Working method of a finished product delivery automation system with AGV path planning
By optimizing AGV path planning using a multi-objective particle swarm optimization algorithm and a dynamic window method, and combining it with a PLC control system and automated equipment, the path planning and data management problems of AGV logistics systems in complex environments are solved, resulting in an efficient, safe, and easily scalable logistics automation system.
Patent Information
- Application Number
- CN202410805646.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-21
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-06-21
AI Technical Summary
Existing AGV logistics systems face difficulties in path planning in complex environments, cannot completely replace manual labor, pose data processing and safety risks, and are costly, making them difficult to adapt to dynamic logistics needs.
A multi-objective particle swarm optimization algorithm combined with a dynamic window method is used for AGV path planning. Combined with a PLC control system, transfer trolley, gantry and trolley, automated logistics operations are realized. Information is collected and analyzed in real time through a code reading system to optimize resource utilization and avoid collisions.
It improves logistics efficiency, reduces human error, ensures information accuracy and security, lowers maintenance costs, and the system is easy to expand and upgrade, adapting to complex environmental needs.
Smart Images

Figure CN118579455B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of warehousing logistics, and in particular to a working method of a finished product delivery automation system with AGV path planning. Background Art
[0002] In traditional logistics, a large number of manual operations have led to employees having to undertake heavy physical labor, such as handling, sorting and stacking of goods. This work mode is not only inefficient, but also puts great pressure on the health of employees. AGV logistics has changed this situation through automated and intelligent technical means. AGV can independently complete the handling, transportation and stacking of goods, greatly reducing the labor intensity of employees, while improving the efficiency and accuracy of logistics operations.
[0003] However, although AGV has replaced some manual operations in many logistics scenarios, current technology still cannot completely get rid of its dependence on manpower. In some complex logistics tasks or specific environments, AGV may not be able to fully adapt and require human intervention or assistance. This limits the effectiveness of AGV in reducing labor intensity and may cause some employees to still have to undertake heavy physical labor; although AGV can reduce direct labor costs, the costs of purchasing, maintaining, upgrading, and system integration of AGV are also a considerable expense. For some logistics companies with smaller scale or limited funds, the introduction of AGV may not be cost-effective. In addition, with the continuous development of technology, the speed of AGV replacement is also accelerating, and companies need to continue to invest funds to maintain the advancement of technology, which also increases the operating costs of companies; although AGV systems are usually highly automated and intelligent, they may still be affected by human factors during the transmission and processing of data streams. For example, human input errors, improper system configuration, or network security issues may lead to misoperation of data streams. These problems may affect the normal operation and logistics efficiency of AGV, and may even cause losses to the enterprise; the AGV logistics system in the existing technology still has some challenges in processing outbound data and inbound system big data. First, with the continuous growth and complexity of logistics business, the amount of data is also increasing rapidly, which puts higher requirements on the data processing capabilities of AGV systems. Secondly, how to ensure the accuracy and integrity of data is also an important issue. Data loss, damage or tampering may occur during data transmission and storage, which will affect the accuracy and traceability of logistics. Finally, how to make full use of big data to optimize logistics operations, improve efficiency and reduce costs is also a question worth exploring.
[0004] AGVs follow pre-defined routes or use navigation algorithms to depict their trajectories. In complex and dynamic environments, AGV path planning becomes a difficult task that requires optimizing multiple goals at the same time, especially energy saving, driving time, and collision avoidance. These traditional methods may not be able to address the dynamic characteristics of the environment and the multifaceted goals associated with AGV operations. The main challenge of AGV path planning is how to achieve a balance while optimizing many goals, avoid collisions, and flexibly adapt to the changing environment. This requires finding a method that can effectively harness the complexity of multi-objective optimization and seamlessly integrate real-time environmental information.
[0005] In summary, although AGV logistics technology has made significant progress in many aspects, there are still some problems and challenges in practical application. In order to solve these problems, we need to continue to promote technological innovation and development, improve the automation and intelligence level of AGV systems, and strengthen data management and network security protection. Summary of the invention
[0006] In order to overcome the deficiencies of the prior art solutions, the present invention provides a working method of a finished product delivery automation system with AGV path planning.
[0007] In order to achieve the above-mentioned purpose, the technical scheme of the present invention is: a working method of a finished product delivery automation system with AGV path planning, the finished product delivery automation system comprises: a PLC control system, a transfer trolley, a gantry, a trolley, a logistics vehicle and an AGV, the PLC control system uses the AGV path planning to realize the control of the AGV, and realizes the movement of the AGV between the transfer trolley, the gantry, and the empty trolley docking station device,
[0008] The working method is specifically as follows:
[0009] A1. Logistics vehicles enter the platform workstation;
[0010] A2. Move the goods from the logistics vehicle to the empty trolley at the platform, and move the goods from the fully loaded trolley to the logistics vehicle;
[0011] A3. Transport the empty trolley to the AGV docking station;
[0012] A4. The barcode reading system automatically collects and analyzes the trolley to select the production line, and the AGV transports the empty trolley to the corresponding production line to connect to the gantry workstation;
[0013] A5. The transfer cart logically takes empty boxes from the empty trolley to the corresponding empty box aisle for returning to the warehouse, and logically takes full boxes from the corresponding finished product aisle to the corresponding trolley aisle;
[0014] A6, AGV transports a full trolley to the docking chain station on the platform;
[0015] A7. The AGV exits the workstation, the chain device starts, and pulls the full trolley to the platform level;
[0016] A8. Push the full vehicle into the docking station of the logistics vehicle;
[0017] A9. Wait for the next logistics truck to ship.
[0018] Among them, AGV path planning includes the following steps:
[0019] S1. Initialization: Initialize the multi-objective particle swarm algorithm with a particle swarm in the desired region space. Each particle represents a potential solution for the AGV path. Generate a dynamic speed window based on the function of the AGV. For each particle in the swarm, perform the following steps:
[0020] S1.1, using a multi-objective particle swarm algorithm to optimize the speed selection within the dynamic window according to multiple objectives;
[0021] S1.2, applying the dynamic window method to select the highest scoring trajectory for the AGV based on the optimized speed;
[0022] S2. Update the position and velocity of the particles using a multi-objective particle swarm algorithm. Repeatedly generate new solutions and check fitness until the termination condition is met, the maximum number of iterations is met, or convergence occurs;
[0023] S3. Select the best solution from the group as the best path for the AGV.
[0024] Beneficial effects:
[0025] The working method of the finished product delivery automation system of the present invention, especially the technology of combining AGV (automatic guided vehicle) path planning, improves logistics efficiency: through the automated path planning and handling of AGV, the time and labor cost of manual handling are greatly reduced, thereby significantly improving the efficiency of logistics processing; reduce human errors: manual operation often has errors and uncertainties, while the automated operation of AGV can accurately and stably complete the handling task, reducing errors caused by human factors; real-time information management: the barcode reading system automatically collects and analyzes the information of the trolley selecting the production line, realizes the real-time tracking and management of logistics information, and improves the efficiency and accuracy of information processing; high flexibility: the path planning of AGV can be It can be flexibly adjusted according to actual needs, adapt to logistics needs in different scenarios, and improve the flexibility and adaptability of the system; optimize resource utilization: the transfer cart sends empty boxes and picks up full boxes in a logical order, ensuring the maximum utilization of resources and avoiding waste of resources; improve safety performance: the automated operation of AGV can ensure that collisions with personnel or other equipment are avoided during the handling process, improving work safety; reduce maintenance costs: the automated operation of equipment such as AGV and transfer carts reduces equipment wear and failure, thereby reducing equipment maintenance costs; easy to expand and upgrade: the entire system adopts a modular design, which is easy to expand and upgrade, and can easily adapt to changes in future logistics needs. The working method of the finished product delivery automation system realizes the automation, informatization and intelligence of logistics processing by introducing AGV path planning technology, improves logistics efficiency, reduces costs, optimizes resource allocation, and improves the safety and flexibility of the system.
[0026] This paper introduces a groundbreaking AGV path planning method that combines multi-objective and dynamic window methods to overcome the limitations of traditional mobile AGV autonomous navigation methods. It helps to calculate the optimal speed and steering angle to ensure safe navigation. By mathematically modeling the multi-objective function using the shortest path and maximum smoothness objective functions, it effectively balances the conflicting objectives and improves the overall performance, improving efficiency, optimizing paths and reducing inefficiencies. The dynamic window method successfully selects the trajectory with the highest score, taking into account the optimization speed and environmental obstacles. The effectiveness of the method is verified through rigorous simulations and comparisons with existing methods, showing advantages such as shorter planning time, shorter path length, fewer turning points and higher route safety. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 It is a flow chart of the working method of the finished product delivery automation system of the present invention;
[0028] Figure 2 A flow chart of the method for AGV path planning of the present invention;
[0029] Figure 3 It is a structural diagram of the finished product delivery automation system of the present invention;
[0030] Figure 4 It is a structural diagram of the gantry and the transfer trolley docking of the present invention; DETAILED DESCRIPTION
[0031] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0032] like Figure 1 As shown, a working method of a finished product delivery automation system with AGV path planning, the finished product delivery automation system includes: a PLC control system, a transfer trolley, a gantry, a trolley, a logistics vehicle and an AGV. The PLC control system uses AGV path planning to control the AGV and realize the movement of the AGV between the transfer trolley, the gantry, and the empty trolley docking station device.
[0033] The working method is specifically as follows:
[0034] A1. Logistics vehicles enter the platform workstation;
[0035] A2. Move the goods from the logistics vehicle to the empty trolley at the platform, and move the goods from the fully loaded trolley to the logistics vehicle;
[0036] A3. Transport the empty trolley to the AGV docking station; specifically:
[0037] A3.1. Manually push the empty trolley to complete the material information labeling production line;
[0038] A3.2, the empty trolley arrives at the AGV docking station;
[0039] A3.3. The code reading mechanism above the docking station identifies the corresponding information of the empty trolley production line;
[0040] A3.4, the empty trolley at the docking station is in the AGV waiting for transportation;
[0041] A3.5. The AGV moves to the docking station, lifts the lifting panel, and the AGV leaves the station with the empty trolley;
[0042] A3.6, AGV automatically sends to the corresponding production line according to the information;
[0043] A3.7, AGV transports the empty trolley and the work is completed;
[0044] A4. The barcode reading system automatically collects and analyzes the trolley to select the production line, and the AGV transports the empty trolley to the corresponding production line to connect to the gantry workstation. Specifically:
[0045] A4.1. The AGV transports the empty trolley to the workstation, and the AGV confirms the arrival and the positioning detection device is started;
[0046] A4.2. The AGV lifting panel descends and exits the workstation, and the gantry mechanism starts;
[0047] A4.3. The empty trolley and the transfer trolley are on the same level and are ready for operation;
[0048] A4.4. The gantry reads the barcodes to identify each empty container return lane and the finished product pickup information;
[0049] A4.5. The gantry mechanism opens the empty trolley aisle in sequence according to the design logic, and the transfer trolley delivers empty boxes and takes full boxes in sequence according to the logical order. After the transfer trolley delivers two finished products to the trolley each time, the gantry closes the guardrail device in sequence; synchronously: the transfer trolley is aligned with the finished product aisle, the signal automatically recognizes the cylinder top rises, and the baffle of the finished product aisle is lowered. After taking all the finished products, the baffle automatically rises to block the last finished product in the aisle;
[0050] A4.6. After the transfer trolley completes the task, it performs tasks at other workstations;
[0051] A4.7, the gantry station is full of trolleys waiting for AGV to transport;
[0052] A5. The transfer cart logically takes empty boxes from the empty trolley to the corresponding empty box aisle for returning to the warehouse, and logically takes full boxes from the corresponding finished product aisle to the corresponding trolley aisle. Specifically:
[0053] A5.1. The empty trolley is waiting for the transfer trolley to pick up the goods;
[0054] A5.2. The gantry mechanism robot opens the first trolley guardrail and pushes the empty container onto the transfer trolley;
[0055] A5.3. The transfer trolley transports the empty container to dock with the empty container cargo lane and rises to the docking height at the same time;
[0056] A5.4. The transfer trolley delivers the empty container to the empty container aisle;
[0057] A5.5. The transfer trolley runs to the finished product lane, picks up the goods and delivers them to the trolley. The transfer trolley delivers empty boxes and picks up full boxes in a logical order.
[0058] A5.6. After the transfer trolley delivers two finished products to the trolley each time, the gantry closes the guardrail device in turn;
[0059] A5.7, and so on, the transfer trolley completes the task of emptying and filling;
[0060] A5.8. After the transfer trolley completes the task, it will perform tasks at other workstations;
[0061] Among them, A5.2, the gantry mechanism robot device opens the first trolley guardrail and pushes the empty box to the transfer trolley, specifically:
[0062] A5.2.1. AGV carries the empty trolley to the gantry station;
[0063] A5.2.2, gantry workstation mechanism starts;
[0064] A5.2.3. The gantry robot moves horizontally to the designated guardrail position;
[0065] A5.2.4. The robot moves horizontally to open the lock pin;
[0066] A5.2.5. The robot pushes the empty box to the transfer cart;
[0067] A5.2.6, the robot mechanism waits for the transfer trolley to arrive;
[0068] A5.2.7. The robot completes the collection of the first empty box and then begins to open the guardrail of the next empty box on the empty trolley;
[0069] A5.2.8. The transfer trolley delivers empty boxes to pick up full boxes in a logical order. After the transfer trolley delivers two finished products to the trolley each time, the gantry closes the guardrail device in turn;
[0070] Among them, A5.5, the transfer trolley runs to the finished product lane, picks up the goods and delivers them to the trolley. The transfer trolley delivers empty boxes and picks up full boxes in a logical order, specifically:
[0071] A5.5.1. The transfer trolley arrives at the corresponding cargo lane and is in the material-retrieving state;
[0072] A5.5.2. The system starts the finished product baffle aisle device, and the cylinder lifts up to open the baffle;
[0073] A5.5.3. The system confirms that the end baffle of the finished product aisle is fully opened;
[0074] A5.5.4. The electric roller at the end of the finished product aisle starts to deliver the finished products to the transfer trolley;
[0075] A5.5.5. The transfer trolley is logically transported to the corresponding trolley recovery channel position;
[0076] A5.5.6. The electric roller of the transfer trolley is started and the finished product is sent into the trolley;
[0077] A5.5.7. The transfer cart starts a task of delivering an empty container and picking up a full container;
[0078] A5.5.8. The cylinder lifts and pushes the baffle, and the transfer trolley is ready to take the finished goods;
[0079] A5.5.9. The cylinder lift baffle mechanism is equipped with a metal proximity sensor to ensure that the baffle is open. In extreme cases, if it is not opened, an alarm will be triggered to remind manual handling.
[0080] Among them, A5.6, after the transfer trolley delivers two finished products to the trolley each time, the gantry closes the guardrail device in sequence, specifically:
[0081] A5.6.1. The transfer trolley delivers the finished product to the trolley cargo lane to confirm that it is in place;
[0082] A5.6.2, gantry workstation mechanism starts;
[0083] A5.6.3. The gantry robot moves horizontally to the designated guardrail position;
[0084] A5.6.4, robot arm closes the lock pin;
[0085] A5.6.5. The robot mechanism opens the next empty box lane for trolleys;
[0086] A5.6.6. The transfer trolley delivers empty boxes to full boxes in a logical order. After the transfer trolley delivers two finished products to the trolley each time, the gantry closes the guardrail device in turn;
[0087] A6, AGV transports a full trolley to the docking chain station on the platform.
[0088] A6.1. AGV transports a full trolley to the chain docking station;
[0089] A6.2. The AGV lifting panel descends, and the full vehicle falls into the slot—the AGV rotates and exits from the side;
[0090] A6.3. The fully loaded trolley falls smoothly into the chain slot;
[0091] A6.4. The chain device starts, the fully loaded vehicle smoothly rises to the platform level, moves along the platform slot direction, and automatically decelerates and stops;
[0092] A7. The AGV exits the workstation, the chain device starts, and pulls the full trolley to the platform level;
[0093] A8. Push the full vehicle into the docking station of the logistics vehicle;
[0094] A9. Wait for the next logistics truck to ship.
[0095] Among them, AGV path planning includes the following steps:
[0096] S1. Initialization: Initialize the multi-objective particle swarm algorithm with a particle swarm in the desired region space. Each particle represents a potential solution for the AGV path. Generate a dynamic speed window based on the function of the AGV. For each particle in the swarm, perform the following steps:
[0097] S1.1, using a multi-objective particle swarm algorithm to optimize the speed selection within the dynamic window according to multiple objectives;
[0098] S1.2, applying the dynamic window method to select the highest scoring trajectory for the AGV based on the optimized speed;
[0099] S2. Update the position and velocity of the particles using a multi-objective particle swarm algorithm. Repeatedly generate new solutions and check fitness until the termination condition is met, the maximum number of iterations is met, or convergence occurs;
[0100] S3, select the best solution from the group as the best path for the AGV;
[0101] Among them, the multi-objective particle swarm optimization algorithm includes:
[0102] M1. Initialization: Randomly generate the positions and velocities of a group of particles in the search space. Set the personal best position and fitness of each particle at the initialization position to the initial value. The particle uses its best fitness to identify the global best position, satisfying the following formula:
[0103]
[0104] Where S and V are the position and velocity of the initialized particle respectively; Based on the cost function The fitness value of the particle position; Ub and Lb are the problem areas at the boundaries of the search space; is the random generator function; d is the dimension; represents the initial velocity in the d dimension.
[0105] M2. Update the speed and position of particles:
[0106] in, ,
[0107] in, and is a generated random number arranged in [0,1]; represents the inertia weight; d is the dimension; and are the variable vectors representing the individual best position and the global best position of the particles of the current generation, respectively; and represents the acceleration coefficient; the particle position S is updated using the following formula; and represents the velocity of the particle at t and t+1; represents the position of the particle at t;
[0108] ,
[0109] in, and represents the position and velocity of the particle at time t, and Represents the position and velocity of the particle at t+1, which is the current generation or iteration and index of the particle id;
[0110] M3, evaluation, the objective function or fitness of each particle's new position is calculated by multiple objective functions, each of which represents a different evaluation goal;
[0111] Optimal AGV path planning is performed through a multi-objective particle swarm in a dynamic window method technique. When managing multiple objectives of a multi-objective particle swarm, a particle swarm with different positions and velocities represents a potential solution; particles traverse the search space to find the best answer that optimizes the specified objective; initialize randomly initialized particle positions and velocities, evaluate the objectives established for AGV path planning, evaluate the fitness of each particle, update the individual best and global best based on the individual best position of each particle, update the global best position using the individual and collective best proxy solutions, update the velocity and position of the particle, and whether the termination criteria are met;
[0112] The multi-objective consists of two objective functions: the shortest path objective function and the maximum smoothness objective function. The shortest path objective function is modeled by calculating the distance along the determined points to guide the AGV forward or backward, turn or move as path planning; the first objective function of the shortest path is expressed as follows:
[0113]
[0114] in, It is the first objective function of the shortest path; and The motion coordinates are and The positions of the i-th steering angle and the i+1-th steering angle in the working area space; is the total number of turning angles of the mobile AGV;
[0115] The second objective function is the maximum smoothness function, as shown below,
[0116] ,
[0117] in, is the second objective function. By using the weight parameter and the advantage optimal solution, the weight parameter of the multi-objective function is expressed as follows:
[0118] ,
[0119] F is a multi-objective function; ω is a weighted parameter of the balance objective;
[0120] Dominance calculation comparisons are performed between particles to determine dominance relations, and each particle in each front is assigned a crowding distance to maintain diversity, and a dominance rank is assigned to the particle based on its dominance calculation relationship.
[0121] M4. Update individual and global optimal values: By comparing the fitness value of each particle target with its individual optimal fitness, if the current fitness is better for any target, update the individual optimal position and fitness value of that objective function, and calculate the global optimal fitness of each target by comparing the fitness value of the particle with whether the current fitness is better for any objective function and updating the global optimal position and fitness of that target.
[0122] M5. Dominance calculation: Dominance calculation is performed by comparing particles to determine dominance relationships; dominance levels are assigned to particles based on their dominance calculation relationships; weights are dynamically adjusted based on the distribution of solutions along the dominance frontier according to the performance of the particles and the characteristics of the optimization problem;
[0123] M6, non-dominated sorting: particles entering different non-dominated fronts are sorted according to their dominance levels, expressed as the following formula:
[0124]
[0125] in, Indicates dominance level; and are the objective function values of the two adjacent non-inferior solutions of the j-th non-inferior solution in the objective function solution set; and are the maximum and minimum values in the objective function solution set, respectively; N is the total number of objective functions; o is the ordinal number of the objective function; k is the total number of non-inferior solutions. The crowding distance assigned to each particle in each frontier to maintain diversity;
[0126] M7, selection and reproduction: select particles from the non-dominant frontier according to the dominance level and crowding distance of the particles; combine the selected particles through crossover and mutation operations to reproduce new particles;
[0127] M8, Termination condition: This step is performed by checking whether the termination condition is met, the maximum number of iterations is reached, or the desired goal is reached; if the termination condition is met, the algorithm is stopped; otherwise, return to M2;
[0128] Among them, the following are the detailed steps of the dynamic window algorithm:
[0129] D1. Perception and positioning: Get sensor readings to perceive the environment and estimate the AGV’s current position and orientation;
[0130] D2. Motion prediction: predict the future motion of the AGV based on the current state of the AGV and the dynamic model; consider the speed, acceleration and steering ability of the AGV;
[0131] D3, Dynamic Window Generation: Define a dynamic window to represent the feasible speed and steering combinations of the AGV; set the maximum and minimum limits of the linear speed and angular velocity according to the capabilities of the AGV;
[0132] D4, trajectory evaluation: Generate a set of candidate trajectories by sampling the velocity within a dynamic window; evaluate each trajectory based on different criteria (such as proximity to obstacles, proximity to the target, and smoothness of motion); assign a score or cost to each route based on the evaluation criteria;
[0133] D5, trajectory selection: select the trajectory with the lowest cost or the highest score as the desired trajectory of the AGV. Ensure that the selected trajectory is collision-free and satisfies the dynamic constraints of the AGV;
[0134] D6, motion execution: convert the desired trajectory into the control command of the AGV actuator. Execute the control command to make the AGV move along the desired trajectory;
[0135] D7, loop: Repeat D1 to D6 continuously, update the AGV's motion plan according to the changing environment and AGV status; adjust the dynamic window and re-evaluate the trajectory to adapt to the dynamic characteristics of the field;
[0136] Among them, the dynamic model of AGV is specifically as follows:
[0137] AGV is responsible for transporting items within a given space. AGV must be able to bypass static obstacles while maintaining the shortest path and minimizing the steering angle. This requirement is necessary to meet multiple goals simultaneously. The AGV workspace is defined as a physical area where obstacles are mapped to the parts of the workspace without collision and the AGV is allowed to move between obstacles. Assuming that the AGV has access to environmental information about its workspace, let α and θ be the angles between the body and the X-axis and the steering wheel, respectively. Assuming that the two wheels are parallel, L is the wheelbase, and the distance traveled by the AGV in the static obstacle at the adjusted speed represents the shortest path between points, which follows the given formula.
[0138]
[0139] In the formula, α and θ are the angle between the vehicle body and the X-axis and the steering wheel angle respectively; and The motion coordinates of the i-th point; and The motion coordinates of the i+1th point, x, y represent the horizontal and vertical movement components respectively; dist is the distance path between the points; is the minimum function; n is the total number of moving points, and v represents the velocity within time t.
[0140] The direction of motion of the AGV body is affected by many physical factors such as gravity acceleration, running speed, and vehicle mass. In addition, in the dynamic environment of workspace obstacles, the speed of the AGV is limited by factors such as curvature, turning radius, maximum turning speed, static friction coefficient, and minimum radius. The AGV steering follows the following steps under motion constraints. Therefore, the AGV running speed is constrained by the acceleration and speed minimum radius limits, and the formula is as follows.
[0141]
[0142] and are the movement speed and the maximum movement speed respectively; u and g are the static friction coefficient and the gravity acceleration parameter respectively; m is the vehicle mass; R and k are the measured length of the deployment area and the turning curvature respectively; L is the wheelbase, Represents the minimum centripetal force.
[0143] The systems, devices, modules or units described in the above embodiments are specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer is, for example, a personal computer, a laptop computer, a tablet computer, a wearable device, or a combination of any of these devices.
[0144] Computer readable media includes permanent and non-permanent, removable and non-removable media that implement information storage by any method or technology. Information is computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information accessed by a computing device. As defined herein, computer readable media does not include transitory media such as modulated data signals and carrier waves.
[0145] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
Claims
1. A working method of a finished product delivery automation system with AGV path planning, characterized in that: The finished product delivery automation system includes: PLC control system, transfer trolley, gantry, trolley, logistics vehicle and AGV. The PLC control system uses AGV path planning to control the AGV and realize the movement of the AGV between the transfer trolley, gantry, and empty trolley docking station device. The working method is specifically as follows: A1. Logistics vehicles enter the platform workstation; A2. Move the goods from the logistics vehicle to the empty trolley at the platform, and move the goods from the fully loaded trolley to the logistics vehicle; A3. Transport the empty trolley to the AGV docking station; A4. The barcode reading system automatically collects and analyzes the trolley to select the production line, and the AGV transports the empty trolley to the corresponding production line to connect to the gantry workstation; A5. The transfer cart logically takes empty boxes from the empty trolley to the corresponding empty box aisle for returning to the warehouse, and logically takes full boxes from the corresponding finished product aisle to the corresponding trolley aisle; A6, AGV transports a full trolley to the docking chain station on the platform; A7. The AGV exits the workstation, the chain device starts, and pulls the full trolley to the platform level; A8. Push the full vehicle into the docking station of the logistics vehicle; A9. Wait for the next logistics vehicle to ship; Among them, AGV path planning includes the following steps: S1. Initialization: Initialize the multi-objective particle swarm algorithm with a particle swarm in the desired region space. Each particle represents a potential solution for the AGV path. Generate a dynamic speed window based on the function of the AGV. For each particle in the swarm, perform the following steps: S1.1, using a multi-objective particle swarm algorithm to optimize the speed selection within the dynamic window according to multiple objectives; S1.2, applying the dynamic window method to select the highest scoring trajectory for the AGV based on the optimized speed; S2, using a multi-objective particle swarm algorithm to update the position and velocity of the particles; repeatedly generate new solutions and check fitness until the termination condition is met, the maximum number of iterations is met, or convergence occurs; S3. Select the best solution from the group as the best path for the AGV.
2. The working method of the finished product delivery automation system with AGV path planning according to claim 1 is characterized in that The step A3 is: A3.
1. Manually push the empty trolley to complete the material information labeling production line; A3.2, the empty trolley arrives at the AGV docking station; A3.
3. The code reading mechanism above the docking station identifies the corresponding information of the empty trolley production line; A3.4, the empty trolley at the docking station is in the AGV waiting for transportation; A3.
5. The AGV moves to the docking station, lifts the lifting panel, and the AGV leaves the station with the empty trolley; A3.6, AGV automatically sends to the corresponding production line according to the information; A3.
7. AGV transports the empty trolley and the work is completed.
3. The working method of the finished product delivery automation system with AGV path planning according to claim 1 is characterized in that The step A4 is: A4.
1. The AGV transports the empty trolley to the workstation, and the AGV confirms the arrival and the positioning detection device is started; A4.
2. The AGV lifting panel descends and exits the workstation, and the gantry mechanism starts; A4.
3. The empty trolley and the transfer trolley are on the same level and are ready for operation; A4.
4. The gantry reads the barcodes to identify each empty container return lane and the finished product pickup information; A4.
5. The gantry mechanism opens the empty trolley aisle in sequence according to the design logic, and the transfer trolley delivers empty boxes and takes full boxes in sequence according to the logical order. After the transfer trolley delivers two finished products to the trolley each time, the gantry closes the guardrail device in sequence; synchronously: the transfer trolley is aligned with the finished product aisle, the signal automatically recognizes the cylinder top rises, and the baffle of the finished product aisle is lowered. After taking all the finished products, the baffle automatically rises to block the last finished product in the aisle; A4.
6. After the transfer trolley completes the task, it performs tasks at other workstations; A4.
7. The gantry workstation is full of carts and is waiting for AGV to transport them.
4. The working method of the finished product delivery automation system with AGV path planning according to claim 1 is characterized in that The step A5 is: A5.
1. The empty trolley is waiting for the transfer trolley to pick up the goods; A5.
2. The gantry mechanism robot opens the first trolley guardrail and pushes the empty container onto the transfer trolley; A5.
3. The transfer trolley transports the empty container to dock with the empty container cargo lane and rises to the docking height at the same time; A5.
4. The transfer trolley delivers the empty container to the empty container aisle; A5.
5. The transfer trolley runs to the finished product lane, picks up the goods and delivers them to the trolley. The transfer trolley delivers empty boxes and picks up full boxes in a logical order. A5.
6. After the transfer trolley delivers two finished products to the trolley each time, the gantry closes the guardrail device in turn; A5.7, and so on, the transfer trolley completes the task of emptying and filling; A5.
8. The transfer cart completes the task and executes the task at another workstation.
5. The working method of the finished product delivery automation system with AGV path planning according to claim 4 is characterized in that The step A5.2 is: A5.2.
1. AGV carries the empty trolley to the gantry station; A5.2.2, gantry workstation mechanism starts; A5.2.
3. The gantry robot moves horizontally to the designated guardrail position; A5.2.
4. The robot moves horizontally to open the lock pin; A5.2.
5. The robot pushes the empty box to the transfer cart; A5.2.6, the robot mechanism waits for the transfer trolley to arrive; A5.2.
7. The robot completes the collection of the first empty box and then begins to open the guardrail of the next empty box on the empty trolley; A5.2.
8. The transfer trolley delivers empty boxes and picks up full boxes in a logical order. After the transfer trolley delivers two finished products to the trolley each time, the gantry closes the guardrail device in turn.
6. The working method of the finished product delivery automation system with AGV path planning according to claim 4 is characterized in that The step A5.5 is: A5.5.
1. The transfer trolley arrives at the corresponding cargo lane and is in the material-retrieving state; A5.5.
2. The system starts the finished product baffle aisle device, and the cylinder lifts up to open the baffle; A5.5.
3. The system confirms that the end baffle of the finished product aisle is fully opened; A5.5.
4. The electric roller at the end of the finished product aisle starts to deliver the finished products to the transfer trolley; A5.5.
5. The transfer trolley is logically transported to the corresponding trolley recovery channel position; A5.5.
6. The electric roller of the transfer trolley is started and the finished product is sent into the trolley; A5.5.
7. The transfer cart starts a task of delivering an empty container and picking up a full container; A5.5.
8. The cylinder lifts and pushes the baffle, and the transfer trolley is ready to take the finished goods; A5.5.
9. The cylinder lift baffle mechanism is equipped with a metal proximity sensor to ensure that the baffle is open. In extreme cases, if it is not opened, an alarm will be triggered to remind manual processing.
7. The working method of the finished product delivery automation system with AGV path planning according to claim 4 is characterized in that The step A5.6 is: A5.6.
1. The transfer trolley delivers the finished product to the trolley cargo lane to confirm that it is in place; A5.6.2, gantry workstation mechanism starts; A5.6.
3. The gantry robot moves horizontally to the designated guardrail position; A5.6.4, robot arm closes the lock pin; A5.6.
5. The robot mechanism opens the next empty box lane for trolleys; A5.6.
6. The transfer trolley delivers empty boxes and picks up full boxes in a logical order. After the transfer trolley delivers two finished products to the trolley each time, the gantry closes the guardrail device in turn.
8. The working method of the finished product delivery automation system with AGV path planning according to claim 1 is characterized in that The step A6 is: A6.
1. AGV transports a full trolley to the chain docking station; A6.
2. The AGV lifting panel descends, and the full vehicle falls into the slot—the AGV rotates and exits from the side; A6.
3. The fully loaded trolley falls smoothly into the chain slot; A6.
4. The chain device starts, the fully loaded vehicle smoothly rises to the platform level, moves along the platform slot, and automatically decelerates and stops.
9. The working method of the finished product delivery automation system with AGV path planning according to claim 1 is characterized in that The multi-objective particle swarm optimization algorithm includes: M1. Initialization: Randomly generate the positions and velocities of a group of particles in the search space; set the personal best position and fitness of each particle at the initialization position to the initial value; the particle uses its best fitness to identify the global best position; M2. Update the speed and position of particles: M3, evaluation, the objective function or fitness of each particle's new position is calculated by multiple objective functions, each of which represents a different evaluation goal; Optimal AGV path planning is performed through a multi-objective particle swarm in a dynamic window method technique; when managing multiple objectives of a multi-objective particle swarm, a particle swarm with different positions and velocities represents a potential solution; particles traverse the search space to find the best answer that optimizes the specified objective; initialize randomly initialized particle positions and velocities, evaluate the objectives established for AGV path planning, evaluate the fitness of each particle, update the individual best and global best based on the individual best position of each particle, update the global best position using the individual and collective best proxy solutions, update the velocity and position of the particle, and whether the termination criteria are met; M4, Update individual and global optimal values: By comparing the fitness value of each particle target with its individual optimal fitness, if the current fitness is better for any target, then update the individual optimal position and fitness value of that objective function, by comparing the fitness value of the particle with whether the current fitness is better for any objective function, and updating the global optimal position and fitness of that target, calculate the global optimal fitness of each target; M5. Dominance calculation: Dominance calculation is performed by comparing particles to determine dominance relationships; dominance levels are assigned to particles based on their dominance calculation relationships; weights are dynamically adjusted based on the distribution of solutions along the dominance frontier according to the performance of the particles and the characteristics of the optimization problem; M6, non-dominated sorting: particles entering different non-dominated fronts are sorted according to their dominance levels; M7, selection and reproduction: select particles from the non-dominant frontier according to the dominance level and crowding distance of the particles; combine the selected particles through crossover and mutation operations to reproduce new particles; M8, Termination condition: This step is performed by checking whether the termination condition is met, the maximum number of iterations is reached, or the desired goal is reached; if the termination condition is met, the algorithm is stopped; otherwise, return to M2.
10. The working method of the finished product delivery automation system with AGV path planning according to claim 1, characterized in that The details of the dynamic window algorithm are as follows: D1. Perception and positioning: Get sensor readings to perceive the environment and estimate the AGV’s current position and orientation; D2. Motion prediction: predict the future motion of the AGV based on the current state of the AGV and the dynamic model; consider the speed, acceleration and steering ability of the AGV; D3, Dynamic Window Generation: Define a dynamic window to represent the feasible speed and steering combinations of the AGV; set the maximum and minimum limits of the linear speed and angular velocity according to the capabilities of the AGV; D4, trajectory evaluation: Generate a set of candidate trajectories by sampling the speed within a dynamic window; evaluate each trajectory according to different criteria; assign a score or cost to each route according to the evaluation criteria; D5, trajectory selection: select the trajectory with the lowest cost or the highest score as the desired trajectory of the AGV; ensure that the selected trajectory is collision-free and satisfies the dynamic constraints of the AGV; D6, motion execution: convert the desired trajectory into control commands for the AGV actuator; execute the control commands to make the AGV move along the desired trajectory; D7, loop: Repeat D1 to D6 continuously, update the AGV's motion plan according to the changing environment and AGV status; adjust the dynamic window and re-evaluate the trajectory to adapt to the dynamic characteristics of the field.
Citation Information
Patent Citations
Storage and distribution integration system and method based on insert-type AGV
CN104555222A
Logistics AGV navigation control method and system based on machine vision
CN117093009A
