Automatic driving vehicle energy-saving dispatching method and system for automatic container terminal

Abstract

The invention relates to an automatic driving vehicle energy-saving scheduling method and system for an automatic container wharf, and the method comprises the steps: obtaining the operation information of a harbor district, obtaining an overall scheduling task demand according to the operation information, and monitoring the operation states of an automatic driving vehicle and operation equipment in real time; obtaining an energy-saving guiding task sequence of the automatic driving vehicle through an energy-saving guiding vehicle team task distribution model according to the overall scheduling task demand and the operation states of the automatic driving vehicle and the operation equipment; according to the energy-saving guiding task sequence of each automatic driving vehicle, based on a multi-vehicle cooperative energy-saving optimization target and a task accurate execution requirement, carrying out joint optimization on a vehicle transportation path and speed through an energy-saving track optimization model, determining energy-saving space-time feature points and passing points of vehicle operation, and generating a multi-vehicle cooperative energy-saving track; and performing energy-saving scheduling on each automatic driving vehicle according to the multi-vehicle collaborative energy-saving track. Compared with the prior art, the energy-saving upper limit and the operation efficiency are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of transportation technology, and in particular to an energy-saving scheduling method and system for automated container terminals using autonomous vehicles. Background Technology

[0002] In recent years, port transportation has increasingly focused on the energy consumption of automated container terminals. Existing operational data shows that energy costs impose a heavy economic burden on ports, and reducing these costs is expected to achieve significant savings. With the construction and widespread adoption of automated terminals, Automated Intelligent Vehicles (AIVs) have been introduced into the transportation system to automate horizontal transport at container terminals. Their energy consumption accounts for more than 40% of the total energy consumption of automated terminal transportation, resulting in considerable cost expenditures. Therefore, how to schedule and manage AIVs and reduce their energy consumption has become a critical issue that urgently needs to be addressed in terminal operations.

[0003] However, existing AIVs scheduling methods have many shortcomings: 1) Early extensive scheduling methods focused primarily on reducing the travel distance of AIVs, neglecting the impact of driving behavior during transport. This often resulted in AIVs traveling at maximum speed, forcing them to wait upon arrival at their destination because loading and unloading equipment was not yet available, thus wasting time and energy. In recent years, some studies have proposed refined scheduling and control strategies to achieve energy savings through dynamic speed adjustments. However, these methods typically rely on pre-determined shortest paths, leading to insufficient road space utilization. When there are many vehicles, roads are prone to congestion and intersection conflicts, with frequent vehicle starts and stops increasing energy consumption, extending transport operation time, causing cranes to wait, and further reducing overall operational efficiency.

[0004] 2) The introduction of autonomous vehicles necessitates the generation of computationally intensive trajectory guidance points in the scheduling system, leading to a dramatic expansion of the scheduling problem's dimensionality. Existing energy-efficient scheduling methods struggle to meet the high timeliness requirements of autonomous driving systems in terms of computational complexity and real-time response performance. In actual operation, emergencies such as urgent tasks, road closures, or traffic congestion frequently occur, requiring scheduling algorithms to generate feasible scheduling schemes within a very short time. However, the solution process for most current scheduling methods is time-consuming, typically requiring several minutes or even hours, leaving AIVs idle while waiting for task allocation. During long-term transportation operations, this idle time can accumulate to tens of hours, severely reducing the overall operational efficiency of the system. Even more detrimental is that delayed scheduling plans often prompt AIVs to adopt high-speed operation strategies to compensate for time losses, further increasing energy consumption and weakening the actual effectiveness of energy-efficient scheduling. For example, patent application CN113486293A discloses an intelligent horizontal transport system for a fully automated container terminal with loading and unloading operations. This system employs a dynamic path planning method that integrates an improved A-star algorithm and a dynamic window algorithm, scheduling tasks based on the principle of the fastest global operation time, and implementing multi-priority traffic control through vehicle-road cooperative technology. Although this method can adjust vehicle paths in real time, it can only plan the theoretically optimal path and cannot actively coordinate the loading and unloading equipment cycle to avoid idling. It can only perform conventional iterative calculations and cannot meet the millisecond-level real-time response requirements of autonomous driving systems. Its algorithm complexity may still lead to scheduling delays and vehicle idleness in large-scale scenarios, thereby increasing energy consumption and weakening overall operational efficiency. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide an energy-saving scheduling method and system for automated container terminals using autonomous vehicles. This invention significantly improves the energy-saving limit and operational efficiency.

[0006] The objective of this invention can be achieved through the following technical solutions: An energy-saving scheduling method for autonomous vehicles in an automated container terminal includes the following steps: Obtain the port area's operational information, derive the overall scheduling task requirements based on the operational information, and monitor the operating status of autonomous vehicles and operating equipment in real time; Based on the overall scheduling task requirements and the operating status of autonomous vehicles and operating equipment, an energy-saving task sequence for each autonomous vehicle is obtained through an energy-saving-oriented fleet task allocation model. Based on the energy-saving task sequence of each autonomous vehicle, and based on the multi-vehicle collaborative energy-saving optimization goal and the requirements for precise task execution, the vehicle transportation path and speed are jointly optimized through the energy-saving trajectory optimization model to determine the energy-saving spatiotemporal feature points and transit points of the vehicle operation, and generate the multi-vehicle collaborative energy-saving trajectory. Energy-saving scheduling is performed on each autonomous vehicle based on the multi-vehicle collaborative energy-saving trajectory.

[0007] Furthermore, the operational information includes road network structure, traffic conditions, historical operating status, vehicle types and quantities, and types and quantities of operating equipment.

[0008] Furthermore, the operating status of the autonomous vehicle includes the vehicle's position and speed, and the operating status of the operating equipment includes the operating status and task completion status.

[0009] Furthermore, the objective function of the energy-saving-oriented fleet task allocation model is: In the formula, Let be the objective function of the energy-saving-oriented fleet task allocation model. For efficiency-related coefficients, For autonomous vehicles Transport containers During the idle time at the crane equipment For a collection of autonomous vehicles, For crane The container collection for the operation, The coefficient is related to energy consumption. For autonomous vehicles Transport containers driving distance, The coefficient is the coefficient related to the conflict. As a node of the port area road network Number of times it was passed over This represents the average number of times all nodes in the port area's road network are traversed. A set of nodes; The autonomous vehicle Transport containers The formula for calculating idle time at the crane equipment is: In the formula, The weighting of the difference between the actual arrival time and the expected arrival time of autonomous vehicles. For containers On the crane The actual operation time at the location. For containers On the crane Expected operation time at the location, For containers The corresponding crane, For autonomous vehicles shipping containers The start time of operation at the first crane. For autonomous vehicles shipping containers The start time of operation at the second crane. For crane type, if crane It is a container The first processing equipment, If the crane It is a container The second processing unit, , For autonomous vehicles Transport containers The time of arrival at the first crane, For containers The start time of the preceding processing task on the first processing device. Handling containers for cranes The operation time cost, For autonomous vehicles The earliest available time within the current scheduling time window, For containers With autonomous vehicles The two variables, if the container By autonomous vehicles transportation, ,otherwise, , For autonomous vehicles At the beginning of the current scheduling time window, For containers The first crane, For average transport speed, For autonomous vehicles Transport containers The time to reach the second crane, For containers The start time of the preceding processing task on the second processing device. For containers The second crane, For cranes and The length of the shortest path between them. The average speed of the autonomous vehicle; The autonomous vehicle Transport containers The formula for calculating the driving distance is: The formula for calculating the potential conflict risk is as follows: In the formula, Let the container and the transport route nodes be binary variables. If the container The transportation route passes through nodes If the result is positive, the value is 1; otherwise, it is 0.

[0010] Furthermore, the constraints of the energy-saving-oriented fleet task allocation model include the uniqueness of container transportation, vehicle transportation capacity limitations, and time feasibility constraints. The uniqueness of the container transport is: In the formula, For containers With autonomous vehicles binary variables, For a collection of autonomous vehicles, For container assembly; The vehicle transport capacity is limited as follows: The time feasibility constraint is: In the formula, For containers On the crane Expected operation time at the location, For crane The container collection for the operation, For cranes, A collection of cranes.

[0011] Furthermore, the energy-saving-oriented fleet task allocation model employs an adaptive large neighborhood search algorithm to solve for the energy-saving-oriented task sequence for each autonomous vehicle. Specific steps include: The overall scheduling task requirements are represented as an encoded sequence using a single-chain encoding structure. The encoded sequence includes a job identifier and an AIV identifier. The order of the job identifiers indicates the order of operations of each container at the quay crane, and the AIV identifier indicates the number of each autonomous vehicle. A new solution is generated by removing some tasks from the encoded sequence using a destruction operator and re-inserting the removed tasks into the encoded sequence using a repair operator. The weights of the destruction and repair operators are dynamically adjusted based on their historical performance during the iteration process to optimize search efficiency. By performing destruction-repair operations in multiple iterations and evaluating the objective function value of the new solution after each iteration, an energy-saving-oriented task sequence is finally output.

[0012] Furthermore, the objective function of the energy-saving trajectory optimization model is: In the formula, The objective function of the energy-saving trajectory optimization model is... As time deviation weight, For autonomous vehicles The difference between the actual arrival time and the planned arrival time at the end of the route. As energy consumption weight, For nodes and nodes The distance between them For autonomous vehicles In road sections or arcs Energy consumption during driving For a set of nodes, A collection of autonomous vehicles; The autonomous vehicle The formula for calculating the difference between the actual arrival time and the planned arrival time at the end of the route is: In the formula, For autonomous vehicles The actual arrival time at the destination node of the route. For autonomous vehicles The planned time to reach the destination node of the route. For autonomous vehicles The endpoint of the route; The autonomous vehicle In road sections or arcs The formula for calculating driving energy consumption is: In the formula, The drag coefficient, The frontal area of ​​the vehicle. air density, For autonomous vehicles In road sections or arcs The speed on, For vehicle quality, For vehicle load capacity, For gravity, The rolling resistance coefficient, For autonomous vehicles With road segment or arc The binary variables, if autonomous vehicles Passing through the section of road or arc , Otherwise, it is 0.

[0013] Furthermore, the constraints of the energy-saving trajectory optimization model include path continuity constraints, speed limit constraints, time feasibility constraints for adjacent road segments, and conflict avoidance constraints. The path continuity constraint is: In the formula, For autonomous vehicles With road segment or arc The binary variables, if autonomous vehicles Passing through the section or arc , Otherwise, it is 0. For autonomous vehicles The starting node of the route, For autonomous vehicles The end point of the route, For a collection of autonomous vehicles, A set of nodes; The speed limit constraint is: In the formula, This represents the minimum speed allowed for autonomous vehicles to pass through nodes. This represents the maximum speed allowed for autonomous vehicles when passing through a node. The time feasibility constraint for the adjacent road segments is: In the formula, For autonomous vehicles Reaching the node Time, For autonomous vehicles Arrival section or arc Time, For autonomous vehicles Reaching the node Time, For nodes and nodes The distance between them For autonomous vehicles Reaching the node Time, For autonomous vehicles The actual departure time of the starting node of the task execution path. For autonomous vehicles The starting point of the transportation route, This is the lower limit of the time when a vehicle passes through a node. This is the maximum time limit for a vehicle to pass through a node. The conflict avoidance constraint is as follows: In the formula, For conflict autonomous vehicles and autonomous vehicles Through nodes The time interval, This refers to the allowed time interval when conflicting vehicles pass through the same node. For autonomous vehicles With road segment or arc The binary variables, if autonomous vehicles Passing through the section or arc , Otherwise, it is 0.

[0014] Furthermore, when scheduling energy-saving operations for each autonomous vehicle based on the multi-vehicle collaborative energy-saving trajectory, the multi-vehicle collaborative energy-saving trajectory is converted into trajectory spatiotemporal features via point commands, and corresponding vehicle operation actions are executed based on the trajectory spatiotemporal features via point commands.

[0015] According to another aspect of the present invention, an energy-saving dispatching system for automated vehicles in an automated container terminal is provided, comprising: The port area data acquisition module is used to acquire the port area's operational information, obtain the overall scheduling task requirements based on the operational information, and monitor the operating status of autonomous vehicles and operating equipment in real time. The energy-saving task planning module is used to obtain the energy-saving task sequence for each autonomous vehicle based on the overall scheduling task requirements and the operating status of autonomous vehicles and operating equipment through an energy-saving-oriented fleet task allocation model. The energy-saving trajectory optimization module is used to jointly optimize the vehicle transportation path and speed based on the energy-saving guidance task sequence of each autonomous vehicle, the multi-vehicle collaborative energy-saving optimization target and the task precision execution requirements, and the energy-saving trajectory optimization model to determine the energy-saving spatiotemporal feature points and transit points of the vehicle operation, and generate the multi-vehicle collaborative energy-saving trajectory. The energy-saving scheduling module is used to perform energy-saving scheduling for each autonomous vehicle based on the multi-vehicle collaborative energy-saving trajectory.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves coordinated optimization of vehicle speed and route selection through energy-saving fleet task allocation and energy-saving trajectory optimization, reducing energy loss and the number of vehicle conflict start-stops. While ensuring the full-load operation of the lifting equipment, it significantly reduces energy consumption and solves the problems of extensive scheduling ignoring the impact of driving behavior and relying on fixed routes, which leads to congestion and energy waste in the existing technology, thus significantly improving the overall energy saving limit.

[0017] 2. This invention achieves multi-vehicle collaboration through conflict avoidance constraints and path continuity constraints, and dynamically adjusts the energy-saving scheduling scheme in conjunction with real-time port operation information. This reduces operation conflicts and the number of stops, and improves the efficiency of container transshipment. It solves the problem of low operational efficiency caused by road congestion and scheduling disconnect in the prior art, and significantly improves operational efficiency.

[0018] 3. This invention adopts an adaptive large neighborhood search algorithm, which solves efficiently through dynamic weight adjustment and single-chain encoding structure. It can maintain excellent performance under different vehicle scales and task distribution conditions. It solves the problems of computational complexity and response delay leading to vehicle idleness in existing scheduling algorithms, and significantly improves the system's fast response capability, computational efficiency and scalability. It is suitable for various port area autonomous driving operation scenarios. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating an energy-saving scheduling method for automated vehicles in an automated container terminal, as proposed in this invention. Figure 2 This is a schematic diagram of a single-chain coding structure; Figure 3This is a schematic diagram of the structure of an energy-saving dispatching system for automated vehicles in an automated container terminal, as proposed in this invention. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0021] The following English abbreviations are involved: Gas-Insulated Switchgear (GIS) Automated Intelligent Vehicles (AIVs) Example 1 This embodiment provides an energy-saving scheduling method for automated driving vehicles in an automated container terminal, such as... Figure 1 As shown, it includes the following steps: S1. Obtain the port area's operational information, determine the overall scheduling task requirements based on the operational information, and monitor the operating status of autonomous vehicles and operating equipment in real time.

[0022] Operational information includes road network structure, traffic conditions, historical operating status, vehicle types and quantities, and types and quantities of operating equipment. The overall scheduling task requirements are obtained based on the analysis of operational information.

[0023] The system dynamically collects the operational status of autonomous vehicles and operating equipment, acquiring real-time information such as container operations and loading / unloading operations to provide data support for subsequent energy-saving scheduling. The operational status of autonomous vehicles includes vehicle position and speed, while the operational status of the operating equipment includes operational status and task completion status.

[0024] S2. Based on the overall scheduling task requirements and the operating status of autonomous vehicles and operating equipment, an energy-saving task sequence for each autonomous vehicle is obtained through an energy-saving-oriented fleet task allocation model.

[0025] Based on the overall operational requirements of the port's autonomous driving fleet and the loading and unloading equipment operation plan, the aim is to generate a task plan that reduces energy consumption while ensuring operational efficiency. Operational efficiency is maintained by minimizing the time deviation between the crane's start time and the vehicle's arrival time, while energy consumption is reduced by minimizing the total vehicle travel distance. Simultaneously, the vehicle's execution capability boundary must be considered to avoid potential conflicts arising from energy-saving task allocation, which could lead to difficulties in vehicle execution after command re-issuance. A balanced distribution of the optimal path vehicle intersection traffic index corresponding to the task is introduced to quantify potential conflict risks, and the executability of commands is improved by minimizing potential risks.

[0026] The objective function of the energy-saving-oriented fleet task allocation model is: In the formula, Let be the objective function of the energy-saving-oriented fleet task allocation model. For efficiency-related coefficients, For autonomous vehicles Transport containers During the idle time at the crane equipment For a collection of autonomous vehicles, For crane The container collection for the operation, The coefficient is related to energy consumption. For autonomous vehicles Transport containers driving distance, The coefficient is the coefficient related to the conflict. As a node of the port area road network Number of times it was passed over This represents the average number of times all nodes in the port area's road network are traversed. It is a set of nodes.

[0027] autonomous vehicles Transport containers The formula for calculating idle time at the crane equipment is: In the formula, The weighting of the difference between the actual arrival time and the expected arrival time of autonomous vehicles. For containers On the crane The actual operation time at the location. For containers On the crane Expected operation time at the location, For containers The corresponding crane, For autonomous vehicles shipping containers The start time of operation at the first crane. For autonomous vehicles shipping containers The start time of operation at the second crane. For crane type, if crane It is a container The first processing equipment, If the crane It is a container The second processing unit, , For autonomous vehicles Transport containers The time of arrival at the first crane, For containers The start time of the preceding processing task on the first processing device. Handling containers for cranes The operation time cost, For autonomous vehicles The earliest available time within the current scheduling time window, For containers With autonomous vehicles The two variables, if the container By autonomous vehicles transportation, ,otherwise, , For autonomous vehicles At the beginning of the current scheduling time window, For containers The first crane, For average transport speed, For autonomous vehicles Transport containers The time to reach the second crane, For containers The start time of the preceding processing task on the second processing device. For containers The second crane, For cranes and The length of the shortest path between them. This represents the average speed of the autonomous vehicle.

[0028] autonomous vehicles Transport containers The formula for calculating the driving distance is: The formula for calculating potential conflict risk is: In the formula, Let the container and the transport route nodes be binary variables. If the container The transportation route passes through nodes If the result is positive, the value is 1; otherwise, it is 0.

[0029] Considering the uniqueness of vehicle-cargo matching and vehicle transportation capacity limitations, the constraints of the energy-saving-oriented fleet task allocation model include the uniqueness of container transportation, vehicle transportation capacity limitations, and time feasibility constraints.

[0030] Each container can only be transported by one vehicle; the uniqueness of container transport is as follows: In the formula, For containers With autonomous vehicles binary variables, For a collection of autonomous vehicles, For container assembly; Each vehicle can only transport one box at a time, and the vehicle's transport capacity is limited to: The expected loading and unloading time for crane equipment must be greater than 0; the time feasibility constraint is: In the formula, For containers On the crane Expected operation time at the location, For crane The container collection for the operation, For cranes, A collection of cranes.

[0031] The energy-saving-oriented fleet task allocation model uses an adaptive large neighborhood search algorithm to solve for the energy-saving-oriented task sequence for each autonomous vehicle. The specific steps include: A single-chain coding structure is used to represent the overall scheduling task requirements as a coding sequence. The encoder and decoder can simultaneously process containers of different sizes and types. The coding sequence includes a job identifier and an AIV identifier. The order of the job identifiers indicates the order of operations for each container at the quay crane, and the AIV identifier represents the number of each autonomous vehicle. Figure 2As shown, containers A and D are 40-foot containers, and containers B and C are 20-foot containers; among them, B, C, and D are import containers, and A is an export container. In the coding sequence: the operation identifier indicates the first operation of the container (import containers correspond to quay crane operations, export containers correspond to yard crane operations); the AIV identifier indicates the autonomous driving truck number. The order of the operation identifiers indicates the operation sequence of each container at the quay crane; the AIV identifier corresponds to the vehicle's task allocation relationship.

[0032] A new solution is generated by removing some tasks from the encoded sequence using a destruction operator and then re-inserting the removed tasks into the encoded sequence using a repair operator. The weights of the destruction and repair operators are dynamically adjusted based on their historical performance during the iteration process to optimize search efficiency. The system performs a destruction-repair operation through multiple iterations, evaluating the objective function value of the new solution after each iteration, and finally outputting an energy-saving-oriented task sequence. Taking the example encoded sequence, the task order for AIV1 is B1→C1→C2→B2, and the task order for AIV2 is A1→A2→D1→D2.

[0033] This task allocation model does not impose constraints on work patterns (such as single-loop or double-loop) or vehicle transportation methods (such as whether full load is required). This means that the model can cover all possible work sequence schemes, and has greater flexibility and adaptability.

[0034] S3. Based on the energy-saving task sequence of each autonomous vehicle, and based on the multi-vehicle collaborative energy-saving optimization goal and the task precision execution requirements, the vehicle transportation path and speed are jointly optimized through the energy-saving trajectory optimization model to determine the energy-saving spatiotemporal feature points and transit points of the vehicle operation, and generate the multi-vehicle collaborative energy-saving trajectory.

[0035] Aiming to simultaneously ensure accurate execution of vehicle energy-saving tasks and energy conservation in actual driving trajectories, an objective function is constructed by weighted summation of task execution time deviation and vehicle trajectory energy consumption. The objective function of the energy-saving trajectory optimization model is: In the formula, The objective function of the energy-saving trajectory optimization model is... As the time deviation weight, For autonomous vehicles The difference between the actual arrival time and the planned arrival time at the end of the route. As energy consumption weight, For nodes and nodes The distance between them For autonomous vehicles In road sections or arcs Energy consumption during driving For a set of nodes, A collection of autonomous vehicles.

[0036] autonomous vehicles The formula for calculating the difference between the actual arrival time and the planned arrival time at the end of the route is: In the formula, For autonomous vehicles The actual arrival time at the destination node of the route. For autonomous vehicles The planned time to reach the destination node of the route. For autonomous vehicles The endpoint of the route.

[0037] autonomous vehicles In road sections or arcs The formula for calculating driving energy consumption is: In the formula, The drag coefficient, The frontal area of ​​the vehicle. air density, For autonomous vehicles In road sections or arcs The speed on, For vehicle quality, For vehicle load capacity, For gravity, The rolling resistance coefficient, For autonomous vehicles With road segment or arc The binary variables, if autonomous vehicles Passing through the section or arc , Otherwise, it is 0.

[0038] The constraints of the energy-saving trajectory optimization model include path continuity constraints, speed limit constraints, time feasibility constraints of adjacent road segments, and conflict avoidance constraints. The travel paths of all vehicles must remain continuous. The path continuity constraint is as follows: In the formula, For autonomous vehicles With road segment or arc The binary variables, if autonomous vehicles Passing through the section or arc , Otherwise, it is 0. For autonomous vehicles The starting node of the route, For autonomous vehicles The end point of the route, For a collection of autonomous vehicles, A set of nodes; The travel time of a vehicle on a road segment must meet the upper and lower speed limits. The speed limit constraints are as follows: In the formula, This represents the minimum speed allowed for autonomous vehicles to pass through nodes. This represents the maximum speed allowed for autonomous vehicles when passing through a node. The departure time of vehicles at intermediate nodes and their adjacent subsequent nodes should meet the time interval requirements, and the time feasibility constraints for adjacent road segments are as follows: In the formula, For autonomous vehicles Reaching the node Time, For autonomous vehicles Arrival section or arc Time, For autonomous vehicles Reaching the node Time, For nodes and nodes The distance between them For autonomous vehicles Reaching the node Time, For autonomous vehicles The actual departure time of the starting node of the task execution path. For autonomous vehicles The starting point of the transportation route, This is the lower limit of the time when a vehicle passes through a node. This is the maximum time limit for a vehicle to pass through a node. For vehicles colliding at an intersection, their passage time must maintain a certain interval. The conflict avoidance constraint is as follows: In the formula, For conflict autonomous vehicles and autonomous vehicles Through nodes The time interval, This refers to the allowed time interval when conflicting vehicles pass through the same node. For autonomous vehicles With road segment or arc The binary variables, if autonomous vehicles Passing through the section or arc , Otherwise, it is 0.

[0039] After linearizing the optimization problem, a mixed-integer nonlinear programming solver is used to solve it, outputting the optimized path of the autonomous truck and the average speed of each road segment. By combining the solutions from all autonomous vehicles, key nodes in the path that need to avoid conflicts and their corresponding vehicle arrival times are identified. A time window is allocated to each conflict node, requiring the arrival times of conflicting vehicles to satisfy conflict avoidance constraints. If not, the speed or path is adjusted for replanning, outputting spatiotemporal sampling point information, including node location, vehicle number, planned arrival time, and suggested speed, resulting in a multi-vehicle cooperative energy-saving trajectory.

[0040] S4. Perform energy-saving scheduling for each autonomous vehicle based on the multi-vehicle collaborative energy-saving trajectory.

[0041] When scheduling energy-saving operations for each autonomous vehicle based on the multi-vehicle collaborative energy-saving trajectory, the multi-vehicle collaborative energy-saving trajectory is converted into trajectory spatiotemporal characteristics via point commands, and corresponding vehicle operation actions are executed based on the trajectory spatiotemporal characteristics via point commands.

[0042] Example 2 This embodiment provides an energy-saving dispatching system for autonomous vehicles in an automated container terminal, such as... Figure 2 As shown, it includes: The port area data acquisition module is used to acquire port area operation information, obtain overall scheduling task requirements based on the operation information, and monitor the operation status of autonomous vehicles and operating equipment in real time. The energy-saving task planning module is used to obtain the energy-saving task sequence for each autonomous vehicle based on the overall scheduling task requirements and the operating status of autonomous vehicles and operating equipment through an energy-saving-oriented fleet task allocation model. The energy-saving trajectory optimization module is used to jointly optimize the vehicle transportation path and speed based on the energy-saving guidance task sequence of each autonomous vehicle, the multi-vehicle collaborative energy-saving optimization target and the task precision execution requirements, and the energy-saving trajectory optimization model to determine the energy-saving spatiotemporal feature points and transit points of the vehicle operation, and generate the multi-vehicle collaborative energy-saving trajectory. The energy-saving scheduling module is used to schedule energy-saving activities for each autonomous vehicle based on the multi-vehicle collaborative energy-saving trajectory.

[0043] By using vehicle status detectors, information such as vehicle location, speed, operation status, and task completion is collected in real time. This information is then used to assess the deviation in the scheduling task execution and feed it back to the energy-saving task planning module and the energy-saving trajectory optimization module to achieve closed-loop iterative optimization.

[0044] The rest is the same as in Example 1.

[0045] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for energy-saving scheduling of automated vehicles in an automated container terminal, characterized in that, Includes the following steps: Obtain the port area's operational information, derive the overall scheduling task requirements based on the operational information, and monitor the operating status of autonomous vehicles and operating equipment in real time; Based on the overall scheduling task requirements and the operating status of autonomous vehicles and operating equipment, an energy-saving task sequence for each autonomous vehicle is obtained through an energy-saving-oriented fleet task allocation model. Based on the energy-saving task sequence of each autonomous vehicle, and based on the multi-vehicle collaborative energy-saving optimization goal and the requirements for precise task execution, the vehicle transportation path and speed are jointly optimized through the energy-saving trajectory optimization model to determine the energy-saving spatiotemporal feature points and transit points of the vehicle operation, and generate the multi-vehicle collaborative energy-saving trajectory. Energy-saving scheduling is performed on each autonomous vehicle based on the multi-vehicle collaborative energy-saving trajectory.

2. The energy-saving scheduling method for automated driving vehicles in an automated container terminal according to claim 1, characterized in that, The operational information includes road network structure, traffic conditions, historical operating status, vehicle types and quantities, and types and quantities of operating equipment.

3. The energy-saving scheduling method for automated driving vehicles in an automated container terminal according to claim 1, characterized in that, The operating status of the autonomous vehicle includes the vehicle's position and speed, and the operating status of the operating equipment includes the operating status and task completion status.

4. The energy-saving scheduling method for automated driving vehicles in an automated container terminal according to claim 1, characterized in that, The objective function of the energy-saving-oriented fleet task allocation model is: In the formula, Let be the objective function of the energy-saving-oriented fleet task allocation model. For efficiency-related coefficients, For autonomous vehicles Transport containers During the idle time at the crane equipment For a collection of autonomous vehicles, For crane The container collection for the operation, The coefficient is related to energy consumption. For autonomous vehicles Transport containers driving distance, The coefficient is the coefficient related to the conflict. As a node of the port area road network Number of times it was passed over This represents the average number of times all nodes in the port area's road network are traversed. A set of nodes; The autonomous vehicle Transport containers The formula for calculating idle time at the crane equipment is: In the formula, The weighting of the difference between the actual arrival time and the expected arrival time of autonomous vehicles. For containers On the crane The actual operation time at the location. For containers On the crane Expected operation time at the location, For containers The corresponding crane, For autonomous vehicles shipping containers The start time of operation at the first crane. For autonomous vehicles shipping containers The start time of operation at the second crane. For crane type, if crane It is a container The first processing equipment, If the crane It is a container The second processing unit, , For autonomous vehicles Transport containers The time of arrival at the first crane, For containers The start time of the preceding processing task on the first processing device. Handling containers for cranes The operation time cost, For autonomous vehicles The earliest available time within the current scheduling time window, For containers With autonomous vehicles The two variables, if the container By autonomous vehicles transportation, ,otherwise, , For autonomous vehicles At the beginning of the current scheduling time window, For containers The first crane, For average transport speed, For autonomous vehicles Transport containers The time to reach the second crane, For containers The start time of the preceding processing task on the second processing device. For containers The second crane, For cranes and The length of the shortest path between them. The average speed of the autonomous vehicle; The autonomous vehicle Transport containers The formula for calculating the driving distance is: The formula for calculating the potential conflict risk is as follows: In the formula, Let the container and the transport route nodes be binary variables. If the container The transportation route passes through nodes If the result is positive, the value is 1; otherwise, it is 0.

5. The energy-saving scheduling method for automated driving vehicles in an automated container terminal according to claim 1, characterized in that, The constraints of the energy-saving-oriented fleet task allocation model include the uniqueness of container transportation, vehicle transportation capacity limitations, and time feasibility constraints. The uniqueness of the container transport is: In the formula, For containers With autonomous vehicles binary variables, For a collection of autonomous vehicles, For container assembly; The vehicle transport capacity is limited as follows: The time feasibility constraint is: In the formula, For containers On the crane Expected operation time at the location, For crane The container collection for the operation, For cranes, A collection of cranes.

6. The energy-saving scheduling method for automated driving vehicles in an automated container terminal according to claim 1, characterized in that, The energy-saving-oriented fleet task allocation model uses an adaptive large neighborhood search algorithm to solve for the energy-saving-oriented task sequence for each autonomous vehicle. The specific steps include: The overall scheduling task requirements are represented as an encoded sequence using a single-chain encoding structure. The encoded sequence includes a job identifier and an AIV identifier. The order of the job identifiers indicates the order of operations of each container at the quay crane, and the AIV identifier indicates the number of each autonomous vehicle. A new solution is generated by removing some tasks from the encoded sequence using a destruction operator and re-inserting the removed tasks into the encoded sequence using a repair operator. The weights of the destruction and repair operators are dynamically adjusted based on their historical performance during the iteration process to optimize search efficiency. By performing destruction-repair operations in multiple iterations and evaluating the objective function value of the new solution after each iteration, an energy-saving-oriented task sequence is finally output.

7. The energy-saving scheduling method for automated driving vehicles in an automated container terminal according to claim 1, characterized in that, The objective function of the energy-saving trajectory optimization model is: In the formula, The objective function of the energy-saving trajectory optimization model is... As the time deviation weight, For autonomous vehicles The difference between the actual arrival time and the planned arrival time at the end of the route. As energy consumption weight, For nodes and nodes The distance between them For autonomous vehicles In road sections or arcs Energy consumption during driving For a set of nodes, A collection of autonomous vehicles; The autonomous vehicle The formula for calculating the difference between the actual arrival time and the planned arrival time at the end of the route is: In the formula, For autonomous vehicles The actual arrival time at the destination node of the route. For autonomous vehicles The planned time to reach the destination node of the route. For autonomous vehicles The endpoint of the route; The autonomous vehicle In road sections or arcs The formula for calculating driving energy consumption is: In the formula, The drag coefficient, The frontal area of ​​the vehicle. air density, For autonomous vehicles In road sections or arcs The speed on, For vehicle quality, For vehicle load capacity, For gravity, The rolling resistance coefficient, For autonomous vehicles With road segment or arc The binary variables, if autonomous vehicles Passing through the section or arc , Otherwise, it is 0.

8. The energy-saving scheduling method for automated driving vehicles in an automated container terminal according to claim 7, characterized in that, The constraints of the energy-saving trajectory optimization model include path continuity constraints, speed limit constraints, time feasibility constraints of adjacent road segments, and conflict avoidance constraints. The path continuity constraint is: In the formula, For autonomous vehicles With road segment or arc The binary variables, if autonomous vehicles Passing through the section or arc , Otherwise, it is 0. For autonomous vehicles The starting node of the route, For autonomous vehicles The end point of the route, For a collection of autonomous vehicles, A set of nodes; The speed limit constraint is: In the formula, This represents the minimum speed allowed for autonomous vehicles to pass through nodes. This represents the maximum speed allowed for autonomous vehicles when passing through a node. The time feasibility constraint for the adjacent road segments is: In the formula, For autonomous vehicles Reaching the node Time, For autonomous vehicles Arrival section or arc Time, For autonomous vehicles Reaching the node Time, For nodes and nodes The distance between them For autonomous vehicles Reaching the node Time, For autonomous vehicles The actual departure time of the starting node of the task execution path. For autonomous vehicles The starting point of the transportation route, This is the lower limit of the time when a vehicle passes through a node. This is the maximum time limit for a vehicle to pass through a node. The conflict avoidance constraint is as follows: In the formula, For conflict autonomous vehicles and autonomous vehicles Through nodes The time interval, This refers to the allowed time interval when conflicting vehicles pass through the same node. For autonomous vehicles With road segment or arc The binary variables, if autonomous vehicles Passing through the section or arc , Otherwise, it is 0.

9. The energy-saving scheduling method for automated driving vehicles in an automated container terminal according to claim 1, characterized in that, When scheduling energy-saving operations for each autonomous vehicle based on the multi-vehicle collaborative energy-saving trajectory, the multi-vehicle collaborative energy-saving trajectory is converted into trajectory spatiotemporal features via point commands, and corresponding vehicle operation actions are executed based on the trajectory spatiotemporal features via point commands.

10. An energy-saving dispatching system for automated vehicles in an automated container terminal, characterized in that, include: The port area data acquisition module is used to acquire the port area's operational information, obtain the overall scheduling task requirements based on the operational information, and monitor the operating status of autonomous vehicles and operating equipment in real time. The energy-saving task planning module is used to obtain the energy-saving task sequence for each autonomous vehicle based on the overall scheduling task requirements and the operating status of autonomous vehicles and operating equipment through an energy-saving-oriented fleet task allocation model. The energy-saving trajectory optimization module is used to jointly optimize the vehicle transportation path and speed based on the energy-saving guidance task sequence of each autonomous vehicle, the multi-vehicle collaborative energy-saving optimization target and the task precision execution requirements, and the energy-saving trajectory optimization model to determine the energy-saving spatiotemporal feature points and transit points of the vehicle operation, and generate the multi-vehicle collaborative energy-saving trajectory. The energy-saving scheduling module is used to perform energy-saving scheduling for each autonomous vehicle based on the multi-vehicle collaborative energy-saving trajectory.

Images