. The optimal reference speed curve is then sent to the vehicle platform. The vehicle platform then sends the optimal driving speed obtained from the optimal reference curve, the preset platoon system constraints, and the preset platoon control error model to the vehicle platoon. The vehicle platoon is then cruise-controlled using the optimal driving speed. The step of determining the optimal reference speed curve of the vehicle convoy based on the real-time location, the navigator vehicle status information, the preset fuel consumption model, and the preset dynamics model includes: determining the optimal reference speed curve of the vehicle convoy through a cloud-based speed planning algorithm based on the real-time location, the navigator vehicle status information, the preset fuel consumption model, and the preset dynamics model, wherein the cloud-based speed planning algorithm is integrated into the cloud-based speed planning algorithm module. The step of determining the optimal reference speed curve of the vehicle convoy using a cloud-based speed planning algorithm includes: using the state partitioning module to partition the speed solution space of the lead vehicle in the vehicle convoy within the planning period to obtain a partitioned state space; based on the partitioned state space and a preset speed planning cost function, using the speed planning cost function module to set a control objective for speed planning; based on the control objective, using the speed solution module to index the preset speed planning cost function, solving for the value of the cost function, and determining the optimal reference speed curve based on the value of the cost function.

2. The method according to claim 1, characterized in that, The step of determining the optimal reference speed curve for the vehicle convoy based on the real-time location, the lead vehicle status information, a preset fuel consumption model, and a preset dynamics model includes: The road gradient information ahead of the vehicle convoy is determined based on the real-time location. Based on the preset dynamic model, the longitudinal force on each vehicle in the vehicle convoy is calculated according to the road slope information ahead of the vehicle convoy and the status information of the lead vehicle. The lead vehicle of the vehicle platoon is divided into states within the planning period. Based on the division results and the preset speed planning cost function, the optimal reference speed curve of the vehicle platoon is determined according to the preset fuel consumption model and the longitudinal force on each vehicle in the vehicle platoon.

3. The method according to claim 1 or 2, characterized in that, The preset fuel consumption model is as follows: ； in, The parameters are fitted to the preset fuel consumption model. This refers to the vehicle's engine torque. This refers to the vehicle's engine speed.

4. The method according to claim 2, characterized in that, The preset dynamic model is as follows: ； in, For the quality of the vehicle, For the vehicle's acceleration, For engine traction, For slope resistance, For rolling resistance, This refers to air resistance during vehicle operation.

5. The method according to claim 2, characterized in that, The preset queue system constraints ； in, This refers to the error in vehicle spacing within the queue. The initial speed error of the vehicles in the queue. For the speed of the vehicle, For the impact force of the vehicle, For vehicle control quantities.

6. The method according to claim 2, characterized in that, The preset speed planning cost function is: ； in, Let be the cost function from the current state to the next state. The cost of moving from the next state to the final state.

7. The method according to claim 1, characterized in that, The vehicles within the same queue communicate via a first communication topology, and the vehicle queue communicates with the cloud via a second communication topology. The first communication topology is: ； The second communication topology is: ； in, For elements in adjacent matrices, N For natural numbers, E Let be a matrix representing connectivity relationships in graph theory.

8. A cloud-based, hierarchical vehicle platooning predictive cruise control device, characterized in that, The device is applied to the cloud-based hierarchical vehicle platooning predictive cruise control method as described in any one of claims 1-7, the device comprising: A receiving module is used to receive predictive cruise control requests from the vehicle convoy; The acquisition module is used to acquire the real-time location of the vehicle convoy and the status information of the lead vehicle based on the predictive cruise control request; and The control module is used to determine the optimal reference speed curve of the vehicle convoy based on the real-time location, the status information of the lead vehicle, a preset fuel consumption model, and a preset dynamics model, and to send the optimal reference speed curve to the vehicle platform. The vehicle platform then sends the optimal driving speed obtained from the optimal reference speed, the preset convoy system constraints, and the preset convoy control error model to the vehicle convoy, and performs cruise control on the vehicle convoy using the optimal driving speed.

9. A server, characterized in that, include: One or more processors; Storage device; an application that stores one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the cloud-based hierarchical vehicle platooning predictive cruise control method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the cloud-based, hierarchical vehicle platooning predictive cruise control method as described in any one of claims 1-7.

Images