A vehicle-cloud cooperative intelligent logistics vehicle driving speed planning method and system considering a task time window

Abstract

The application discloses a kind of task time window's car cloud coordination intelligent logistics car driving speed planning method and system, method includes: based on intersection queuing dissipation time prediction and road speed limit, calculate the feasible time region of vehicle through each signal light intersection, and extract each intersection green light passing time window set in combination with signal timing;With total energy consumption minimization and total travel time minimization as target, with distribution time window, green light passing time window, road speed limit and vehicle dynamics as constraint to establish speed planning model, nonlinear model is converted into mixed integer linear programming model by variable substitution and energy consumption model fitting;According to the urgency degree of task time window and traffic flow resistance coefficient, determine the priority of optimization target, and solve using the rolling optimization mechanism triggered by time-space hybrid, output each road section speed planning result.The application is used for the speed automatic planning of urban electric logistics car, can reduce driving energy consumption and improve logistics distribution timeliness.

Description

Technical Field

[0001] This invention belongs to the field of intelligent driving technology, and in particular relates to a method and system for planning the driving speed of a vehicle-cloud collaborative intelligent logistics vehicle that takes into account the task time window. Background Technology

[0002] In the complex urban logistics and transportation environment, road intersection conditions and traffic lights are constantly changing. Maintaining transportation efficiency while optimizing energy consumption is a significant challenge. This paper explores an adaptive speed planning method for urban logistics vehicles that considers task time windows, aiming to reduce energy consumption. By calculating and evaluating green light passage windows and vehicle speed in real time, this method jointly optimizes green light passage time window selection and vehicle speed to maximize transportation efficiency. This research is urgently needed and requires further development.

[0003] Existing vehicle speed planning primarily focuses on either energy consumption optimization or driving efficiency, failing to comprehensively consider the balance between the two. For example, existing technology discloses a method for intelligent driving-human co-driving mode switching in urban logistics vehicles that considers driving risk and energy consumption. This method includes extracting driving data for driver mode switching, obtaining the mode switching interval considering driver style and weekly vehicle trajectory prediction, and optimizing the mode switching timing based on driving risk and maximum energy consumption. This method uses historical characteristic data of the driver mode switching process and weekly vehicle trajectory prediction, combined with style recognition results, to obtain the risk interval and risk change curve for mode switching decisions using a risk assessment model. Based on these two factors, it obtains the time interval for mode switching decisions and finally establishes an optimization model based on human-driving mode speed prediction to optimize the switching timing. This method comprehensively considers driving risk and driving energy consumption, effectively solving the problems of high driving energy consumption and insufficient safety caused by unreasonable human-machine co-driving mode switching in complex urban scenarios for intelligent driving logistics vehicles.

[0004] However, due to the complex traffic conditions at urban intersections, the real-time changes in traffic lights, and the fact that the aforementioned driving decisions do not consider transportation efficiency, this target mode switching decision suffers from inefficiency. On one hand, urban logistics delivery tasks have clear arrival time requirements. Existing speed planning methods only focus on road traffic and driving energy consumption, failing to incorporate the time window constraint of the delivery task into the speed planning process, making it difficult to meet the timeliness requirements of logistics delivery. On the other hand, existing methods use a phased speed planning approach at intersections with continuous traffic lights, failing to uniformly model the selection of green light windows at multiple intersections with vehicle speed trajectories, easily getting trapped in local optima and failing to maximize overall traffic efficiency. Furthermore, the optimization objectives of existing methods typically use fixed weight allocations or fixed priority rankings, failing to adaptively adjust according to the urgency of the task and the dynamic changes in road traffic conditions, making it difficult to achieve the optimal balance between delivery timeliness and driving energy consumption. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention proposes a vehicle-cloud collaborative intelligent logistics vehicle speed planning method and system that considers task time windows, thereby resolving the issues existing in the prior art.

[0006] Firstly, to achieve the above objectives, the present invention provides a vehicle-cloud collaborative intelligent logistics vehicle speed planning method considering task time windows, comprising the following steps: The cloud platform obtains signal timing parameters at each traffic light intersection along the vehicle's route, as well as road speed limits and traffic flow information for each road segment. Based on the predicted queue dissipation time at intersections and the speed limits of each road segment, the cloud calculates the feasible time range for vehicles to pass through each traffic light intersection, and extracts the set of green light passage time windows for each traffic light intersection from the feasible time range according to the signal timing parameters of each traffic light intersection. With the optimization objectives of minimizing total vehicle energy consumption and total driving time, and with constraints such as time windows at each delivery point, green light time windows at each traffic light intersection, road speed limits, and vehicle dynamics, a multi-objective vehicle speed planning model is established. Discrete decision variables for selecting green light passage time windows are introduced into the multi-objective vehicle speed planning model, and surface fitting and variable substitution are performed on the vehicle energy consumption term to transform the multi-objective vehicle speed planning model into a mixed integer linear programming model with objective priority ranking. The time urgency index is calculated based on the remaining time and distance of the current delivery task, and the traffic flow impedance coefficient is calculated based on the traffic flow status of the road ahead. The priority ranking of each optimization objective in the mixed integer linear programming model is determined based on the combination of the values ​​of the time urgency index and the traffic flow impedance coefficient. The mixed-integer linear programming model is solved using a rolling optimization method. The model is resolved based on the vehicle position and speed status uploaded by the vehicle terminal, and the speed planning results for each road segment are output and sent to the vehicle terminal for execution.

[0007] Optionally, the process of calculating the feasible time range for a vehicle to pass through each traffic light intersection based on the predicted queue dissipation time and the road speed limits of each road segment includes: using an intelligent driver model to predict the acceleration of each vehicle in front of it in the current intersection queue; recursively calculating the queue length change curve over time based on the predicted acceleration values ​​of each vehicle in front; and taking the moment when the queue length is reduced to zero as the queue dissipation time; and calculating the shortest and longest passage times for the vehicle to reach the stop line of the target intersection based on the distance between the vehicle's current position and the stop line of the target intersection, the maximum and minimum road speed limits, and the set expected acceleration and deceleration, and using the shortest and longest passage times to constitute the feasible time range.

[0008] Optionally, the process of extracting the set of green light passage time windows for each traffic light intersection from the feasible time region based on the signal timing parameters of each traffic light intersection includes: obtaining the start and end times of the green light signal for each traffic light intersection within a complete signal cycle; performing intersection operations on the time intervals of the feasible time region with each green light period within the signal cycle to extract the overlapping parts of the feasible time region and each green light period; and using the set of all overlapping parts as the set of green light passage time windows for the traffic light intersection.

[0009] Optionally, the process of performing surface fitting on the vehicle energy consumption item includes: sampling the motor power expression within a preset speed range and a preset acceleration range, calculating the motor power value corresponding to each sampling point, and constructing a fitting dataset containing speed, acceleration, and motor power; fitting the fitting dataset with a surface function to obtain an approximate analytical expression of motor power with respect to speed and acceleration; and introducing a speed square variable and a speed reciprocal auxiliary variable to replace the original variables in the multi-objective vehicle speed planning model to convert the approximate analytical expression into a linear expression.

[0010] Optionally, the process of converting the multi-objective vehicle speed planning model into a mixed-integer linear programming model includes: configuring a 0-1 type discrete decision variable for each green light passage time window at each traffic light intersection in the multi-objective vehicle speed planning model, and setting constraints to limit that there is one and only one discrete decision variable at the same traffic light intersection with a value of 1; replacing the speed variable in the travel time expression of each road segment with the speed square variable, and introducing a linear approximation to express the travel time; determining whether the corresponding green light passage time window is selected based on the value of each discrete decision variable, and using the boundary value of the selected green light passage time window as the constraint boundary of the arrival time at the corresponding intersection.

[0011] Optionally, the process of determining the priority ranking of each optimization objective in the mixed-integer linear programming model based on the combination of the time urgency index and the traffic flow impedance coefficient includes: comparing the time urgency index with a first threshold and comparing the traffic flow impedance coefficient with a second threshold; when the time urgency index is greater than the first threshold and the traffic flow impedance coefficient is less than the second threshold, setting the priority of the total travel time minimization objective higher than the total travel energy consumption minimization objective; when the time urgency index is greater than the first threshold and the traffic flow impedance coefficient is greater than the second threshold, setting the priority of the total travel energy consumption minimization objective higher than the total travel time minimization objective; when the time urgency index is less than the first threshold and the traffic flow impedance coefficient is less than the second threshold, setting the total travel energy consumption minimization objective and the total travel time minimization objective to have the same priority.

[0012] Optionally, the process of triggering model re-solution based on vehicle position and speed includes: continuously monitoring vehicle position and speed; triggering the first re-solution when the vehicle reaches a preset waypoint; repeatedly triggering re-solution at fixed time intervals when the vehicle has not reached the preset waypoint and the speed is lower than a preset speed threshold; and maintaining the current speed planning result without triggering re-solution when the vehicle has not reached the preset waypoint and the speed is not lower than the preset speed threshold.

[0013] Secondly, the present invention also provides a vehicle-cloud collaborative intelligent logistics vehicle speed planning system that considers task time windows, for implementing a vehicle-cloud collaborative intelligent logistics vehicle speed planning method that considers task time windows, the system comprising: The information acquisition module is used to acquire signal timing parameters at each traffic light intersection along the vehicle's route, as well as road speed limit information and road traffic flow information for each road segment. The passage window generation module is used to calculate the feasible time range for vehicles to pass through each traffic light intersection based on the predicted queue dissipation time of the intersection and the road speed limit of each road segment, and to extract the set of green light passage time windows for each traffic light intersection from the feasible time range according to the signal timing parameters of each traffic light intersection. The model building module is used to establish a multi-objective vehicle speed planning model with the optimization objectives of minimizing total vehicle energy consumption and total driving time, and with constraints such as time windows at each delivery point, green light time windows at each traffic light intersection, road speed limits, and vehicle dynamics. The model transformation module is used to introduce discrete decision variables for selecting the green light passage time window into the multi-objective vehicle speed planning model, and to perform surface fitting and variable substitution on the vehicle energy consumption term, so as to transform the multi-objective vehicle speed planning model into a mixed integer linear programming model with objective priority ranking. The priority decision module is used to calculate the time urgency index based on the remaining time and distance of the current delivery task, and to calculate the traffic flow impedance coefficient based on the traffic flow status of the road ahead. Based on the combination of the values ​​of the time urgency index and the traffic flow impedance coefficient, the priority ranking of each optimization objective in the mixed integer linear programming model is determined. The rolling optimization solution module is used to solve the mixed integer linear programming model using a rolling optimization method. It triggers model re-solution based on vehicle position and speed status, and outputs the speed planning results for each road segment.

[0014] Thirdly, the present invention also provides a computer terminal device, comprising: One or more processors; A memory, coupled to the processor, for storing 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 steps of the vehicle-cloud collaborative intelligent logistics vehicle speed planning method that considers task time windows in the first aspect above.

[0015] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the steps of the vehicle-cloud collaborative intelligent logistics vehicle speed planning method considering task time windows in the first aspect described above.

[0016] Compared with the prior art, the present invention has the following advantages and technical effects: This invention provides a vehicle-cloud collaborative intelligent logistics vehicle speed planning method and system that considers task time windows. Addressing the characteristic of urban logistics delivery tasks with clearly defined arrival time requirements, it introduces task time window constraints from multiple delivery points into the vehicle speed planning process. By comprehensively considering the earliest and latest permitted arrival times at each delivery point, it achieves full-process speed optimization oriented towards the timeliness requirements of logistics delivery, avoiding the problem of existing speed planning methods that only focus on road traffic and ignore the timeliness of delivery tasks. This invention comprehensively considers intersection queue dissipation time, signal timing, road speed limits, and vehicle dynamics constraints, constructing a set of feasible green light passage time windows for continuous signalized intersections. It unifies the green light passage time window selection with vehicle speed trajectory planning, achieving joint optimization of discrete passage window decisions and continuous speed decisions. This avoids the local optimum problem caused by staged planning at each intersection, improving the overall traffic efficiency of vehicles passing through continuous signalized intersections. This invention addresses the problem of dynamic changes in task urgency and road traffic conditions during urban logistics delivery. Based on the task time window urgency index and traffic flow impedance coefficient, it adaptively adjusts the priority of driving time targets and energy consumption targets. Combined with a spatiotemporal hybrid triggering rolling optimization mechanism, it updates vehicle speed trajectory in real time, enabling vehicles to achieve dynamic coordination between delivery timeliness and driving energy consumption according to changes in delivery tasks and traffic environment. Attached Figure Description

[0017] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a schematic diagram of the green light transit time window planning according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the logistics and distribution route according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the spatiotemporal hybrid triggering mechanism according to an embodiment of the present invention; Figure 4 This is a flowchart illustrating a vehicle-cloud collaborative intelligent logistics vehicle speed planning method that considers task time windows, according to an embodiment of the present invention. Detailed Implementation

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0020] Example 1 like Figures 1-4 As shown, this embodiment provides a vehicle-cloud collaborative intelligent logistics vehicle speed planning method that considers task time windows, including: The cloud platform obtains signal timing parameters at each traffic light intersection along the vehicle's route, as well as road speed limits and traffic flow information for each road segment. Based on the predicted queue dissipation time at intersections and the speed limits of each road segment, the cloud calculates the feasible time range for vehicles to pass through each traffic light intersection, and extracts the set of green light passage time windows for each traffic light intersection from the feasible time range according to the signal timing parameters of each traffic light intersection. With the optimization objectives of minimizing total vehicle energy consumption and total driving time, and with constraints such as time windows at each delivery point, green light time windows at each traffic light intersection, road speed limits, and vehicle dynamics, a multi-objective vehicle speed planning model is established. Discrete decision variables for selecting green light passage time windows are introduced into the multi-objective vehicle speed planning model, and surface fitting and variable substitution are performed on the vehicle energy consumption term to transform the multi-objective vehicle speed planning model into a mixed integer linear programming model with objective priority ranking. The time urgency index is calculated based on the remaining time and distance of the current delivery task, and the traffic flow impedance coefficient is calculated based on the traffic flow status of the road ahead. The priority ranking of each optimization objective in the mixed integer linear programming model is determined based on the combination of the values ​​of the time urgency index and the traffic flow impedance coefficient. The mixed-integer linear programming model is solved using a rolling optimization method. The model is resolved based on the vehicle position and speed status uploaded by the vehicle terminal, and the speed planning results for each road segment are output and sent to the vehicle terminal for execution.

[0021] Furthermore, the process of calculating the feasible time range for vehicles to pass through each traffic light intersection based on the predicted queue dissipation time and the road speed limits of each road segment includes: using an intelligent driver model to predict the acceleration of each vehicle in the current intersection queue; recursively calculating the queue length change curve over time based on the predicted acceleration values ​​of each vehicle in front; and taking the moment when the queue length shrinks to zero as the queue dissipation time; and calculating the shortest and longest passage times for vehicles to reach the stop line of the target intersection based on the distance between the vehicle's current position and the stop line of the target intersection, the maximum and minimum road speed limits, and the set expected acceleration and deceleration, and using the shortest and longest passage times to constitute the feasible time range.

[0022] Specifically, the implementation process of this embodiment includes: The first step is to reconstruct the green light passage window constraint for continuous traffic light intersections: based on the predicted queue dissipation time and road speed limits, the feasible time range for vehicles to pass through continuous traffic light intersections is calculated, and then the set of green light passage time windows at each intersection is obtained. ; Intersection queue dissipation time prediction: Number each vehicle in the current intersection queue, starting from the vehicle closest to the stop line, from 1 to... The queue length is equal to the first... The distance from the vehicle to the stop line, i.e. The trajectory of vehicle j can be obtained by predicting its acceleration. IDM is used to predict the acceleration of each vehicle. According to IDM, the acceleration of vehicle j follows the following pattern: (1); (2); in, This is the maximum acceleration; To reduce speed to meet comfort expectations; The desired speed is equal to the maximum permissible speed on the road. The acceleration index; Let be the relative distance between vehicle j and the vehicle in front of it. - ; Let J be the desired space between vehicle J and the vehicle in front of it. The minimum acceptable stationary distance; For vehicles j Safety interval time; and vehicles j and the vehicle in front j -1 speed. Based on the acceleration prediction of the vehicles in the queue, the queue length is: (3); in, For the two cars The initial distance to the stop line; For vehicles Initial velocity. Queue dissipation time. pass estimate; Calculate the feasible time range for a vehicle to pass through a continuously signalized intersection: Assuming that the acceleration of the vehicle in the feasible passage interval is constant, formulas (7) and (8) can be used to predict the passage through the intersection. Time: (4); (5); in, and These are the minimum and maximum transit times, respectively; The average speed of the vehicle traveling along the given route is determined from historical data. and These represent the desired acceleration and deceleration, respectively, to meet driving comfort requirements. If the sampling time is 1 second, then... .

[0023] Furthermore, the process of extracting the set of green light passage time windows for each traffic light intersection from the feasible time region based on the signal timing parameters of each traffic light intersection includes: obtaining the start and end times of the green light signal for each traffic light intersection within a complete signal cycle; performing intersection operations on the time intervals of the feasible time region with each green light period within the signal cycle to extract the overlapping parts of the feasible time region and each green light period; and using the set of all overlapping parts as the set of green light passage time windows for that traffic light intersection.

[0024] Specifically, the implementation process of this embodiment includes: The green lights at each intersection are aggregated through a time window. Based on the feasible time range for vehicles to pass through intersections with consecutive traffic lights, and combined with the signal timing and cycle information of each traffic light, calculate the set of feasible time intervals for passing through each intersection when the green light is on. ; (6); in, Let S represent the set of feasible green light passage time windows at the s-th intersection.

[0025] Furthermore, the process of performing surface fitting on the vehicle energy consumption item includes: sampling the motor power expression within a preset speed range and a preset acceleration range, calculating the motor power value corresponding to each sampling point, and constructing a fitting dataset containing speed, acceleration, and motor power; fitting the fitting dataset with a surface function to obtain an approximate analytical expression of motor power with respect to speed and acceleration; and introducing a speed square variable and a speed reciprocal auxiliary variable to replace the original variables in the multi-objective vehicle speed planning model, thereby converting the approximate analytical expression into a linear expression.

[0026] Specifically, the implementation process of this embodiment includes: The second step involves constructing a multi-objective mixed-integer linear speed planning model considering task time windows: This involves considering multiple urban logistics delivery points with time window constraints, combined with the set of feasible time intervals for passing through each intersection's green light. A high-efficiency and energy-saving multi-objective nonlinear model for vehicle speed planning was established. Through variable substitution and energy consumption model fitting, the nonlinear multi-objective model was transformed into a mixed integer linear programming model with objective priority ranking. High-efficiency and energy-saving multi-objective nonlinear model for vehicle speed planning: This model, based on total energy consumption and total travel time, and considering delivery time windows, feasible time intervals at various intersections, road speed limits, and vehicle dynamics constraints, is established in the distance domain as a multi-objective nonlinear model for vehicle speed planning. (7); (8); (9); in, Indicates energy consumption target, The time objective is represented by dividing the route into n segments, with a total of n+1 waypoints. Indicates the length of the k-th segment. Let the speed be at the k-th waypoint. This represents the time to reach the s-th intersection, where s = 0 indicates the initial time is 0. This represents the time taken to travel from the previous stop line s-1 at the intersection to the current stop line s at the intersection. This indicates the lower limit of the time window at intersection s. This indicates the upper limit of the time window at intersection s. For set A certain interval in, This indicates the total number of intersections. This represents the set of locations of each intersection among all waypoints. This represents the time of arrival at the m-th delivery point. , Let these represent the earliest arrival time and the latest arrival time of the m-th delivery point, respectively. This indicates the number of delivery points.

[0027] Energy consumption model fitting: First, according to the motor power expression (10), the speed v increases from 0 to 20 m / s, and the acceleration a increases from -2 m / s. 2 up to 1.6 m / s 2 Sampling is performed to calculate the corresponding motor power, and the fitted dataset is constructed accordingly; surface function fitting is performed according to equation (11); (10); (11); in, Indicates motor torque. Indicates the motor speed. This represents the motor efficiency map, which is a function of motor torque and speed. The sign function is 1 when the torque is greater than 0 and -1 when the torque is less than 0.

[0028] Furthermore, the process of transforming the multi-objective vehicle speed planning model into a mixed-integer linear programming model includes: configuring a 0-1 type discrete decision variable for each green light passage time window at each traffic light intersection in the multi-objective vehicle speed planning model, and setting constraints to limit that only one of the discrete decision variables at the same traffic light intersection has a value of 1; replacing the speed variable in the travel time expression of each road segment with the speed square variable, and introducing a linear approximation to express the travel time; determining whether the corresponding green light passage time window is selected based on the value of each discrete decision variable, and using the boundary value of the selected green light passage time window as the constraint boundary of the arrival time at the corresponding intersection.

[0029] Specifically, the implementation process of this embodiment includes: The second step also includes a mixed-integer linear programming model with objective priority ranking: through variable substitution. and and model approximation Transform models (7)-(9) into mixed integer linear programming models (12)-(14); transform linear multi-objective functions (12)-(13) into objectives with objective priorities, and transform them into objective programming models; (12); (13); (14); (15); in, , , , For from set The first two time windows selected in the text Choose variables for 0-1; , , , The positive and negative deviations of the two targets are respectively. and The objective for both objective functions is set to 0. and Prioritize the two objectives.

[0030] Furthermore, the process of determining the priority ranking of each optimization objective in the mixed-integer linear programming model based on the combination of the time urgency index and the traffic flow impedance coefficient includes: comparing the time urgency index with a first threshold and comparing the traffic flow impedance coefficient with a second threshold; when the time urgency index is greater than the first threshold and the traffic flow impedance coefficient is less than the second threshold, the priority of the total travel time minimization objective is set higher than the total travel energy consumption minimization objective; when the time urgency index is greater than the first threshold and the traffic flow impedance coefficient is greater than the second threshold, the priority of the total travel energy consumption minimization objective is set higher than the total travel time minimization objective; when the time urgency index is less than the first threshold and the traffic flow impedance coefficient is less than the second threshold, the priority of the total travel energy consumption minimization objective and the total travel time minimization objective are set to have the same priority.

[0031] Specifically, the implementation process of this embodiment includes: The third step is to solve the rolling optimization problem based on adaptive target priority: design adaptive target priority rules according to the urgency of the task time window and the traffic flow impedance coefficient, and establish a rolling optimization mechanism triggered by spatiotemporal hybrid. The urgency of the task time window: Assuming the current time is t, the remaining distance to the target delivery point is... The time window for the target node is The required minimum average vehicle speed is defined as follows: Combined with the vehicle's physical maximum speed limit Define the time urgency index : ; in, The optimal economic speed for the current road section. This indicates extreme urgency; it is necessary to travel at speeds close to the physical limits to avoid defaulting. This indicates that there is ample time and an economical driving speed can be used.

[0032] Traffic flow impedance coefficient: The average traffic flow velocity of the road segment ahead is obtained from the cloud. Define the traffic flow impedance coefficient. : ; in, This refers to the free-flow speed of vehicles on this section of road. This indicates severe congestion, where the longitudinal speed planning of vehicles is severely limited by the vehicle in front. It indicates smooth traffic flow and allows for flexible speed planning.

[0033] Adaptive target priority rule: Based on the time urgency index and traffic flow impedance coefficient, a target priority ranking rule is set: when Greater than 0.5 and When it is less than 0.5, the time objective takes priority; when Greater than 0.5 and When it is greater than 0.5, the energy consumption target takes priority; when Less than 0.5 and When it is greater than 0.5, the energy consumption target takes priority; when Less than 0.5 and When the value is less than 0.5, the energy consumption target and the time target are balanced.

[0034] Furthermore, the process of triggering model re-solution based on vehicle position and speed includes: continuously monitoring vehicle position and speed; triggering the first re-solution when the vehicle reaches a preset waypoint; repeatedly triggering re-solution at fixed time intervals when the vehicle has not reached the preset waypoint and the speed is lower than a preset speed threshold; and maintaining the current speed planning result and not triggering re-solution when the vehicle has not reached the preset waypoint and the speed is not lower than the preset speed threshold.

[0035] Specifically, the implementation process of this embodiment includes: Spatiotemporal hybrid triggering rolling optimization mechanism: Based on continuously monitored vehicle position and speed, when the vehicle reaches the predetermined waypoint, the waypoint triggering algorithm is activated to solve the vehicle speed planning model to update the vehicle speed curve of subsequent road segments; if the waypoint has not yet been reached but the vehicle speed is below 6 km / h, a per-second triggering mechanism is used to frequently solve the vehicle speed planning model under low-speed conditions to update the vehicle speed; otherwise, the previously generated optimal vehicle speed trajectory will remain unchanged and will not trigger a new model solution.

[0036] Example 2 Based on the same general inventive concept, this invention also provides a vehicle-cloud collaborative intelligent logistics vehicle speed planning system that considers task time windows. The vehicle-cloud collaborative intelligent logistics vehicle speed planning system considering task time windows described below can be referred to in correspondence with the vehicle-cloud collaborative intelligent logistics vehicle speed planning method considering task time windows described above. The system includes: The information acquisition module is used to acquire signal timing parameters at each traffic light intersection along the vehicle's route, as well as road speed limit information and road traffic flow information for each road segment. The passage window generation module is used to calculate the feasible time range for vehicles to pass through each traffic light intersection based on the predicted queue dissipation time of the intersection and the road speed limit of each road segment, and to extract the set of green light passage time windows for each traffic light intersection from the feasible time range according to the signal timing parameters of each traffic light intersection. The model building module is used to establish a multi-objective vehicle speed planning model with the optimization objectives of minimizing total vehicle energy consumption and total driving time, and with constraints such as time windows at each delivery point, green light time windows at each traffic light intersection, road speed limits, and vehicle dynamics. The model transformation module is used to introduce discrete decision variables for selecting the green light passage time window into the multi-objective vehicle speed planning model, and to perform surface fitting and variable substitution on the vehicle energy consumption term, so as to transform the multi-objective vehicle speed planning model into a mixed integer linear programming model with objective priority ranking. The priority decision module is used to calculate the time urgency index based on the remaining time and distance of the current delivery task, and to calculate the traffic flow impedance coefficient based on the traffic flow status of the road ahead. Based on the combination of the values ​​of the time urgency index and the traffic flow impedance coefficient, the priority ranking of each optimization objective in the mixed integer linear programming model is determined. The rolling optimization solution module is used to solve the mixed integer linear programming model using a rolling optimization method. It triggers model re-solution based on vehicle position and speed status, and outputs the speed planning results for each road segment.

[0037] In this embodiment, a computer terminal device is provided, including: One or more processors; A memory, coupled to the processor, for storing 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 steps of the above-described vehicle-cloud collaborative intelligent logistics vehicle speed planning method that takes into account task time windows.

[0038] In this embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the above-described vehicle-cloud collaborative intelligent logistics vehicle speed planning method that considers task time windows.

[0039] This invention provides a vehicle-cloud collaborative intelligent logistics vehicle speed planning method and system that considers task time windows. Addressing the characteristic of urban logistics delivery tasks with clearly defined arrival time requirements, it introduces task time window constraints from multiple delivery points into the vehicle speed planning process. By comprehensively considering the earliest and latest permitted arrival times at each delivery point, it achieves full-process speed optimization oriented towards the timeliness requirements of logistics delivery, avoiding the problem of existing speed planning methods that only focus on road traffic and ignore the timeliness of delivery tasks. This invention comprehensively considers intersection queue dissipation time, signal timing, road speed limits, and vehicle dynamics constraints, constructing a set of feasible green light passage time windows for continuous signalized intersections. It unifies the green light passage time window selection with vehicle speed trajectory planning, achieving joint optimization of discrete passage window decisions and continuous speed decisions. This avoids the local optimum problem caused by staged planning at each intersection, improving the overall traffic efficiency of vehicles passing through continuous signalized intersections. This invention addresses the problem of dynamic changes in task urgency and road traffic conditions during urban logistics delivery. Based on the task time window urgency index and traffic flow impedance coefficient, it adaptively adjusts the priority of driving time targets and energy consumption targets. Combined with a spatiotemporal hybrid triggering rolling optimization mechanism, it updates vehicle speed trajectory in real time, enabling vehicles to achieve dynamic coordination between delivery timeliness and driving energy consumption according to changes in delivery tasks and traffic environment.

[0040] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for planning the driving speed of a vehicle-cloud collaborative intelligent logistics vehicle considering task time windows, characterized in that, Includes the following steps: The cloud platform obtains signal timing parameters at each traffic light intersection along the vehicle's route, as well as road speed limits and traffic flow information for each road segment. Based on the predicted queue dissipation time at intersections and the speed limits of each road segment, the cloud calculates the feasible time range for vehicles to pass through each traffic light intersection, and extracts the set of green light passage time windows for each traffic light intersection from the feasible time range according to the signal timing parameters of each traffic light intersection. With the optimization objectives of minimizing total vehicle energy consumption and total driving time, and with constraints such as time windows at each delivery point, green light time windows at each traffic light intersection, road speed limits, and vehicle dynamics, a multi-objective vehicle speed planning model is established. Discrete decision variables for selecting green light passage time windows are introduced into the multi-objective vehicle speed planning model, and surface fitting and variable substitution are performed on the vehicle energy consumption term to transform the multi-objective vehicle speed planning model into a mixed integer linear programming model with objective priority ranking. The time urgency index is calculated based on the remaining time and distance of the current delivery task, and the traffic flow impedance coefficient is calculated based on the traffic flow status of the road ahead. The priority ranking of each optimization objective in the mixed integer linear programming model is determined based on the combination of the values ​​of the time urgency index and the traffic flow impedance coefficient. The mixed-integer linear programming model is solved using a rolling optimization method. The model is resolved based on the vehicle position and speed status uploaded by the vehicle terminal, and the speed planning results for each road segment are output and sent to the vehicle terminal for execution.

2. The method according to claim 1, characterized in that, The process of calculating the feasible time range for vehicles to pass through each traffic light intersection based on the predicted queue dissipation time and the road speed limits of each road segment includes: using an intelligent driver model to predict the acceleration of each vehicle in the current intersection queue; recursively calculating the queue length change curve over time based on the predicted acceleration values ​​of each vehicle in front; and taking the moment when the queue length shrinks to zero as the queue dissipation time; and calculating the shortest and longest passage times for vehicles to reach the stop line of the target intersection based on the distance between the vehicle's current position and the stop line of the target intersection, the maximum and minimum road speed limits, and the set expected acceleration and deceleration, and using the shortest and longest passage times to constitute the feasible time range.

3. The method according to claim 2, characterized in that, The process of extracting the set of green light passage time windows for each traffic light intersection from the feasible time region based on the signal timing parameters of each traffic light intersection includes: obtaining the start and end times of the green light signal for each traffic light intersection within a complete signal cycle; performing intersection operations on the time intervals of the feasible time region with each green light period within the signal cycle to extract the overlapping parts of the feasible time region and each green light period; and using the set of all overlapping parts as the set of green light passage time windows for the traffic light intersection.

4. The method according to claim 1, characterized in that, The process of surface fitting for vehicle energy consumption includes: sampling the motor power expression within a preset speed range and a preset acceleration range, calculating the motor power value corresponding to each sampling point, and constructing a fitting dataset containing speed, acceleration, and motor power; fitting the fitting dataset with a surface function to obtain an approximate analytical expression for motor power with respect to speed and acceleration; and introducing a speed square variable and a speed reciprocal auxiliary variable into the multi-objective vehicle speed planning model to replace the original variables, thereby converting the approximate analytical expression into a linear expression.

5. The method according to claim 4, characterized in that, The process of transforming the multi-objective vehicle speed planning model into a mixed-integer linear programming model includes: configuring a 0-1 type discrete decision variable for each green light passage time window at each traffic light intersection in the multi-objective vehicle speed planning model, and setting constraints to limit that only one of the discrete decision variables at the same traffic light intersection has a value of 1; replacing the speed variable in the travel time expression of each road segment with the speed square variable, and introducing a linear approximation to express the travel time; determining whether the corresponding green light passage time window is selected based on the value of each discrete decision variable, and using the boundary value of the selected green light passage time window as the constraint boundary of the arrival time at the corresponding intersection.

6. The method according to claim 1, characterized in that, The process of determining the priority ranking of each optimization objective in the mixed-integer linear programming model based on the combination of the time urgency index and the traffic flow impedance coefficient includes: comparing the time urgency index with a first threshold and comparing the traffic flow impedance coefficient with a second threshold; when the time urgency index is greater than the first threshold and the traffic flow impedance coefficient is less than the second threshold, the priority of the total travel time minimization objective is set higher than the total travel energy consumption minimization objective; when the time urgency index is greater than the first threshold and the traffic flow impedance coefficient is greater than the second threshold, the priority of the total travel energy consumption minimization objective is set higher than the total travel time minimization objective; when the time urgency index is less than the first threshold and the traffic flow impedance coefficient is less than the second threshold, the priority of the total travel energy consumption minimization objective and the total travel time minimization objective are set to have the same priority.

7. The method according to claim 1, characterized in that, The process of triggering model re-solution based on vehicle position and speed includes: continuously monitoring vehicle position and speed; triggering the first re-solution when the vehicle reaches a preset waypoint; repeatedly triggering re-solution at fixed time intervals when the vehicle has not reached the preset waypoint and the speed is lower than a preset speed threshold; and maintaining the current speed planning result without triggering re-solution when the vehicle has not reached the preset waypoint and the speed is not lower than the preset speed threshold.

8. A vehicle-cloud collaborative intelligent logistics vehicle speed planning system considering task time windows, characterized in that, The system for implementing the method of any one of claims 1-7 comprises: The information acquisition module is used to acquire signal timing parameters at each traffic light intersection along the vehicle's route, as well as road speed limit information and road traffic flow information for each road segment. The passage window generation module is used to calculate the feasible time range for vehicles to pass through each traffic light intersection based on the predicted queue dissipation time of the intersection and the road speed limit of each road segment, and to extract the set of green light passage time windows for each traffic light intersection from the feasible time range according to the signal timing parameters of each traffic light intersection. The model building module is used to establish a multi-objective vehicle speed planning model with the optimization objectives of minimizing total vehicle energy consumption and total driving time, and with constraints such as time windows at each delivery point, green light time windows at each traffic light intersection, road speed limits, and vehicle dynamics. The model transformation module is used to introduce discrete decision variables for selecting the green light passage time window into the multi-objective vehicle speed planning model, and to perform surface fitting and variable substitution on the vehicle energy consumption term, so as to transform the multi-objective vehicle speed planning model into a mixed integer linear programming model with objective priority ranking. The priority decision module is used to calculate the time urgency index based on the remaining time and distance of the current delivery task, and to calculate the traffic flow impedance coefficient based on the traffic flow status of the road ahead. Based on the combination of the values ​​of the time urgency index and the traffic flow impedance coefficient, the priority ranking of each optimization objective in the mixed integer linear programming model is determined. The rolling optimization solution module is used to solve the mixed integer linear programming model using a rolling optimization method. It triggers model re-solution based on vehicle position and speed status, and outputs the speed planning results for each road segment.

9. A computer terminal device, characterized in that, include: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors perform the steps of the 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, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-7.

Images