Plant factory automation logistics collaborative operation system and scheduling method

Abstract

This invention provides an automated logistics collaborative operation system and scheduling method for plant factories. The system includes a three-dimensional cultivation storage area, a horizontal shuttle subsystem, a vertical transfer subsystem, a central processing station subsystem, and an intelligent scheduling and control station, forming a relay-style logistics topology combining multi-level shuttles and stacker cranes. The multi-level shuttles are responsible for horizontal storage and retrieval of overall cultivation units between cultivation rack layers; the stacker cranes are responsible for vertical collection and layer changing at the ends of the aisles. The two systems achieve spatiotemporal coordination through the intelligent scheduling and control station. Lateral flexible clamping mechanisms and anti-drip transport modes ensure the safety and cleanliness of live crops during high-speed transport. Combined with scheduling algorithms, the entire process from sowing, germination, and transplanting to harvesting and cleaning is automated, significantly improving the production efficiency and logistics throughput of plant factories and solving the problem of efficient and stable transport of massive cultivation units in high-density plant factories.

Description

Technical Field

[0001] This invention relates to the field of intelligent warehousing and agricultural automation technology, specifically to an automated logistics collaborative operation system and scheduling method for plant factories. Background Technology

[0002] As plant factories expand in scale, especially towards 12 or higher floors, logistics costs have become a key bottleneck restricting their commercialization. Traditional plant factories rely heavily on manual handling or simple conveyor belt systems. However, when dealing with ultra-large factories with a single area of ​​4502㎡ and 12 floors, manual handling is not only inefficient but also poses significant safety hazards due to working at heights. More importantly, pathogens brought in by personnel entering and leaving the clean production area are one of the main causes of crop yield reduction.

[0003] While existing industrial automated logistics solutions (such as AS / RS automated storage and retrieval systems) are mature, they are not well-suited for direct application in plant factories. The unique characteristics of transporting live organisms: Industrial products are rigid, while plants are flexible and living. Severe vibrations during transport are unacceptable, as they can damage the roots or leaves. Furthermore, hydroponic plants must be moved along with their nutrient solution tanks, posing a risk of liquid spillage that could contaminate the equipment and the plants below.

[0004] The complexity of the operational process: Ordinary warehousing typically involves only two steps: "inbound" and "outbound." Plant factories, however, involve an extremely complex cyclical process, including sowing, germination in storage, transplanting (potentially multiple times), re-entry for growth, harvesting, and empty tray cleaning and return to storage. This high frequency of inventory turnover places extremely high demands on the throughput capacity of the logistics system.

[0005] Environmental adaptability: Plant factories are characterized by high humidity (often between 70% and 90%) and are filled with corrosive gases (such as ozone used for disinfection or volatiles from acidic nutrient solutions), making ordinary logistics equipment prone to rust and damage.

[0006] To address the aforementioned pain points, companies like Sananbio have launched automation solutions such as Uplift, but further improvements are still needed to refine the details of multi-device collaborative scheduling in 12-layer ultra-high-resolution architectures. Currently, no descriptions or reports of technologies similar to this invention have been found, and no similar domestic or international materials have been collected. Summary of the Invention

[0007] To address the aforementioned shortcomings in the prior art, this invention provides an automated logistics collaborative operation system and scheduling method for plant factories.

[0008] According to a first aspect of the present invention, an automated logistics collaborative operation system for plant factories is provided, comprising: The three-dimensional cultivation storage area, which serves as the core area for plant growth, includes multiple rows and layers of cultivation racks; each layer is also equipped with guide rails, forming cultivation rack aisles for the operation of the horizontal flow rotor system. The tunnel transfer subsystem is set at the end of the cultivation rack tunnel as a vertical operating mechanism between the ground and any floor height, and is used to exchange plants with the central processing station subsystem in the vertical space. The horizontal flow rotor system includes shuttles configured in each cultivation layer as horizontal operating mechanisms for exchanging plants with the tunnel transfer subsystem in the horizontal space; at the same time, the system uses an on-board vision detection module to collect crop growth images in real time and send them to the intelligent dispatch control station. The central processing station subsystem, serving as the task execution center at the ground level, is connected to the tunnel transfer subsystem via a ground transport network to transport plants to the corresponding task execution units for subsequent task operations. The intelligent dispatch control station, as the control center of the ground layer, monitors the plant growth status by acquiring crop growth images collected by the horizontal flow rotor system, and generates corresponding assignment schemes to coordinate the actions of the above subsystems; or, based on the logistics task instructions given by the superior production management system, it generates corresponding assignment schemes to coordinate the actions of the above subsystems.

[0009] Preferably, the high-density cultivation rack is equipped with cultivation troughs and planting boards, forming cultivation units with the crops planted on them.

[0010] Preferably, the tunnel transfer subsystem uses a stacker crane or vertical lift equipped with a lifting platform to connect the plants delivered by the shuttle cars on each floor and transport them to the ground or other floors.

[0011] Preferably, the shuttle has an omnidirectional wheel set at the bottom of its chassis and side clamping mechanisms that can extend laterally. The end of the clamping mechanism has a slot that matches the bottom of the cultivation rack for horizontally hooking out the cultivation rack.

[0012] Preferably, the shuttle vehicle has a height of 120-150 mm, and the height of the shuttle vehicle is not greater than one-quarter of the net height of a single-layer cultivation rack in the three-dimensional cultivation area.

[0013] Preferably, the interior of the shuttle car adopts a layout structure in which a flat servo motor array is driven and connected to a horizontal transmission gearbox.

[0014] Preferably, the shuttle vehicle is provided with an active servo leveling push rod and an on-board six-axis inertial measurement unit at the chassis suspension or the root of the lateral clamping mechanism.

[0015] Preferably, the ground conveying network includes: a roller conveyor line and a chain conveyor; wherein, the roller conveyor line connects the tunnel transfer subsystem with each task execution unit, and the chain conveyor is configured on the roller conveyor line for connecting plants; The task execution unit includes: a seeder, a transplanter, a harvester, a washing machine, and a six-axis industrial robot unit; wherein: The seeder adopts a cross-line gantry structure and is suspended at the seedling starting end of the ground conveying network. After visually aligning with the holes of the planting board, it automatically adsorbs and precisely releases seeds into the planting substrate through changes in air pressure and negative pressure, realizing high-throughput unmanned continuous sowing. The transplanting machine uses a multi-axis industrial robotic arm module that is hung upside down or placed sideways next to the transplanting conveyor line. It is used to evaluate the physiological health characteristics of seedlings in real time through image algorithms, pick up healthy plants that meet the standards without damage, and accurately transplant them to the mature plant planting board according to the preset extended plant spacing, while automatically removing inferior seedlings. The harvester is installed on a frame that runs straight through the ground roller conveyor network. It is used to cut the cultivation unit containing mature crops flush with the base of the plant stem, send the edible part to the packaging station, and output the waste root tray to the washing machine. The cleaning machine forms a closed loop with the ground conveying network, and is used to perform high-pressure washing and high-temperature disinfection on the harvested plants and / or empty cultivation troughs and planting boards after harvesting. After being reassembled by the automatic tray assembly machine, the plants are sent back to the three-dimensional cultivation warehouse for use by the tunnel transfer subsystem. The six-axis industrial robot unit is arranged on the side of the transplanting line and the packaging line. In conjunction with the vision recognition system, it performs precise transplanting of plants, removal of residual leaves, and packing of finished products.

[0016] Preferably, the system further includes: an independent stacker crane located between the three-dimensional cultivation storage area and the central processing station subsystem, forming a buffer storage area for temporarily storing cultivation units to be processed.

[0017] According to a second aspect of the present invention, a scheduling method for an automated logistics collaborative operation system in a plant factory is provided, comprising: The intelligent dispatch control station calculates the optimal path and equipment assignment scheme based on the task instructions given by the superior production management system or based on monitoring information. According to the assignment scheme, the corresponding shuttle car picks up the designated cultivation rack along the cultivation rack alley and transports it to the alley transfer subsystem at the end of the cultivation rack alley for plant handover. The tunnel transfer subsystem transports the plants to the ground and then through the ground transport network to the task execution unit specified in the assignment plan, where they are carried out corresponding task operations. After the task is completed, the new cultivation units and / or cleaned cultivation racks are transferred to the designated location through the ground conveying network, the tunnel transfer subsystem, and each shuttle car, thus completing the logistics scheduling.

[0018] Preferably, the intelligent scheduling and control station calculates the optimal path and equipment assignment scheme based on task instructions given by the superior production management system, or based on monitoring information, including: Acquire spatial location information of a designated cultivation rack within the three-dimensional cultivation area and real-time plant growth status data. The growth status data includes at least: canopy projection area, leaf color index, and cumulative light intensity obtained based on environmental sensors. Based on the real-time growth status data, calculate the optimal task operation time window for each cultivation rack. Based on the spatial location information of the designated cultivation racks within the three-dimensional cultivation area, a path optimization calculation model based on genetic algorithm or reinforcement learning is adopted to minimize the total travel time and energy consumption of the shuttle and stacker crane. The optimal task operation time window and the real-time plant growth status data are introduced as constraints to generate a scheduling path. The system detects whether there is time overlap on the same lane node in the generated scheduling path; if so, it calculates the scheduling urgency based on the difference between the real-time plant growth status data and the optimal task operation time window, and prioritizes the conflicting scheduling paths according to the scheduling urgency to obtain the optimal path and equipment assignment scheme.

[0019] Preferably, the above method further includes: The intelligent scheduling and control station adopts a dynamic storage location management mechanism, which dynamically adjusts the storage location of plants in the automated warehouse according to the plant growth stage and environmental needs, in order to achieve optimal matching of light and heat resources.

[0020] Preferably, the above method further includes: The intelligent dispatch control station employs anti-drip closed-loop control logic to prevent liquid spillage from the shuttle. This anti-drip closed-loop control logic is achieved through the coordinated action of an active servo leveling push rod installed at the chassis suspension or the root of the lateral clamping mechanism of the shuttle, and an onboard six-axis inertial measurement unit. When transporting cultivation units containing nutrient solution, the intelligent dispatch control station limits the acceleration and deceleration amplitudes of the shuttle and the stacker crane in real time. Simultaneously, during the dynamic acceleration or deceleration phase of the shuttle, the intelligent dispatch control station calculates the compensation angle based on the instantaneous acceleration command and controls the active servo leveling push rod to extend and retract synchronously, causing the cultivation unit carrying fluid to generate a small reverse compensation angle to actively balance the dynamic tilt of the internal liquid surface.

[0021] By adopting the above technical solution, the present invention has at least one of the following beneficial effects compared with the prior art: The automated logistics collaborative operation system and scheduling method for plant factories provided by this invention adopts a collaborative relay mode in which shuttle cars run horizontally and stacker cranes run vertically. The shuttle cars are extremely flexible inside the cultivation racks, able to reach the innermost storage locations; the stacker cranes leverage their advantages of large vertical travel and heavy load capacity; the two exchange plants at a connecting platform at the end of the aisle. This combination is several times more efficient than using a stacker crane alone (which has limited fork extension depth) or an AGV alone (which cannot reach high levels).

[0022] The automated logistics collaborative operation system and scheduling method for plant factories provided by this invention define a standard logistics unit, namely an integrated circulation unit combining crops, cultivation troughs, and planting boards. Throughout the entire production cycle, except for the transplanting and harvesting stages, the crops never leave the cultivation trough where they grow. This integrated circulation unit not only avoids disturbing the root system but also solves the problem of transporting liquid-laden crops through the anti-drip design of the cultivation trough.

[0023] The automated logistics collaborative operation system and scheduling method for plant factories provided by this invention takes into account that plant growth is slow (measured in days) while machine operation is fast (measured in seconds). Therefore, a buffer zone is set up between the cultivation area and the manual operation area. The stacker crane first sends the pallets to be processed (such as those to be replanted) to the buffer zone for temporary storage. After accumulating a certain quantity, they are then sent to the production line in batches, realizing a dynamic buffering strategy like a reservoir. This reservoir design greatly balances the difference in cycle time between the front and back ends, improving equipment utilization. Attached Figure Description

[0024] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the overall architecture of the automated logistics collaborative operation system for plant factories according to a preferred embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the structure of the three-dimensional cultivation storage area in a preferred embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of the shuttle vehicle in a preferred embodiment of the present invention.

[0027] Figure 4 This is a flowchart illustrating the scheduling method of the automated logistics collaborative operation system for plant factories in a preferred embodiment of the present invention.

[0028] In the diagram, 1 represents the three-dimensional cultivation storage area, 2 represents the tunnel transfer subsystem, 3 represents the intelligent dispatch control station, 41 represents the ground conveying network, 42 ​​represents the seeder, 43 represents the transplanter, 44 represents the harvester, 11 represents the aluminum alloy guide rail, 12 represents the cultivation unit, 101 represents the chassis, 102 represents the omnidirectional wheel, 103 represents the lateral clamping mechanism, and 104 represents the active servo leveling push rod. Detailed Implementation

[0029] The embodiments of the present invention are described in detail below: These embodiments are implemented based on the technical solution of the present invention, and provide detailed implementation methods and specific operation processes. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention.

[0030] While existing industrial automated logistics solutions (such as AS / RS automated storage and retrieval systems) are mature, they are not well-suited for direct application to plant factories. This is due to several factors, including the unique characteristics of live animal transportation, the complexity of operational processes and environmental adaptability, as well as insufficient attention to the details of multi-device collaborative scheduling under a 12-layer ultra-high architecture.

[0031] To address the aforementioned issues, one embodiment of the present invention provides an automated logistics collaborative operation system for plant factories. This system adopts a zoned collaborative mode that features unmanned cultivation areas, robotic operation areas, and highly efficient transfer areas. By designing a multi-level system architecture, it realizes an automated logistics system suitable for high-density plant factories, based on the collaborative operation of shuttle cars and stacker cranes.

[0032] Specifically, such as Figure 1 As shown, the automated logistics collaborative operation system for plant factories provided in this embodiment may include: The three-dimensional cultivation storage area 1, serving as the core area for plant growth, includes multiple rows and layers of high-density cultivation racks; each layer is also equipped with high-precision aluminum alloy guide rails 11, forming the cultivation rack aisles for the operation of the multi-layer shuttle system. Figure 2 As shown.

[0033] The tunnel transfer subsystem 2 is a vertical operating mechanism located at the end of the cultivation rack tunnel, used to exchange plants with the central processing station subsystem in the vertical space.

[0034] The horizontal flow rotor system includes shuttles configured in each cultivation layer as horizontal operating mechanisms for exchanging plants with the tunnel transfer subsystem in the horizontal space; at the same time, the system uses an on-board vision detection module to collect crop growth images in real time and send them to the intelligent dispatch control station.

[0035] The central processing station subsystem, serving as the task execution center at the ground level, is connected to the tunnel transfer subsystem via a ground transport network to transport plants to the corresponding task execution units for subsequent task operations.

[0036] The Intelligent Dispatch Control Station (WCS) 3, as the control center of the ground layer, monitors the plant growth status by acquiring crop growth images collected by the horizontal flow rotor system, and generates corresponding assignment schemes to coordinate the actions of the above subsystems; or, based on the logistics task instructions given by the superior production management system (PMS), it generates corresponding assignment schemes to coordinate the actions of the above subsystems.

[0037] In some preferred embodiments, the aforementioned vertical cultivation storage area may further include: The high-density cultivation rack is equipped with cultivation troughs and planting boards, which together with the crops planted on them form cultivation units 12.

[0038] In some preferred embodiments, the aforementioned tunnel transfer subsystem may further include: Stacker cranes or vertical lifts equipped with lifting platforms are used to connect plants delivered by shuttle cars from each level and transport them to the ground or other levels.

[0039] In some preferred embodiments, the stacker crane described above may further include: The stacker crane is equipped with a lifting loading platform for receiving cultivation units delivered by multi-level shuttle cars and transporting them to the ground floor or other levels.

[0040] In some preferred embodiments, the above-described horizontal flow rotor system may further include: The shuttle vehicle has an omnidirectional wheel set (omnidirectional wheels 102) at the bottom of its chassis 101. Lateral clamping mechanisms 103 extend laterally from both sides of the vehicle. The ends of these clamping mechanisms have slots that fit into the bottom of the cultivation rack, used to horizontally hook the cultivation rack out. Figure 3 As shown.

[0041] In some preferred embodiments, the shuttle vehicle may further include: The shuttle's height is between 120 mm and 150 mm, and does not exceed one-quarter of the clearance height of a single-layer cultivation rack in the vertical cultivation area. The shuttle's interior employs a layout structure with a flat servo motor array and a horizontal transmission gearbox, achieving an ultra-thin body configuration while meeting the mechanical strength requirements for the extension and retraction of the lateral clamping mechanism, thereby maximizing the effective vertical planting density of the vertical cultivation area. Active servo leveling push rods 104 and onboard six-axis inertial measurement units (IMUs) are installed at the shuttle's chassis suspension or at the root of the lateral clamping mechanism. The Active Servo Leveling Push Rod 104 is a miniature servo electric linear actuator designed for low-profile shuttle applications (shuttle height ≤ 150mm). Its typical structure includes: a motor, employing a flat brushless DC servo motor (BLDC) with an axial length ≤ 20mm and a Hall encoder; a transmission mechanism using a planetary gear reducer and precision ball screw, with a reduction ratio of approximately 5:1 to 10:1, achieving millimeter-level displacement accuracy; and a telescopic rod made of stainless steel, with optional chrome plating, featuring a ball hinge or U-shaped fork at the front end for flexible connection to the bottom support plate of the cultivation unit. Built-in sensors include: a Hall position sensor for detecting the absolute position of the telescopic rod; and limit switches for protection at both ends. The housing is made of unibody aluminum alloy with an IP54 sealing rating, providing moisture and corrosion resistance. Installation is as follows: one end of the push rod is fixed to the base of the shuttle chassis or the lateral clamping mechanism via a hinge support, while the other end is hinged to the underside of the cultivation unit's support platform. When the shuttle is not carrying any cargo, the push rod is in the neutral position; during operation, it actively adjusts its extension and retraction based on IMU feedback. In a specific application example, the push rod can achieve a ±5mm extension and retraction stroke with a response time ≤50ms, effectively compensating for liquid surface tilt under typical acceleration (≤0.3g) of a plant factory shuttle. An onboard six-axis inertial measurement unit (IMU) is used to measure shuttle body data.

[0042] In some preferred embodiments, the aforementioned central processing station subsystem may further include: The ground transport network 41 includes: a roller conveyor line and a chain conveyor; wherein, the roller conveyor line connects the tunnel transfer subsystem with each task execution unit, and the chain conveyor is configured on the roller conveyor line for connecting plants; The task execution units of the central processing station include: a seeder 42, a transplanter 43, a harvester 44, a washing machine, and a six-axis industrial robot unit; among which: The automated seeder adopts a cross-line gantry structure and is suspended at the seedling starting end of the ground conveying network. It integrates an array of photoelectric sensing modules and an air suction seed discharge pipe controlled by a solenoid valve. After visually aligning with the holes of the planting board, it automatically adsorbs and precisely releases seeds into the planting substrate through changes in air pressure, achieving high-throughput unmanned continuous sowing. The vision-guided transplanting machine uses a multi-axis industrial robotic arm module that is hung upside down or placed sideways next to the transplanting conveyor line, and is equipped with a pneumatic flexible end gripper and a multispectral machine vision positioning system. It is used to evaluate the physiological health characteristics of seedlings in real time through image algorithms, to pick up healthy plants that meet the standards without damage, and to accurately transplant them to the mature plant planting board according to the preset extended plant spacing, while automatically removing inferior seedlings. The continuous harvester is installed on a frame that runs straight through the ground roller conveyor network. It is equipped with a reciprocating high-speed cutting blade assembly with an adaptive height adjustment mechanism and a canopy separation conveyor belt. It is used to cut the cultivation unit carrying mature crops flush from the base of the plant stem, send the edible part to the packaging station, and output the waste root tray to the washing machine. The cleaning machine, together with the ground conveying network, forms a closed loop and is used to perform high-pressure washing and high-temperature disinfection on harvested plants and / or empty cultivation troughs and planting boards after harvesting. After being reassembled by an automatic tray-assembly machine, the plants are transported back to the three-dimensional cultivation warehouse for use by the tunnel transfer subsystem. The six-axis industrial robot unit is positioned on the side of the transplanting and packaging lines. In conjunction with the vision recognition system, it performs precise transplanting of plants, removal of residual leaves, and packing of finished products.

[0043] In some preferred embodiments, the system further includes: An independent stacker crane located between the vertical cultivation storage area and the central processing station subsystem forms a buffer storage area for temporarily storing cultivation units awaiting processing.

[0044] Based on the automated logistics collaborative operation system for plant factories provided in the above embodiments of the present invention, an embodiment of the present invention also provides a scheduling method for the system.

[0045] Combination such as Figure 4 The schematic diagram shown illustrates the working principle of the automated logistics collaborative operation system for plant factories provided in this embodiment. The scheduling method may include: S1, the Intelligent Dispatch Control Station (WCS) calculates the optimal path and equipment assignment scheme based on the task instructions given by the superior Production Management System (PMS) or based on monitoring information. S2, according to the assignment plan, the corresponding shuttle car picks up the designated cultivation rack along the cultivation rack alley and transports it to the alley transfer subsystem at the end of the cultivation rack alley for plant handover. S3, the tunnel transfer subsystem transports the plants to the ground and then through the ground transport network to the task execution unit specified in the assignment plan, where they perform the corresponding task operations; S4. After the task operation is completed, the new cultivation unit and / or the cleaned cultivation rack are transferred to the designated location through the ground transport network, the tunnel transfer subsystem and each shuttle car, thus completing the logistics scheduling.

[0046] In some preferred embodiments, the above-mentioned S1, where the intelligent scheduling control station calculates the optimal path and equipment assignment scheme based on the task instructions given by the superior production management system, or based on monitoring information, may further include: S11, acquire the spatial location information of the designated cultivation rack in the three-dimensional cultivation area and the real-time growth status data of the plants. The growth status data includes at least: canopy projection area, leaf color index and cumulative light intensity acquired based on environmental sensors. S12, based on real-time growth status data, calculates the optimal task operation time window for each cultivation rack, such as the transplanting time window and the harvesting time window; S13, based on the spatial location information of the designated cultivation racks in the three-dimensional cultivation area, adopts a path optimization calculation model based on genetic algorithm or reinforcement learning, with the goal of minimizing the total travel time and energy consumption of the shuttle and stacker crane, and introduces the optimal task operation time window and real-time plant growth status data as constraints to generate a scheduling path; S14, detect whether there is time overlap on the same roadway node in the generated scheduling paths; if so, calculate the scheduling urgency based on the difference between the real-time plant growth status data and the optimal task operation time window, and prioritize the conflicting scheduling paths according to the scheduling urgency to obtain the optimal path and equipment assignment scheme. Further: S141, Time Window Conflict Detection: After each shuttle task generates a path, the time interval (time window) for the shuttle to enter and leave each alleyway node is predicted. When the time windows of two or more tasks overlap at the same node, it is determined as a potential conflict.

[0047] S142, Dynamic Priority Calculation: For each task, calculate its scheduling urgency. This value is based on the difference between the current time and the lower limit of the optimal task operation time window (e.g., if a plant is close to its optimal transplanting or harvesting window, its urgency increases). The smaller the difference, the higher the urgency.

[0048] S143, Conflict resolution strategy: High-urgency tasks are given priority and their paths remain unchanged; low-urgency tasks implement a waiting strategy (pausing in place for one time slice) or a detour strategy (replanning alternative paths, such as adjusting the order of nodes); if the waiting time exceeds a set threshold (such as 30 seconds), the system will force the replanning of the entire path of the task.

[0049] S144, Rolling Time Domain Optimization: Every fixed time interval (e.g., 500 milliseconds), the system re-detects time window conflicts for all incomplete tasks and dynamically adjusts their priorities. This approach avoids localized congestion caused by a single fixed scheduler.

[0050] This step implements a coordinated scheduling strategy that combines time windows, dynamic priorities, and limited waiting time, effectively avoiding congestion.

[0051] In existing automated storage and retrieval systems (AS / RS), standard genetic algorithms (GA) or reinforcement learning (RL) are typically used to solve problems such as the Traveling Salesman Problem (TSP) or the Vehicle Routing Problem (VRP) to achieve the shortest path and minimum energy consumption. However, in the automated logistics scheduling of plant factories, since the priority of plant entry and exit must depend on their vital signs, a bio-physical co-evaluation scheduling model is constructed in this step. The input of this model not only includes the coordinate space data of each storage location, but also must integrate big data fed back by the onboard vision detection module of the multi-level shuttle in real time, including the current canopy projection area, leaf color index, and cumulative light intensity obtained based on environmental sensors for each plant tray.

[0052] In the mathematical model of the algorithm execution, the system first calculates the optimal task operation time window required for each cultivation unit based on these physiological data, including: the transplanting time window and the harvesting time window. During the chromosome fitness evaluation phase of the genetic algorithm (or the reward function mechanism of reinforcement learning), if the generated logistics assignment scheme causes a tray of plants to miss the optimal light and heat conversion window, or causes seedlings to be incorrectly assigned to high-temperature zones, the model will impose a large penalty function. Therefore, in the iterative optimization process, the model's objective is no longer simply to minimize mechanical energy consumption, but rather to "maximize the global biological fitness of the entire plant population in the entire storage area while ensuring the efficient operation of logistics equipment without collisions or deadlocks." This step directly transforms agricultural biological parameters into a cross-domain fusion algorithm with operations research mathematical constraints, fully satisfying the adaptability of plants, a commodity with special attributes, in logistics scheduling.

[0053] In some preferred embodiments, the above method may further include: The intelligent dispatch control station employs a drip-proof closed-loop control logic to prevent liquid spillage from the shuttle. This drip-proof closed-loop control logic is achieved through the coordinated action of an active servo leveling push rod installed on the shuttle chassis suspension or at the base of the lateral clamping mechanism, and an onboard six-axis inertial measurement unit (IMU). When transporting cultivation units containing nutrient solution, the intelligent dispatch control station continuously limits the acceleration and deceleration amplitudes of the shuttle and stacker crane; the acceleration and deceleration amplitudes... The control envelope is defined based on hydrostatics. In the formula, It is the acceleration due to gravity. The maximum liquid level tilt angle is calculated based on the safe dry string height of the cultivation trough. Simultaneously, during the dynamic acceleration or deceleration phase of the shuttle, the intelligent scheduling and control station calculates the compensation angle based on the instantaneous acceleration command and controls the active servo leveling push rod to extend and retract synchronously, so that the cultivation unit carrying the fluid generates a small reverse compensation tilt angle α, in order to actively balance the dynamic tilt of the internal liquid surface and completely prevent the nutrient solution from overflowing.

[0054] Furthermore, when the shuttle accelerates or decelerates, inertia causes the nutrient solution level within the cultivation unit to tilt. To suppress surface fluctuations, the push rod actively pushes the cultivation unit to generate a small reverse tilt angle based on the instantaneous acceleration value, keeping the solution level relatively horizontal. The control and signal transmission relationship (closed-loop control flow) at this time is as follows: The intelligent dispatch control station (WCS) sends motion commands (such as target speed and acceleration curve) to the shuttle, and at the same time, sends the instantaneous acceleration command as a feedforward signal to the on-board controller.

[0055] The onboard six-axis inertial measurement unit (IMU) measures the actual vehicle attitude (pitch angle, roll angle) and three-axis acceleration of the shuttle in real time, and feeds the measured data back to the onboard controller.

[0056] The vehicle-mounted controller (integrated into the WCS or a standalone PLC) receives the acceleration feedforward value from the WCS and the measured attitude data from the IMU, calculates the compensation angle, and outputs PWM or analog signals to drive the motor of the active servo leveling push rod.

[0057] The active servo leveling push rod extends or shortens according to the control signal, pushing the bottom of the cultivation unit to achieve angle compensation.

[0058] In some preferred embodiments, the above method may further include: The intelligent dispatch and control station adopts a dynamic storage location management mechanism, which dynamically adjusts the storage location of plants in the automated warehouse according to the plant growth stage and environmental needs, so as to achieve optimal matching of light and heat resources.

[0059] The scheduling method provided in the above embodiments of the present invention involves a Production Management System (PMS) generating a production plan based on order requirements and decomposing it into a series of logistics task instructions. The Intelligent Scheduling Control Center (WCS) receives these instructions or calculates the optimal path and equipment assignment scheme based on its monitoring information. The WCS control station dispatches a multi-level shuttle to the seedling area to retrieve seedling trays that have reached the appropriate age and transports them to the end of the aisle; it also dispatches a stacker crane to lower the seedling trays onto the ground conveyor line; the conveyor line then sends the seedling trays into the automatic transplanter, completing the transplanting operation scheduling. After transplanting, the WCS control station dispatches a stacker crane to lift the transplanted mature plant trays to a designated cultivation layer; it then dispatches a shuttle on that layer to store the mature plant trays in an available storage location. The WCS control station dispatches the shuttle and stacker crane to transport mature crop trays to the harvesting and packaging machine, completing the harvesting operation scheduling. After cleaning, the empty trays are combined and then returned to the warehouse or buffer warehouse by the logistics system, completing the empty tray return scheduling.

[0060] The technical solution provided by the above embodiments of the present invention will be further described in detail below with reference to a specific application example.

[0061] In this specific application example, taking the actual needs of Qingmei Digital Plant Factory in Xuanqiao Town, Pudong New Area as an example, the technical solution provided by the above embodiments of the present invention is used to implement the corresponding automated logistics collaborative operation system and scheduling method.

[0062] In this specific application example, the constructed automated logistics collaborative operation system includes the following four levels: The vertical cultivation storage area is tiered, with the core area for plant growth consisting of multiple rows and layers of high-density cultivation racks. Each layer is equipped with high-precision aluminum alloy guide rails for the operation of the multi-layer shuttle.

[0063] The aisle transfer subsystem is located at the end of the cultivation rack aisles. For the 12-story elevated warehouse of the Qingmei Digital Plant Factory in Xuanqiao Town, Pudong New Area, this specific application example utilizes a heavy-duty stacker crane. The stacker crane moves up and down the towering aisles, acting as a "vertical elevator," responsible for exchanging plants between the ground and any floor level.

[0064] The horizontal flow rotor system consists of multiple layers, as described above. Unlike ordinary shuttles, this shuttle is equipped with a special lateral clamping mechanism. Instead of directly supporting the bottom of the plants, it uses laterally extending forks to precisely hook into specialized slots at the bottom of the cultivation rack. This design allows the shuttle body to be made very thin, thereby reducing the interlayer spacing and increasing planting density.

[0065] The central processing station level is the "brain" of the factory, located on the ground floor. It is connected to the tunnel transfer subsystem via roller conveyors and performs tasks through integrated seeders, transplanters, harvesters, washers, and six-axis industrial robot units.

[0066] The following section uses the plant replanting task as an example to illustrate the system's scheduling method.

[0067] The system scheduling method for this plant replanting task includes the following steps: Step 1, Instruction Issuance: The Intelligent Scheduling and Control Station (WCS) in the plant factory monitors that the seedlings in the nursery area have grown for 15 days and have reached the transplanting standard. The system automatically generates a transplanting task order.

[0068] Step 2, High-altitude Retrieval (Shuttle Operation): The multi-level shuttle located on the 8th floor (seedling layer) receives a wireless command and quickly departs from its standby position. It smoothly glides along the guide rails to the target storage location (e.g., row 8, column 32). After the shuttle comes to a stop, its metal forks extend laterally and precisely insert into the slots at the bottom of the seedling trays. As the forks retract, the seedling trays filled with tender green seedlings are smoothly pulled onto the back of the shuttle.

[0069] Step 3, Vertical Handover (Stacker Crane Operation): The shuttle, carrying seedling trays, travels to the end of the aisle. The massive stacker crane is already waiting there, its loading platform precisely positioned at the 8th floor level. The shuttle pushes the seedling trays onto the stacker crane's loading platform. After the handover is complete, the stacker crane, like a high-speed elevator, carries the seedling trays down to the ground floor's delivery port.

[0070] Step 4, Robotic Transplanting (Workstation Operation): The seedling trays leave the stacker crane and enter the ground-based roller conveyor line, where they are transported to the automated transplanting workshop. Here, a six-axis robot is already prepared. Its vision system (camera) quickly scans the seedling trays, identifying the height and leaf width of each seedling, instantly determining which are strong and which are weak. The robot's flexible grippers skillfully pick up the healthy seedlings and transplant them into the more spaced planting cups on the planting boards. The weak seedlings that are rejected are discarded.

[0071] Step 5, Return to Storage for Growth: The transplanted planting boards are loaded into clean cultivation troughs and assembled into new cultivation units. They are then transported back to the stacker crane via a conveyor line, lifted to the 5th layer (mature plant cultivation layer), and sent to a new storage location by a shuttle car on that layer to begin the next stage of growth.

[0072] In this specific application example, the entire process was seamless, requiring no human intervention. The shuttle, stacker crane, and robot formed a well-coordinated team, efficiently completing the demanding agricultural operations under the command of the WCS control station, thus realizing a true depiction of an unmanned plant factory.

[0073] The automated logistics collaborative operation system and scheduling method for plant factories provided in the above embodiments of the present invention define each cultivation unit in the three-dimensional cultivation storage area as a combination of "crop + planting board + cultivation trough"; the lateral telescopic clamping mechanism of the multi-layer shuttle car is designed with hooks or supports that match the bottom features of the cultivation trough, so as to realize non-contact (non-contact with the plant body) overall grasping of the cultivation unit.

[0074] The central processing station subsystem uses a six-axis industrial robot unit to achieve precise transplanting of seedlings, removal of residual leaves, and packing of finished products.

[0075] By setting up a separate cache library area, it is possible to balance the differences between the low-frequency rhythms of the crop growth cycle and the high-frequency rhythms of automated equipment operations.

[0076] The multi-level shuttle is equipped with an onboard vision inspection module, including a high-definition camera and a depth sensor, which can collect crop growth images (leaf area, plant height, leaf color) in real time during the inspection process and upload them to the intelligent dispatch and control station via wireless network for growth status analysis.

[0077] The cleaning machine forms a closed loop with the ground conveyor network; the harvested plants and / or empty cultivation troughs and planting boards after harvesting automatically enter the cleaning machine, and after high-pressure rinsing and high-temperature disinfection, they are reassembled by the automatic tray assembly machine and sent back to the three-dimensional cultivation warehouse for use by the stacker crane.

[0078] The WCS control station employs a path optimization strategy based on genetic algorithms or reinforcement learning to minimize the total travel time and energy consumption of shuttles and stacker cranes, while avoiding congestion and conflicts among multiple vehicles in the same lane.

[0079] The design incorporates anti-drip control logic to limit the acceleration and deceleration amplitudes of the shuttle and stacker crane when transporting cultivation tanks containing nutrient solution, and to control the vehicles to maintain a slight elevation angle during horizontal movement to prevent liquid spillage.

[0080] Design a dynamic storage location management mechanism to dynamically adjust the storage location of crops in the automated warehouse based on crop growth stages (seedlings occupy less space, mature plants occupy more space) and environmental requirements (temperature differences between different layers), thereby achieving optimal matching of light and heat resources.

[0081] Any matters not covered in the above embodiments of the present invention are well-known in the art.

[0082] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various modifications or variations within the scope of the claims, which do not affect the essence of the present invention.

Claims

1. An automated logistics collaborative operation system for plant factories, characterized in that, include: The three-dimensional cultivation storage area, which serves as the core area for plant growth, includes multiple rows and layers of cultivation racks; each layer is also equipped with guide rails, forming cultivation rack aisles for the operation of the horizontal flow rotor system. The tunnel transfer subsystem is set at the end of the cultivation rack tunnel as a vertical operating mechanism between the ground and any floor height, and is used to exchange plants with the central processing station subsystem in the vertical space. The horizontal flow rotor system includes shuttles configured in each cultivation layer as horizontal operating mechanisms for exchanging plants with the tunnel transfer subsystem in the horizontal space; at the same time, the system uses an on-board vision detection module to collect crop growth images in real time and send them to the intelligent dispatch control station. The central processing station subsystem, serving as the task execution center at the ground level, is connected to the tunnel transfer subsystem via a ground transport network to transport plants to the corresponding task execution units for subsequent task operations. The intelligent dispatch control station, as the control center of the ground layer, monitors the plant growth status by acquiring crop growth images collected by the horizontal flow rotor system, and generates corresponding assignment schemes to coordinate the actions of the above subsystems; or, based on the logistics task instructions given by the superior production management system, it generates corresponding assignment schemes to coordinate the actions of the above subsystems.

2. The automated logistics collaborative operation system for plant factories according to claim 1, characterized in that, The high-density cultivation rack is equipped with cultivation troughs and planting boards, which together form cultivation units with the crops planted on them.

3. The automated logistics collaborative operation system for plant factories according to claim 1, characterized in that, The tunnel transfer subsystem uses a stacker crane or vertical lift equipped with a lifting platform to connect the plants delivered by shuttle cars on each floor and transport them to the ground or other floors.

4. The automated logistics collaborative operation system for plant factories according to claim 1, characterized in that, The shuttle has an omnidirectional wheel set at the bottom of its chassis and side clamping mechanisms that can extend laterally. The end of the clamping mechanism has a slot that matches the bottom of the cultivation rack, which is used to hook the cultivation rack out horizontally.

5. The automated logistics collaborative operation system for plant factories according to claim 4, characterized in that, The height of the shuttle vehicle is 120-150 mm, and the height of the vehicle body is not greater than one-quarter of the net height of a single-layer cultivation rack in the three-dimensional cultivation area. The interior of the shuttle car adopts a layout structure that connects a flat servo motor array with a horizontal transmission gearbox. The shuttle is equipped with an active servo leveling push rod and an on-board six-axis inertial measurement unit at the chassis suspension or the root of the lateral clamping mechanism.

6. The automated logistics collaborative operation system for plant factories according to claim 1, characterized in that, The ground transport network includes: a roller conveyor line and a chain conveyor; wherein, the roller conveyor line connects the tunnel transport subsystem with each task execution unit, and the chain conveyor is configured on the roller conveyor line for connecting plants; The task execution unit includes: a seeder, a transplanter, a harvester, a washing machine, and a six-axis industrial robot unit; wherein: The seeder adopts a cross-line gantry structure and is suspended at the seedling starting end of the ground conveying network. After visually aligning with the holes of the planting board, it automatically adsorbs and precisely releases seeds into the planting substrate through changes in air pressure and negative pressure, realizing high-throughput unmanned continuous sowing. The transplanting machine uses a multi-axis industrial robotic arm module that is hung upside down or placed sideways next to the transplanting conveyor line. It is used to evaluate the physiological health characteristics of seedlings in real time through image algorithms, pick up healthy plants that meet the standards without damage, and accurately transplant them to the mature plant planting board according to the preset extended plant spacing, while automatically removing inferior seedlings. The harvester is installed on a frame that runs straight through the ground roller conveyor network. It is used to cut the cultivation unit containing mature crops flush with the base of the plant stem, send the edible part to the packaging station, and output the waste root tray to the washing machine. The cleaning machine forms a closed loop with the ground conveying network, and is used to perform high-pressure washing and high-temperature disinfection on the harvested plants and / or empty cultivation troughs and planting boards after harvesting. After being reassembled by the automatic tray assembly machine, the plants are sent back to the three-dimensional cultivation warehouse for use by the tunnel transfer subsystem. The six-axis industrial robot unit is arranged on the side of the transplanting line and the packaging line. In conjunction with the vision recognition system, it performs precise transplanting of plants, removal of residual leaves, and packing of finished products.

7. The automated logistics collaborative operation system for plant factories according to any one of claims 1-6, characterized in that, Also includes: An independent stacker crane located between the three-dimensional cultivation storage area and the central processing station subsystem forms a buffer storage area for temporarily storing cultivation units to be processed.

8. A scheduling method for an automated logistics collaborative operation system in a plant factory, characterized in that, include: The intelligent dispatch control station calculates the optimal path and equipment assignment scheme based on the task instructions given by the superior production management system, or based on monitoring information. According to the assignment scheme, the corresponding shuttle car picks up the designated cultivation rack along the cultivation rack alley and transports it to the alley transfer subsystem at the end of the cultivation rack alley for plant handover. The tunnel transfer subsystem transports the plants to the ground and then through the ground transport network to the task execution unit specified in the assignment plan, where they are carried out corresponding task operations. After the task is completed, the new cultivation units and / or cleaned cultivation racks are transferred to the designated location through the ground conveying network, the tunnel transfer subsystem, and each shuttle car, thus completing the logistics scheduling.

9. The scheduling method of the automated logistics collaborative operation system for plant factories according to claim 8, characterized in that, The intelligent scheduling and control station calculates the optimal path and equipment assignment scheme based on task instructions from the superior production management system, or based on monitoring information, including: Acquire spatial location information of a designated cultivation rack within the three-dimensional cultivation area and real-time plant growth status data. The growth status data includes at least: canopy projection area, leaf color index, and cumulative light intensity obtained based on environmental sensors. Based on the real-time growth status data, calculate the optimal task operation time window for each cultivation rack. Based on the spatial location information of the designated cultivation racks within the three-dimensional cultivation area, a path optimization calculation model based on genetic algorithm or reinforcement learning is adopted to minimize the total travel time and energy consumption of the shuttle and stacker crane. The optimal task operation time window and the real-time plant growth status data are introduced as constraints to generate a scheduling path. The system detects whether there is time overlap on the same lane node in the generated scheduling path; if so, it calculates the scheduling urgency based on the difference between the real-time plant growth status data and the optimal task operation time window, and prioritizes the conflicting scheduling paths according to the scheduling urgency to obtain the optimal path and equipment assignment scheme.

10. The scheduling method of the automated logistics collaborative operation system for plant factories according to claim 8, characterized in that, It also includes any one or more of the following: The intelligent dispatch control station adopts a dynamic storage location management mechanism, which dynamically adjusts the storage location of plants in the automated warehouse according to the plant growth stage and environmental needs, in order to achieve optimal matching of light and heat resources. The intelligent dispatch control station employs anti-drip closed-loop control logic to prevent liquid spillage from the shuttle. This anti-drip closed-loop control logic is achieved through the coordinated action of an active servo leveling push rod installed at the chassis suspension or the root of the lateral clamping mechanism of the shuttle, and an onboard six-axis inertial measurement unit. When transporting cultivation units containing nutrient solution, the intelligent dispatch control station limits the acceleration and deceleration amplitudes of the shuttle and the stacker crane in real time. Simultaneously, during the dynamic acceleration or deceleration phase of the shuttle, the intelligent dispatch control station calculates the compensation angle based on the instantaneous acceleration command and controls the active servo leveling push rod to extend and retract synchronously, causing the cultivation unit carrying fluid to generate a small reverse compensation angle to actively balance the dynamic tilt of the internal liquid surface.

Images