Iterative learning predictive control method and device for energy-saving driving of intelligent network-connected motorcade
Through the iterative learning of predictive control methods for energy-saving driving of intelligent connected fleets, combined with global and local optimization models, the problem of insufficient decision-making optimization in traditional methods is solved, and a more economical fleet driving method is achieved.
Patent Information
- Application Number
- CN202410803894.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-20
- Publication Date
- 2024-11-08
AI Technical Summary
Traditional driving decision generation methods have shortcomings in terms of optimization, making it difficult to update decision results in real time, resulting in poor performance of autonomous driving fleets in driving strategies that consume the least energy.
The iterative learning prediction control method of the energy-saving driving of an intelligent connected fleet is adopted. By obtaining the global optimization model and current driving information of the target vehicle in the fleet, the initial control volume is obtained, and the driving control strategy iteratively optimizes during driving, combining global and local optimization models. To improve decision optimization.
The optimization of autonomous driving decisions is improved, and a more economical fleet driving method is achieved. Through preliminary analysis of the global optimization model and real-time adjustment of the local optimization model, the fleet's energy-saving driving strategy is optimized.
Smart Images

Figure CN118915430A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of vehicle dynamics control technology, and in particular to an iterative learning predictive control method and device for energy-saving driving of an intelligent networked fleet. Background Art
[0002] With the development of autonomous driving technology, how to make a fleet of multiple vehicles drive with the least energy consumption has become an urgent problem that needs to be solved.
[0003] Traditional driving decision generation methods usually use global optimization models or local optimization algorithms in a rolling time domain framework to output driving decisions. However, traditional driving decision generation methods have poor optimization performance. Summary of the invention
[0004] Based on this, it is necessary to provide an iterative learning predictive control method and device for energy-saving driving of an intelligent connected fleet that can improve optimization in response to the above technical problems.
[0005] In a first aspect, the present application provides an iterative learning predictive control method for energy-saving driving of an intelligent connected vehicle fleet, comprising:
[0006] Obtain the global optimization model and current driving information of the target vehicle in the fleet, and obtain the initial control amount of the target vehicle based on the driving information and the global optimization model;
[0007] Controlling the target vehicle to travel according to the initial control amount, and obtaining an initial travel trajectory of the target vehicle during the travel of the target vehicle;
[0008] According to the initial driving trajectory, an initial local optimization model of the target vehicle is obtained, and according to the initial local optimization model, the driving control strategy of the target vehicle is iteratively optimized.
[0009] In one embodiment, obtaining a global optimization model of a target vehicle in a fleet includes:
[0010] Obtain the fuel consumption model and dynamic model of the target vehicle, where the fuel consumption model is related to the total fuel consumption of the vehicles in the fleet;
[0011] According to the dynamic model, a spatial discrete model is obtained, and a global optimization model is determined according to the fuel consumption model and the spatial discrete model.
[0012] In one embodiment, the initial control amount includes a control amount sequence, and according to the initial control amount, controlling the target vehicle to travel includes:
[0013] The target vehicle is controlled to travel according to the first control variable in the control variable sequence.
[0014] In one embodiment, obtaining an initial local optimization model of the target vehicle according to the initial driving trajectory includes:
[0015] According to the dynamics model, a time discrete model is obtained;
[0016] An initial local optimization model is obtained according to the time discrete model, the fuel consumption model and the initial driving trajectory.
[0017] In one embodiment, the initial driving trajectory includes an initial state quantity and an initial fuel consumption corresponding to the initial state quantity. According to the time discrete model, the fuel consumption model and the initial driving trajectory, an initial local optimization model is obtained, including:
[0018] Obtain the terminal constraint set corresponding to the initial state quantity according to the initial state quantity;
[0019] Obtain a terminal fuel consumption function corresponding to the initial fuel consumption according to the initial fuel consumption;
[0020] An initial local optimization model is obtained according to the time discrete model, the fuel consumption model, the approximate terminal constraint set corresponding to the terminal constraint set, and the approximate terminal fuel consumption function corresponding to the terminal fuel consumption function.
[0021] In one embodiment, obtaining the initial control amount of the target vehicle according to the driving information and the global optimization model includes:
[0022] Solve the global optimization model to obtain the global optimization speed;
[0023] Determine the static local model based on the driving information, the global optimization speed and the time discrete model;
[0024] The static local optimization model is solved to obtain the initial control quantity.
[0025] In one embodiment, iteratively optimizing the driving control strategy of the target vehicle according to the initial local optimization model includes:
[0026] Solve the initial local optimization model to obtain the optimal control quantity of the target vehicle;
[0027] According to the optimized control amount, the target vehicle is controlled to continue to move, and during the process of the target vehicle continuing to move, an intermediate moving trajectory of the target vehicle is obtained;
[0028] According to the intermediate driving trajectory, an intermediate local optimization model of the target vehicle is obtained, and according to the intermediate local optimization model, the driving control strategy of the target vehicle is continuously iteratively optimized until the target vehicle finishes driving in the current target road section.
[0029] In one embodiment, the method further comprises:
[0030] When the target vehicle has finished traveling in the current target road section, determining whether the target local optimization model obtained by the last iterative optimization of the target vehicle in the target road section has converged;
[0031] If the target local optimization model converges, when the target vehicle re-drives the target road section, the driving control strategy of the target vehicle is obtained according to the target local optimization model;
[0032] If the target local optimization model has not converged, when the target vehicle re-drives the target section, the local optimization model of the target vehicle is reconstructed according to the target driving trajectory corresponding to the target local optimization model, and the driving control strategy of the target vehicle is obtained according to the reconstructed local optimization model.
[0033] In a second aspect, the present application also provides an iterative learning prediction control device for energy-saving driving of an intelligent networked fleet, comprising:
[0034] A first acquisition module is used to acquire a global optimization model and current driving information of a target vehicle in a fleet, and acquire an initial control amount of the target vehicle according to the driving information and the global optimization model;
[0035] A second acquisition module, used for controlling the target vehicle to travel according to the initial control amount, and obtaining an initial travel trajectory of the target vehicle during the travel of the target vehicle;
[0036] The control module is used to obtain an initial local optimization model of the target vehicle according to the initial driving trajectory, and iteratively optimize the driving control strategy of the target vehicle according to the initial local optimization model.
[0037] In a third aspect, an embodiment of the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method of the first aspect when executing the computer program.
[0038] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of the first aspect described above.
[0039] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the method in the first aspect are implemented.
[0040] The above-mentioned intelligent networked fleet energy-saving driving iterative learning predictive control method and device can obtain the global optimization model and current driving information of the target vehicle in the fleet, and obtain the initial control amount of the target vehicle according to the driving information and the global optimization model. Then, according to the initial control amount, the target vehicle is controlled to drive, and the initial driving trajectory of the target vehicle is obtained during the driving process of the target vehicle. In this way, the initial local optimization model of the target vehicle can be obtained according to the initial driving trajectory, and the driving control strategy of the target vehicle can be iteratively optimized according to the initial local optimization model. In this way, the global optimization model is first used to perform a global analysis on the form information, and the initial control amount can be preliminarily obtained. In the process of controlling the operation of the target vehicle using the initial control amount, the initial driving trajectory can be obtained, and the initial local optimization model is constructed through the initial driving trajectory. The form control strategy of the target vehicle during driving is continuously optimized using the initial local optimization model. The global optimization model is used to perform a global analysis, and the initial local optimization model is used to make real-time decisions, thereby improving the optimization of automatic driving decisions. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0042] Figure 1 A diagram of an application environment of an iterative learning predictive control method for energy-saving driving of an intelligent network-connected fleet in one embodiment;
[0043] Figure 2 A schematic diagram of a flow chart of an iterative learning predictive control method for energy-saving driving of an intelligent network-connected fleet in one embodiment;
[0044] Figure 3 A schematic diagram of vehicle forces in another embodiment;
[0045] Figure 4 is a fitting diagram of the relationship between the vehicle spacing and the air resistance coefficient in another embodiment;
[0046] Figure 5 is a schematic flow chart of step 201 in another embodiment;
[0047] Figure 6 is a schematic flow chart of step 203 in another embodiment;
[0048] Figure 7 A schematic diagram of a flow chart of iteratively optimizing a driving control strategy of a target vehicle in another embodiment;
[0049] Figure 8 A flowchart of an exemplary iterative learning predictive control method for energy-saving driving of an intelligent networked fleet in one embodiment;
[0050] Fig. 9 It is a structural block diagram of an iterative learning prediction control device for energy-saving driving of an intelligent network-connected fleet in one embodiment;
[0051] Fig.10 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION
[0052] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0053] With the progress and development of society, people's demand for travel safety, convenience and economy is increasing. In order to achieve intelligent, safe, energy-saving and environmentally friendly driving of road vehicles, autonomous driving technology has been proposed and gradually developed.
[0054] In autonomous driving technology, by integrating on-board sensors, wireless communications, computer control, navigation and positioning technologies, and adopting reasonable control strategies, the vehicle's autonomous driving system can autonomously calculate the optimal control decisions, thereby reducing driver error and using road traffic information to avoid bad driving behavior and achieve cruise control, lane keeping, energy-saving driving, etc.
[0055] With the development of autonomous driving technology, how to make a fleet of multiple vehicles drive with the least energy consumption has become an urgent problem that needs to be solved.
[0056] Traditional driving decision generation methods usually use global optimization models or local optimization algorithms in a rolling time domain framework to output driving decisions.
[0057] Global optimization models include Dijkstra algorithm, A* algorithm, artificial neural network algorithm, particle swarm optimization algorithm, genetic algorithm and dynamic programming. These algorithms can solve the global optimal control strategy of the vehicle, but because the global optimization model is an offline algorithm, it is impossible to analyze the driving environment in real time and it is difficult to update the decision results in time; although the local optimization algorithm can generate decisions in real time according to the current environment, it can only analyze the driving conditions in a shorter time range in the future.
[0058] Therefore, the driving decisions generated by traditional driving decision generation methods are less optimized.
[0059] In view of this, the embodiment of the present application provides an iterative learning predictive control method and device for energy-saving driving of an intelligent networked fleet. By obtaining the global optimization model and current driving information of the target vehicle in the fleet, and according to the driving information and the global optimization model, the initial control amount of the target vehicle can be obtained. Then, according to the initial control amount, the target vehicle is controlled to drive, and during the driving process of the target vehicle, the initial driving trajectory of the target vehicle is obtained. In this way, the initial local optimization model of the target vehicle can be obtained according to the initial driving trajectory, and the driving control strategy of the target vehicle can be iteratively optimized according to the initial local optimization model. In this way, the global optimization model is first used to perform a global analysis on the form information, and the initial control amount can be preliminarily obtained. In the process of controlling the operation of the target vehicle using the initial control amount, the initial driving trajectory can be obtained, and the initial local optimization model is constructed through the initial driving trajectory. The form control strategy of the target vehicle during driving is continuously optimized using the initial local optimization model. The global optimization model is used to perform a global analysis, and the initial local optimization model is used to make real-time decisions, thereby improving the optimization of the automatic driving decision.
[0060] The intelligent networked fleet energy-saving driving iterative learning predictive control method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the data storage system can store the data that the server 101 needs to process. The data storage system can be integrated on the server 101, or it can be placed on the cloud or other network servers. The server 101 can be deployed on any vehicle in the fleet. The server 101 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.
[0061] In an exemplary embodiment, Figure 2 As shown in the figure, an iterative learning predictive control method for energy-saving driving of an intelligent networked fleet is provided. Figure 1 The server 101 in the example is used as an example to illustrate, including the following steps 201 to 203. Among them:
[0062] Step 201, obtaining a global optimization model and current driving information of a target vehicle in a fleet, and obtaining an initial control amount of the target vehicle according to the driving information and the global optimization model.
[0063] In the embodiment of the present application, the fleet is the fleet in which the target vehicle is located, the fleet includes multiple vehicles, and the target vehicle can be any vehicle in the fleet.
[0064] The convoy is traveling on a preset road. In the embodiment of the present application, the vehicles in the convoy can be set to adopt a time interval following strategy, that is, the front and rear wheels maintain a constant time interval Δt, and the distance intervals between the vehicles are equal:
[0065] S i,t =S i-1,t-Δt (1)
[0066] Among them, S i,t Represents the distance between the i-th vehicle and the vehicle in front.
[0067] Therefore, each vehicle in the convoy can maintain the same speed:
[0068] V i,z =V i-1,z (2)
[0069] The distance between vehicles is d iz :
[0070] d iz =v i-1 Δt-l i-1 (3)
[0071] The global optimization model may be a dynamic programming model for comprehensively analyzing the current driving information of the target vehicle. In an embodiment of the present application, in order to achieve the purpose of energy-saving driving, the ultimate optimization goal of the global optimization model may be to minimize fuel consumption.
[0072] Driving information may be information of the target vehicle on a preset path, including vehicle mass, vehicle longitudinal speed, vehicle longitudinal displacement, braking force, driving force, ramp force, rolling resistance, air resistance, etc., wherein the server may obtain vehicle mass, vehicle longitudinal speed, vehicle longitudinal displacement, braking force, and driving force through sensors, and regarding ramp force, rolling resistance, and air resistance, the server may calculate them based on vehicle mass, vehicle longitudinal speed, vehicle longitudinal displacement, braking force, driving force, etc.
[0073] For example, refer to Figure 3 The ramp force is the force exerted on the target vehicle due to the ramp angle when it is driving on the ramp. When the target vehicle is in an uphill state, the ramp force is a resistance force. When the target vehicle is in a downhill state, the ramp force is a driving force. g,j (α(s i )) is calculated as follows:
[0074] F g,i (α(s i ))=-m i gsin(α(s i )) (4)
[0075] Among them, α(s i ) is the only vertical i is the ramp angle at , and m is the target vehicle mass.
[0076] About α(s i ) is calculated as follows:
[0077] α(s i )=a0+a1s i +a2s i 2 (5)
[0078] Among them, α0, α1 and α2 are time-related parameters, which are obtained by fitting and updated as the position of the target vehicle changes.
[0079] When the target vehicle is driving on a slope, the tires will be affected by the hysteresis effect, that is, the target vehicle will be affected by the rolling resistance. The rolling resistance F r,i (α(s i )) is calculated as follows:
[0080] F r,i (α(s i ))=-C r,i m i gcos(α(s i )) (6)
[0081] Wherein, Cr,i is the rolling resistance coefficient.
[0082] The calculation formula for the construction period resistance Fd,i (vi, di) is as follows:
[0083]
[0084] Among them, v i is the current speed of the target vehicle, d i is the function of the air resistance caused by the vehicle spacing of the target vehicle, A i is the cross-sectional area of the target vehicle, ρ a is the air density, C D (d i ) is the air resistance coefficient.
[0085] The relationship between the air resistance coefficient and the vehicle spacing can be referred to the following formula:
[0086]
[0087] Among them, C D,0 is the air resistance coefficient when there is only one car in the convoy, C D,0(1-φ(di)) is the air resistance coefficient when the number of vehicles in the convoy exceeds 1. D,0 (1-φ(di)) can be obtained by experimental fitting, refer to Figure 4 , is the relationship between vehicle spacing and air resistance coefficient.
[0088] In an embodiment of the present application, after the server analyzes the current driving information through a global optimization model, it can preliminarily obtain an initial control amount, and the control amount can be a parameter used to control the driving of the target vehicle.
[0089] Step 202: Control the target vehicle to travel according to the initial control amount, and obtain the initial travel trajectory of the target vehicle during the travel of the target vehicle.
[0090] The control amount may include a driving force and a braking force, wherein the driving force is used to represent the force for the target vehicle to move forward, and the braking force represents the force for preventing the target vehicle from moving forward.
[0091] In the embodiment of the present application, after the initial control amount is acquired, the server may control the target vehicle to travel according to the driving force and braking force included in the initial control amount.
[0092] Regarding the process in which the server controls the target vehicle to travel according to the initial control amount, in a possible implementation, the server may control the driving force of the target vehicle by controlling the engine speed and torque of the target vehicle, thereby controlling the longitudinal speed of the target vehicle. e,i The calculation formula is as follows:
[0093] F e,i =IηT i / r i (9)
[0094] Among them, I is the total transmission ratio of the vehicle, η is the transmission efficiency, ri is the wheel class of the target vehicle, and Ti is the torque of the target vehicle.
[0095] The longitudinal velocity of the target vehicle is calculated as follows:
[0096] v i =ω i r i / I (10)
[0097] Where ωi is the engine speed.
[0098] In a possible implementation, the initial control amount may include a control amount sequence, which includes control amounts of multiple stages. In an embodiment of the present application, the server may control the driving of the target vehicle according to the first control amount in the control amount sequence. Since the first control amount is the control amount closest to the current stage, the first control amount is more optimized.
[0099] After obtaining the initial control amount, the server can use the initial control amount to control the driving of the target vehicle, and obtain the initial driving trajectory corresponding to the initial control amount during the driving of the target vehicle. In this embodiment of the present application, the initial driving trajectory may include the driving status corresponding to each time point and the fuel consumption corresponding to each driving status.
[0100] Step 203: acquiring an initial local optimization model of the target vehicle according to the initial driving trajectory, and iteratively optimizing the driving control strategy of the target vehicle according to the initial local optimization model.
[0101] The initial driving trajectory is a driving trajectory obtained by the server using the initial control amount to control the driving of the target vehicle. According to the initial driving trajectory, the server can construct an initial local optimization model.
[0102] In an embodiment of the present application, the driving control strategy of the target vehicle can be continuously iteratively optimized through the initial local optimization model, so as to obtain the driving control strategy that best suits the preset path.
[0103] In a possible implementation, after the target vehicle has traveled a preset target road, it is an iterative process. In the embodiment of the present application, the server can continuously optimize the driving control strategy by continuously solving the initial local optimization model. The driving control strategy may include the predicted control amount corresponding to each prediction step.
[0104] Regarding the method of stopping iteration by the server, optionally, the server presets an iteration number threshold, and when the iteration number reaches the iteration number threshold, it is determined that the initial local optimization model has converged; optionally, for one iteration process, the server obtains the driving control strategy corresponding to the iteration based on the initial local optimization model, and the server can control the target vehicle to drive with the driving control strategy corresponding to the iteration, thereby obtaining the driving trajectory corresponding to the driving control strategy of the iteration, so as to obtain the driving trajectory corresponding to multiple iteration processes, and detect the volatility of the driving trajectory. When the volatility is less than the preset threshold, the server can determine that the initial local model corresponding to the iteration process has converged.
[0105] When the local model converges, the server can update and optimize the driving control strategy in real time based on the local model.
[0106] In the above embodiment, the global optimization model is first used to perform a global analysis of the form information, and the initial control amount can be preliminarily obtained. In the process of using the initial control amount to control the operation of the target vehicle, the initial driving trajectory can be obtained, and the initial local optimization model is constructed through the initial driving trajectory. The initial local optimization model is used to continuously optimize the form control strategy during the driving process of the target vehicle. The global optimization model is used to perform a global analysis, and the initial local optimization model is used to make real-time decisions, thereby improving the optimization of autonomous driving decisions.
[0107] In one embodiment, based on the above Figure 2 The embodiment shown, see Figure 5 , this embodiment relates to the process of obtaining the global optimization model of the target vehicle in the fleet. Figure 5 As shown, step 201 may include step 501 and step 502 .
[0108] Step 501, obtaining a fuel consumption model and a dynamic model of a target vehicle.
[0109] In a possible implementation, the server may obtain the fuel consumption model and the dynamic model of the target vehicle. Optionally, the server may obtain the fuel consumption model and the dynamic model through an external input device. Optionally, the server may obtain the fuel consumption model and the dynamic model in a local storage space.
[0110] Since the target vehicle is subject to longitudinal force on the road, the dynamic model of the target vehicle can refer to the following formula:
[0111]
[0112] Among them, v′ i is the longitudinal velocity v i The derivative of is the longitudinal displacement s i The derivative of . Subscript i∈M={1,……,j}∈N+, j is the total number of vehicles in the fleet.
[0113] In the embodiment of the present application, the dynamic model of the vehicle can be simplified into the following formula:
[0114]
[0115] Among them, x represents the state quantity, which can include longitudinal velocity and longitudinal displacement. The state quantity of the i-th vehicle can be expressed as X i =[v i ,s i ] T , control quantity u i It can include the driving force and braking force of the vehicle. The control amount of the i-th vehicle can be recorded as ui =[F e,i , F b,i ] T .
[0116] Then the state quantity satisfies the following constraints:
[0117] x i ∈X={v i :0≤v i ≤v max} (13)
[0118] The control quantity satisfies the following constraints:
[0119] u i ∈U={(F e,i , F b,i ): 0≤F e,i ≤F max , F min ≤F b,i ≤0} (14)
[0120] In the embodiment of the present application, the fuel consumption model is related to the total fuel consumption of the vehicles in the fleet, and the fuel consumption of the vehicle is related to the engine speed and torque of the vehicle. Therefore, the fuel consumption model of the vehicle can refer to the following formula:
[0121]
[0122] Among them, b1, b2, b3, b4, b5, and b6 are preset fuel consumption parameters, which can be obtained through the minimum quadratic method of the fuel consumption map.
[0123] Since the driving force, braking force and speed of the target vehicle can be obtained according to the torque and engine speed, the fuel consumption model can be simplified as the following formula:
[0124] f fuel,i =g i (x i ,u i ) (16)
[0125] Step 502: acquiring a spatial discrete model according to the dynamics model, and determining a global optimization model according to the fuel consumption model and the spatial discrete model.
[0126] In the embodiment of the present application, the spatial discrete model is a model obtained by discretizing the dynamic model in space using the explicit Euler method, and the spatial discrete model can be recorded as:
[0127] x i,z+1 =f i,Δs (x i,z ,u i,z ) (17)
[0128] Among them, z is the discrete control variable and the discrete space is Δs.
[0129] In a possible implementation, the global optimization model may refer to the following formula:
[0130]
[0131] Constraints:
[0132]
[0133]
[0134]
[0135] in, is the optimal control input sequence for the fleet, N F The total time limit for completing a preset path, k represents a time stage, and S represents the end point of the road.
[0136] However, it is difficult to solve the global optimization model. Therefore, in the embodiment of the present application, the global optimization model is converted. That is, the global optimization model in the embodiment of the present application can be obtained according to the spatial discrete model and the fuel consumption model:
[0137]
[0138] Among them, N DP Represents the total step length, the total distance of the preset path is S d , total step length N DP =S d / Δs, w1 and w2 are the weight coefficients corresponding to fuel consumption and longitudinal speed respectively, and their sum is 1.
[0139] In a possible implementation, based on the global optimization model and driving information obtained above, the process of obtaining the initial control amount of the target vehicle may include the following steps:
[0140] First, the server solves the global optimization model to obtain the global optimization speed.
[0141] According to the driving information, the global optimization model is solved to obtain the global optimization speed, which includes the speeds corresponding to multiple steps in the preset path. The specific formula can be referred to as follows:
[0142]
[0143] Then, the server can determine the static local model based on the driving information, the global optimization speed, and the time discrete model.
[0144]
[0145] Among them, Q, R and ρ are weights, ε>0 is the preset relaxation factor, N m is the prediction step size, is the reference state quantity, the reference speed included in the reference state quantity is the global optimization speed, and the longitudinal displacement corresponding to the global optimization speed can be obtained according to the global optimization speed, f safe (x i-1,k|t , x i,k|t ) is the vehicle safety distance constraint condition. By solving the static local optimization model, the server can obtain the optimized control quantity and state quantity with the smallest difference between the reference state quantity and the reference control quantity.
[0146] The optimized control amount can refer to the following formula:
[0147]
[0148] The state quantity corresponding to the optimized control quantity can refer to the following formula:
[0149]
[0150] Finally, the server solves the static local optimization model to obtain an initial control quantity, which is the above-mentioned optimized control quantity.
[0151] The operation of the target vehicle is controlled based on the initial control amount. In the embodiment of the present application, the server may control the driving of the target vehicle according to the first control amount in the control amount sequence corresponding to the initial control amount.
[0152] In an embodiment of the present application, the server can track and record the state quantity corresponding to each time point of the target vehicle within a step stage and the fuel consumption corresponding to each state quantity as the initial trajectory corresponding to the initial control quantity during the driving process of the target vehicle.
[0153] The state quantities included in the initial trajectory can refer to the following formula:
[0154]
[0155] The fuel consumption included in the initial trajectory can be referred to the following formula:
[0156]
[0157] In one embodiment, based on the above Figure 5 The embodiment shown, see Figure 6 This embodiment relates to a process of obtaining an initial local optimization model of a target vehicle based on an initial driving trajectory. Figure 6As shown, step 203 may include step 601 and step 602 .
[0158] Step 601, obtaining a time discrete model according to a dynamic model.
[0159] In the embodiment of the present application, the time discrete model is a model obtained by discretizing the dynamic model in time using the explicit Euler method, and the time discrete model can be expressed as:
[0160]
[0161] Among them, Δt is the discrete time and k is the discrete time variable.
[0162] Step 602: Obtain an initial local optimization model according to the time discrete model, the fuel consumption model and the initial driving trajectory.
[0163] In an embodiment of the present application, the server may use the initial driving trajectory as the data used in the first iterative process. After the server obtains the initial local optimization model based on the initial driving trajectory, the time discrete model and the fuel consumption model, the server may solve the initial local optimization model to obtain the driving control strategy corresponding to the first iterative process. The driving control strategy includes the control amount corresponding to the first iterative process. After using the driving control strategy to control the driving of the target vehicle, the server can obtain the driving trajectory corresponding to the next iterative process.
[0164] Therefore, each iteration process corresponds to a driving trajectory, and the initial state corresponding to the qth iteration process can be expressed as the following formula:
[0165]
[0166] The initial fuel consumption corresponding to the initial state quantity can be expressed as the following formula:
[0167]
[0168] Among them, q represents the qth iteration process.
[0169] In one possible implementation, the initial driving trajectory includes an initial state quantity and an initial fuel consumption corresponding to the initial state quantity. In this embodiment of the present application, the server can obtain a terminal constraint set corresponding to the initial state quantity based on the initial state quantity, and obtain a terminal fuel consumption function corresponding to the initial fuel consumption based on the initial fuel consumption. Finally, the initial local optimization model is obtained based on the time discrete model, the fuel consumption model, the approximate terminal constraint set corresponding to the terminal constraint set, and the approximate terminal fuel consumption function corresponding to the terminal fuel consumption function.
[0170] The terminal constraint set corresponding to the initial state quantity can be defined as the following formula:
[0171]
[0172] Among them, the state quantity of the terminal converges to X i,F is the state quantity of the terminal.
[0173] In a possible implementation, the amount of data in the terminal constraint set is huge. If it is not processed, it will lead to excessive calculation. Therefore, the terminal constraint set can be approximated, referring to the following formula:
[0174]
[0175] Where C = [0, 1], Nf(t <N f ≤N F ) is the constraint set step size, so that the approximate set of the terminal constraint set only includes part of the data The amount of calculation is reduced and the calculation efficiency is improved.
[0176] In the embodiment of the present application, the server can obtain the fuel consumption function according to the initial fuel consumption:
[0177]
[0178] because is the fuel consumption corresponding to the kth moment in the qth iteration. Therefore, the fuel consumption function can be set as follows:
[0179]
[0180] That is, the fuel consumption of the target vehicle is 0 when it is at the destination, and the fuel consumption is greater than 0 at other times. The fuel consumption function represents the sum of the fuel consumption of the target vehicle when it reaches the destination at time t.
[0181] Based on the above fuel consumption function, the server can define the terminal fuel consumption function as:
[0182]
[0183] in, Represents each state x in the historical trajectory i The number of iterations and the set of time index values can be expressed as the following formula:
[0184]
[0185] Based on this, we can obtain the historical trajectory and x i Equal state quantity iteration number and time index value.
[0186] At this point, the fuel consumption function and the set can be approximated to obtain an approximate terminal fuel consumption function:
[0187]
[0188] The approximate terminal fuel consumption function can be solved as an optimization problem in the title time domain, thereby constraining the fuel consumption of the driving control strategy.
[0189] In summary, the server can obtain the initial local optimization model corresponding to the t-time stage of the q-th iteration according to the time discrete model, the fuel consumption model, the approximate terminal constraint set corresponding to the terminal constraint set, and the approximate terminal fuel consumption function corresponding to the terminal fuel consumption function:
[0190]
[0191] By solving the initial local optimization model, the optimal control quantity corresponding to time t of the qth iteration can be obtained.
[0192] In one embodiment, based on the above Figure 6 The embodiment shown, see Figure 7 This embodiment involves a process of iteratively optimizing the driving control strategy of the target vehicle based on the initial local optimization model. Figure 7 As shown, step 203 may include steps 701 to 703 .
[0193] Step 701, solving the initial local optimization model to obtain the optimized control quantity of the target vehicle.
[0194] Step 702: Control the target vehicle to continue driving according to the optimized control amount, and obtain the intermediate driving trajectory of the target vehicle during the process of the target vehicle continuing to drive.
[0195] Step 703, according to the intermediate driving trajectory, an intermediate local optimization model of the target vehicle is obtained, and according to the intermediate local optimization model, the driving control strategy of the target vehicle is continuously optimized iteratively until the target vehicle finishes driving in the current target road section.
[0196] In an embodiment of the present application, for an iterative process, the server can obtain an optimized control quantity sequence and a predicted state quantity sequence corresponding to the optimized control quantity sequence by solving the initial local optimization model at time t of the qth iteration.
[0197] Among them, the optimized control quantity sequence can refer to the following formula:
[0198]
[0199] The predicted state quantity sequence corresponding to the optimized control quantity sequence can refer to the following formula:
[0200]
[0201] The server may optimize the control quantity sequence including the first control quantity As the optimal control quantity corresponding to time t.
[0202] After obtaining the optimized control amount, the server can control the target vehicle to continue driving according to the optimized control amount, and obtain the intermediate driving trajectory of the target vehicle during the target vehicle's continued driving.
[0203] At this time, the server may use the intermediate driving trajectory as data for constructing an intermediate local optimization model at time t+1 of the qth iteration to construct an intermediate local optimization model.
[0204] Repeat this process until the target section ends, and the server can obtain the optimized control amount for each time period corresponding to the target section, the intermediate driving trajectory corresponding to each optimized control amount, and the target local optimization model after the iteration.
[0205] At this point, the server can determine whether the target local optimization model has converged after the iteration process.
[0206] In a possible implementation, when the target vehicle finishes traveling in the current target road section, the server may determine whether the target local optimization model obtained by the last iterative optimization of the target vehicle in the target road section has converged.
[0207] Regarding the method for the server to determine whether the target local optimization model has converged, optionally, the server can calculate the number of times the target vehicle has traveled on the target section. If the number reaches a travel number threshold, the server determines that the target local optimization model has converged; optionally, the server can obtain historical driving data of the target vehicle on the target section, and the historical driving data includes the optimization control amount of each time period obtained during each historical driving process, and the intermediate driving trajectory corresponding to each optimization control amount. The server can detect the difference between each intermediate driving trajectory included in each historical driving data and the intermediate driving trajectory obtained after the current driving is completed. The server can set a difference threshold. If the difference is less than the difference threshold, the server determines that the target local optimization model has converged, or, if the number of times the difference is less than the difference threshold is greater than a preset number, for example 10 times, that is, the difference in the intermediate driving trajectories obtained by the target vehicle traveling on the target section 10 times is less than the difference threshold, then the server determines that the target local optimization model has converged.
[0208] If the target local optimization model converges, when the target vehicle re-travels the target road section, the driving control strategy of the target vehicle is obtained according to the target local optimization model.
[0209] If the target local optimization model has not converged, when the target vehicle re-drives the target section, the local optimization model of the target vehicle is reconstructed according to the target driving trajectory corresponding to the target local optimization model, and the driving control strategy of the target vehicle is obtained according to the reconstructed local optimization model.
[0210] In one embodiment, referring to Figure 8 , provides an exemplary iterative learning predictive control method for energy-saving driving of intelligent networked fleets, which can be applied to Figure 1 In the implementation environment shown.
[0211] Step 801, obtaining the fuel consumption model and dynamic model of the target vehicle.
[0212] Among them, the fuel consumption model is related to the total fuel consumption of the vehicles in the fleet.
[0213] Step 802: acquiring a spatial discrete model according to the dynamics model, and determining a global optimization model according to the fuel consumption model and the spatial discrete model.
[0214] Step 803, obtaining current driving information.
[0215] Step 804, solving the global optimization model to obtain the global optimization speed.
[0216] Step 805: Determine a static local model according to the driving information, the global optimization speed and the time discrete model.
[0217] Step 806, solving the static local optimization model to obtain the initial control quantity.
[0218] The initial control quantity includes a control quantity sequence.
[0219] Step 807, controlling the target vehicle to travel according to the first control variable in the control variable sequence.
[0220] Step 808: during the driving process of the target vehicle, an initial driving trajectory of the target vehicle is obtained.
[0221] The initial driving trajectory includes an initial state quantity and an initial fuel consumption corresponding to the initial state quantity.
[0222] Step 809: Obtain a time discrete model according to the dynamics model.
[0223] Step 810, obtaining a terminal constraint set corresponding to the initial state quantity according to the initial state quantity.
[0224] Step 811, obtaining a terminal fuel consumption function corresponding to the initial fuel consumption according to the initial fuel consumption.
[0225] Step 812, obtaining an initial local optimization model according to the time discrete model, the fuel consumption model, the approximate terminal constraint set corresponding to the terminal constraint set, and the approximate terminal fuel consumption function corresponding to the terminal fuel consumption function.
[0226] Step 813, solving the initial local optimization model to obtain the optimized control quantity of the target vehicle.
[0227] Step 814, controlling the target vehicle to continue driving according to the optimized control amount, and obtaining the intermediate driving trajectory of the target vehicle during the process of the target vehicle continuing to drive.
[0228] Step 815, according to the intermediate driving trajectory, obtain the intermediate local optimization model of the target vehicle, and according to the intermediate local optimization model, continue to iteratively optimize the driving control strategy of the target vehicle until the target vehicle finishes driving in the current target road section.
[0229] Step 816, when the target vehicle has finished traveling in the current target road section, it is determined whether the target local optimization model obtained by the last iterative optimization of the target vehicle in the target road section has converged.
[0230] Step 817: If the target local optimization model converges, when the target vehicle re-travels the target road section, the driving control strategy of the target vehicle is obtained according to the target local optimization model.
[0231] Step 818, if the target local optimization model has not converged, then when the target vehicle re-drives the target section, the local optimization model of the target vehicle is reconstructed according to the target driving trajectory corresponding to the target local optimization model, and the driving control strategy of the target vehicle is obtained according to the reconstructed local optimization model.
[0232] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.
[0233] Based on the same inventive concept, the embodiment of the present application also provides an intelligent networked fleet energy-saving driving iterative learning prediction control device for implementing the above-mentioned intelligent networked fleet energy-saving driving iterative learning prediction control method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the intelligent networked fleet energy-saving driving iterative learning prediction control device provided below can refer to the limitations of the intelligent networked fleet energy-saving driving iterative learning prediction control method above, and will not be repeated here.
[0234] In an exemplary embodiment, Fig. 9 As shown, a device for iterative learning prediction control of energy-saving driving of an intelligent network-connected vehicle fleet is provided, comprising: a first acquisition module 901, a second acquisition module 902 and a control module 903, wherein:
[0235] The first acquisition module 901 is used to acquire the global optimization model and current driving information of the target vehicle in the fleet, and acquire the initial control amount of the target vehicle according to the driving information and the global optimization model;
[0236] The second acquisition module 902 is used to control the target vehicle to travel according to the initial control amount, and to obtain the initial travel trajectory of the target vehicle during the travel of the target vehicle;
[0237] The control module 903 is used to obtain an initial local optimization model of the target vehicle according to the initial driving trajectory, and iteratively optimize the driving control strategy of the target vehicle according to the initial local optimization model.
[0238] In one embodiment, the first acquisition module 901 includes:
[0239] A fuel consumption model and dynamic model acquisition unit, used to acquire the fuel consumption model and dynamic model of the target vehicle, wherein the fuel consumption model is related to the total fuel consumption of the vehicles in the fleet;
[0240] The global determination unit is used to obtain a spatial discrete model according to a dynamic model, and to determine a global optimization model according to a fuel consumption model and the spatial discrete model.
[0241] In one embodiment, the second acquisition module 902 includes:
[0242] The control initial control unit is used to control the driving of the target vehicle according to the first control quantity in the control quantity sequence.
[0243] In one embodiment, the control module 903 includes:
[0244] A time discrete unit is used to obtain a time discrete model according to a dynamic model;
[0245] The local model acquisition unit is used to acquire an initial local optimization model according to the time discrete model, the fuel consumption model and the initial driving trajectory.
[0246] In one embodiment, the initial driving trajectory includes an initial state quantity and an initial fuel consumption corresponding to the initial state quantity, and the local model acquisition unit is further used to perform the following steps:
[0247] Obtain the terminal constraint set corresponding to the initial state quantity according to the initial state quantity;
[0248] Obtain a terminal fuel consumption function corresponding to the initial fuel consumption according to the initial fuel consumption;
[0249] An initial local optimization model is obtained according to the time discrete model, the fuel consumption model, the approximate terminal constraint set corresponding to the terminal constraint set, and the approximate terminal fuel consumption function corresponding to the terminal fuel consumption function.
[0250] In one embodiment, the second acquisition module 902 includes:
[0251] A global solving unit, used to solve the global optimization model and obtain the global optimization speed;
[0252] A static local determination unit, used for determining a static local model according to driving information, a global optimization speed and a time discrete model;
[0253] The initial control quantity acquisition unit is used to solve the static local optimization model to obtain the initial control quantity.
[0254] In one embodiment, the control module 903 includes:
[0255] An optimized control quantity acquisition unit is used to solve the initial local optimization model to obtain the optimized control quantity of the target vehicle;
[0256] An intermediate control unit, used to control the target vehicle to continue to travel according to the optimized control amount, and to obtain an intermediate travel trajectory of the target vehicle during the process of the target vehicle continuing to travel;
[0257] The iterative optimization unit is used to obtain an intermediate local optimization model of the target vehicle according to the intermediate driving trajectory, and continue to iteratively optimize the driving control strategy of the target vehicle according to the intermediate local optimization model until the target vehicle completes driving in the current target section.
[0258] In one embodiment, the control module 903 further includes:
[0259] A judgment unit, used to judge whether the target local optimization model obtained by the last iterative optimization of the target vehicle in the target section has converged when the target vehicle has finished traveling in the current target section;
[0260] A convergence unit, for obtaining a driving control strategy of the target vehicle according to the target local optimization model when the target vehicle re-drives the target road section if the target local optimization model converges;
[0261] The non-convergence unit is used to reconstruct the local optimization model of the target vehicle according to the target driving trajectory corresponding to the target local optimization model when the target vehicle re-drives the target section if the target local optimization model has not converged, and obtain the driving control strategy of the target vehicle according to the reconstructed local optimization model.
[0262] Each module in the above-mentioned intelligent networked fleet energy-saving driving iterative learning prediction control device can be implemented in whole or in part by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.
[0263] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig.10 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store iterative learning predictive control data for energy-saving driving of an intelligent networked fleet. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for iterative learning predictive control of energy-saving driving of an intelligent networked fleet is implemented.
[0264] Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0265] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:
[0266] Acquire a global optimization model and current driving information of a target vehicle in the fleet, and acquire an initial control amount of the target vehicle according to the driving information and the global optimization model;
[0267] Controlling the target vehicle to travel according to the initial control amount, and obtaining an initial travel trajectory of the target vehicle during the travel of the target vehicle;
[0268] An initial local optimization model of the target vehicle is obtained according to the initial driving trajectory, and a driving control strategy of the target vehicle is iteratively optimized according to the initial local optimization model.
[0269] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0270] Obtaining a fuel consumption model and a dynamic model of the target vehicle, wherein the fuel consumption model is related to the total fuel consumption of the vehicles in the fleet;
[0271] According to the dynamic model, a spatial discrete model is obtained, and according to the fuel consumption model and the spatial discrete model, the global optimization model is determined.
[0272] In one embodiment, the initial control amount includes a control amount sequence, and the processor further implements the following steps when executing the computer program:
[0273] The target vehicle is controlled to travel according to a first control variable in the control variable sequence.
[0274] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0275] According to the dynamic model, a time discrete model is obtained;
[0276] The initial local optimization model is acquired according to the time discrete model, the fuel consumption model and the initial driving trajectory.
[0277] In one embodiment, the initial driving trajectory includes an initial state quantity and an initial fuel consumption corresponding to the initial state quantity, and the processor further implements the following steps when executing the computer program:
[0278] Acquire a terminal constraint set corresponding to the initial state quantity according to the initial state quantity;
[0279] Acquire a terminal fuel consumption function corresponding to the initial fuel consumption according to the initial fuel consumption;
[0280] The initial local optimization model is obtained according to the time discrete model, the fuel consumption model, the approximate terminal constraint set corresponding to the terminal constraint set, and the approximate terminal fuel consumption function corresponding to the terminal fuel consumption function.
[0281] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0282] Solving the global optimization model to obtain a global optimization speed;
[0283] Determining a static local model according to the driving information, the global optimization speed, and the time discrete model;
[0284] The static local optimization model is solved to obtain the initial control variable.
[0285] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0286] Solving the initial local optimization model to obtain an optimized control amount of the target vehicle;
[0287] According to the optimized control amount, controlling the target vehicle to continue to travel, and obtaining an intermediate travel trajectory of the target vehicle during the process of the target vehicle continuing to travel;
[0288] According to the intermediate driving trajectory, an intermediate local optimization model of the target vehicle is obtained, and according to the intermediate local optimization model, the driving control strategy of the target vehicle is continuously iteratively optimized until the target vehicle finishes driving in the current target road section.
[0289] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0290] When the target vehicle has finished traveling in the current target road section, determining whether the target local optimization model obtained by the last iterative optimization of the target vehicle in the target road section has converged;
[0291] If the target local optimization model converges, when the target vehicle re-travels the target road section, a driving control strategy of the target vehicle is obtained according to the target local optimization model;
[0292] If the target local optimization model has not converged, when the target vehicle re-travels the target road section, the local optimization model of the target vehicle is reconstructed according to the target driving trajectory corresponding to the target local optimization model, and the driving control strategy of the target vehicle is obtained according to the reconstructed local optimization model.
[0293] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:
[0294] Acquire a global optimization model and current driving information of a target vehicle in the fleet, and acquire an initial control amount of the target vehicle according to the driving information and the global optimization model;
[0295] Controlling the target vehicle to travel according to the initial control amount, and obtaining an initial travel trajectory of the target vehicle during the travel of the target vehicle;
[0296] An initial local optimization model of the target vehicle is obtained according to the initial driving trajectory, and a driving control strategy of the target vehicle is iteratively optimized according to the initial local optimization model.
[0297] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0298] Obtaining a fuel consumption model and a dynamic model of the target vehicle, wherein the fuel consumption model is related to the total fuel consumption of the vehicles in the fleet;
[0299] According to the dynamic model, a spatial discrete model is obtained, and according to the fuel consumption model and the spatial discrete model, the global optimization model is determined.
[0300] In one embodiment, the initial control amount includes a control amount sequence, and when the computer program is executed by the processor, the following steps are further implemented:
[0301] The target vehicle is controlled to travel according to a first control variable in the control variable sequence.
[0302] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0303] According to the dynamic model, a time discrete model is obtained;
[0304] The initial local optimization model is acquired according to the time discrete model, the fuel consumption model and the initial driving trajectory.
[0305] In one embodiment, the initial driving trajectory includes an initial state quantity and an initial fuel consumption corresponding to the initial state quantity, and when the computer program is executed by the processor, the following steps are further implemented:
[0306] Acquire a terminal constraint set corresponding to the initial state quantity according to the initial state quantity;
[0307] Acquire a terminal fuel consumption function corresponding to the initial fuel consumption according to the initial fuel consumption;
[0308] The initial local optimization model is obtained according to the time discrete model, the fuel consumption model, the approximate terminal constraint set corresponding to the terminal constraint set, and the approximate terminal fuel consumption function corresponding to the terminal fuel consumption function.
[0309] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0310] Solving the global optimization model to obtain a global optimization speed;
[0311] Determining a static local model according to the driving information, the global optimization speed, and the time discrete model;
[0312] The static local optimization model is solved to obtain the initial control variable.
[0313] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0314] Solving the initial local optimization model to obtain an optimized control amount of the target vehicle;
[0315] According to the optimized control amount, controlling the target vehicle to continue to travel, and obtaining an intermediate travel trajectory of the target vehicle during the process of the target vehicle continuing to travel;
[0316] According to the intermediate driving trajectory, an intermediate local optimization model of the target vehicle is obtained, and according to the intermediate local optimization model, the driving control strategy of the target vehicle is continuously iteratively optimized until the target vehicle finishes driving in the current target road section.
[0317] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0318] When the target vehicle has finished traveling in the current target road section, determining whether the target local optimization model obtained by the last iterative optimization of the target vehicle in the target road section has converged;
[0319] If the target local optimization model converges, when the target vehicle re-travels the target road section, a driving control strategy of the target vehicle is obtained according to the target local optimization model;
[0320] If the target local optimization model has not converged, when the target vehicle re-travels the target road section, the local optimization model of the target vehicle is reconstructed according to the target driving trajectory corresponding to the target local optimization model, and the driving control strategy of the target vehicle is obtained according to the reconstructed local optimization model.
[0321] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:
[0322] Acquire a global optimization model and current driving information of a target vehicle in the fleet, and acquire an initial control amount of the target vehicle according to the driving information and the global optimization model;
[0323] Controlling the target vehicle to travel according to the initial control amount, and obtaining an initial travel trajectory of the target vehicle during the travel of the target vehicle;
[0324] An initial local optimization model of the target vehicle is obtained according to the initial driving trajectory, and a driving control strategy of the target vehicle is iteratively optimized according to the initial local optimization model.
[0325] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0326] Obtaining a fuel consumption model and a dynamic model of the target vehicle, wherein the fuel consumption model is related to the total fuel consumption of the vehicles in the fleet;
[0327] According to the dynamic model, a spatial discrete model is obtained, and according to the fuel consumption model and the spatial discrete model, the global optimization model is determined.
[0328] In one embodiment, the initial control amount includes a control amount sequence, and when the computer program is executed by the processor, the following steps are further implemented:
[0329] The target vehicle is controlled to travel according to a first control variable in the control variable sequence.
[0330] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0331] According to the dynamic model, a time discrete model is obtained;
[0332] The initial local optimization model is acquired according to the time discrete model, the fuel consumption model and the initial driving trajectory.
[0333] In one embodiment, the initial driving trajectory includes an initial state quantity and an initial fuel consumption corresponding to the initial state quantity, and when the computer program is executed by the processor, the following steps are further implemented:
[0334] Acquire a terminal constraint set corresponding to the initial state quantity according to the initial state quantity;
[0335] Acquire a terminal fuel consumption function corresponding to the initial fuel consumption according to the initial fuel consumption;
[0336] The initial local optimization model is obtained according to the time discrete model, the fuel consumption model, the approximate terminal constraint set corresponding to the terminal constraint set, and the approximate terminal fuel consumption function corresponding to the terminal fuel consumption function.
[0337] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0338] Solving the global optimization model to obtain a global optimization speed;
[0339] Determining a static local model according to the driving information, the global optimization speed, and the time discrete model;
[0340] The static local optimization model is solved to obtain the initial control variable.
[0341] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0342] Solving the initial local optimization model to obtain an optimized control amount of the target vehicle;
[0343] According to the optimized control amount, controlling the target vehicle to continue to travel, and obtaining an intermediate travel trajectory of the target vehicle during the process of the target vehicle continuing to travel;
[0344] According to the intermediate driving trajectory, an intermediate local optimization model of the target vehicle is obtained, and according to the intermediate local optimization model, the driving control strategy of the target vehicle is continuously iteratively optimized until the target vehicle finishes driving in the current target road section.
[0345] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:
[0346] When the target vehicle has finished traveling in the current target road section, determining whether the target local optimization model obtained by the last iterative optimization of the target vehicle in the target road section has converged;
[0347] If the target local optimization model converges, when the target vehicle re-travels the target road section, a driving control strategy of the target vehicle is obtained according to the target local optimization model;
[0348] If the target local optimization model has not converged, when the target vehicle re-travels the target road section, the local optimization model of the target vehicle is reconstructed according to the target driving trajectory corresponding to the target local optimization model, and the driving control strategy of the target vehicle is obtained according to the reconstructed local optimization model.
[0349] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.
[0350] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited thereto.
[0351] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.
[0352] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. An iterative learning predictive control method for energy-saving driving of an intelligent networked fleet, characterized in that: The method comprises: Acquire a global optimization model and current driving information of a target vehicle in the fleet, and acquire an initial control amount of the target vehicle according to the driving information and the global optimization model; Controlling the target vehicle to travel according to the initial control amount, and obtaining an initial travel trajectory of the target vehicle during the travel of the target vehicle; An initial local optimization model of the target vehicle is obtained according to the initial driving trajectory, and a driving control strategy of the target vehicle is iteratively optimized according to the initial local optimization model.
2. The method according to claim 1, characterized in that The step of obtaining a global optimization model of a target vehicle in a fleet includes: Obtaining a fuel consumption model and a dynamic model of the target vehicle, wherein the fuel consumption model is related to the total fuel consumption of the vehicles in the fleet; According to the dynamic model, a spatial discrete model is obtained, and according to the fuel consumption model and the spatial discrete model, the global optimization model is determined.
3. The method according to claim 2, characterized in that The step of obtaining an initial local optimization model of the target vehicle according to the initial driving trajectory includes: According to the dynamic model, a time discrete model is obtained; The initial local optimization model is acquired according to the time discrete model, the fuel consumption model and the initial driving trajectory.
4. The method according to claim 3, characterized in that The initial driving trajectory includes an initial state quantity and an initial fuel consumption corresponding to the initial state quantity. The initial local optimization model is obtained according to the time discrete model, the fuel consumption model and the initial driving trajectory, including: Acquire a terminal constraint set corresponding to the initial state quantity according to the initial state quantity; Acquire a terminal fuel consumption function corresponding to the initial fuel consumption according to the initial fuel consumption; The initial local optimization model is obtained according to the time discrete model, the fuel consumption model, the approximate terminal constraint set corresponding to the terminal constraint set, and the approximate terminal fuel consumption function corresponding to the terminal fuel consumption function.
5. The method according to claim 3, characterized in that: The obtaining of the initial control amount of the target vehicle according to the driving information and the global optimization model includes: Solving the global optimization model to obtain a global optimization speed; Determining a static local model according to the driving information, the global optimization speed, and the time discrete model; The static local optimization model is solved to obtain the initial control variable.
6. The method according to claim 1, characterized in that The iterative optimization of the driving control strategy of the target vehicle according to the initial local optimization model includes: Solving the initial local optimization model to obtain an optimized control amount of the target vehicle; According to the optimized control amount, controlling the target vehicle to continue to travel, and obtaining an intermediate travel trajectory of the target vehicle during the process of the target vehicle continuing to travel; According to the intermediate driving trajectory, an intermediate local optimization model of the target vehicle is obtained, and according to the intermediate local optimization model, the driving control strategy of the target vehicle is continuously iteratively optimized until the target vehicle finishes driving in the current target road section.
7. The method according to claim 6, characterized in that The method further comprises: When the target vehicle has finished traveling in the current target road section, determining whether the target local optimization model obtained by the last iterative optimization of the target vehicle in the target road section has converged; If the target local optimization model converges, when the target vehicle re-travels the target road section, a driving control strategy of the target vehicle is obtained according to the target local optimization model; If the target local optimization model has not converged, when the target vehicle re-travels the target road section, the local optimization model of the target vehicle is reconstructed according to the target driving trajectory corresponding to the target local optimization model, and the driving control strategy of the target vehicle is obtained according to the reconstructed local optimization model.
8. An iterative learning prediction control device for energy-saving driving of an intelligent networked fleet, characterized in that: The device comprises: A first acquisition module is used to acquire a global optimization model and current driving information of a target vehicle in a fleet, and acquire an initial control amount of the target vehicle according to the driving information and the global optimization model; A second acquisition module, used for controlling the target vehicle to travel according to the initial control amount, and obtaining an initial travel trajectory of the target vehicle during the travel of the target vehicle; A control module is used to obtain an initial local optimization model of the target vehicle according to the initial driving trajectory, and iteratively optimize the driving control strategy of the target vehicle according to the initial local optimization model.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.
