A robot-based pharmaceutical feeding intelligent control method and system

Abstract

The present application belongs to the technical field of industrial robot control, and particularly relates to a pharmaceutical feeding intelligent control method and system based on a robot, which comprises the following steps: obtaining the remaining effective time of a feeding task, determining a time gravity gain coefficient and a basic virtual gravity vector which dynamically change with time; monitoring the robot distribution in a feeding temporary storage area, calculating the reverse impedance force vector suffered by a robot which attempts to enter the temporary storage area; determining the dynamic traffic penetration rate of the robot according to the consumed time of the task and the real-time density of the temporary storage area; correcting the reverse impedance force vector based on the dynamic traffic penetration rate, combining the basic virtual gravity vector to generate a final navigation resultant force vector, and driving the robot to move. The present application realizes selective release of tasks which are close to overtime, prevents physical congestion deadlock, and effectively avoids feeding overtime and liquid waste caused by queuing and waiting.

Description

Technical Field

[0001] This invention relates to the field of industrial robot control technology. More specifically, this invention relates to a robot-based intelligent control method and system for pharmaceutical feeding. Background Technology

[0002] In the intelligent production system of modern pharmaceutical industry, the material feeding stage serves as a crucial hub connecting the storage system and the production reactor, and its operational efficiency directly affects the production quality of medicines. This is especially true in the field of traditional Chinese medicine, where the entire process of transporting materials from the warehouse to the feeding workshop via elevators and then into the reactor demands extremely high timeliness and accuracy. Limited by pharmaceutical process specifications and the physicochemical properties of the materials themselves, material feeding tasks typically face strict time windows. For example, materials must be fed into the reactor within a specific timeframe after leaving the warehouse; exceeding this time window will directly lead to the spoilage of the medicine or a production quality incident.

[0003] However, existing automated logistics systems face severe resource bottlenecks and scheduling challenges when handling such time-sensitive tasks. The entrance to the material handling workshop typically has only a limited number of temporary material storage locations. When multiple production lines generate material handling requests concurrently, a large number of mobile robots converge on the storage area, leading to high-density traffic congestion in localized areas. Existing technologies usually employ the Artificial Potential Field (APF) method for robot path planning and obstacle avoidance. This method works by setting up a virtual repulsive field based on physical distance to prevent collisions. However, this traditional repulsive field mechanism has the characteristic of indiscriminate repulsion. When a high-intensity repulsive field is generated in the storage area due to robot aggregation, this field will equally block all external robots attempting to enter the area. This mechanism causes robots performing urgent tasks, such as carrying critical materials nearing their expiration date, to be unable to overcome the repulsive force of the congested area, forcing them to queue outside the congestion zone, thus leading to serious problems of material handling delays and material losses. Summary of the Invention

[0004] To address the technical problem that the indiscriminate repulsion of existing artificial potential field methods leads to emergency tasks being unable to overcome congestion, resulting in material feeding delays and material scrapping, this invention provides solutions in the following aspects.

[0005] In a first aspect, the present invention provides a robot-based intelligent control method for pharmaceutical material dispensing, comprising: acquiring the remaining effective time of the dispensing task and the target coordinates of the dispensing port; determining the time gravity gain coefficient at the current moment based on the remaining effective time; and calculating the basic virtual gravity vector pointing towards the dispensing port on the robot based on the time gravity gain coefficient; monitoring the robot distribution in the dispensing temporary storage area in real time, constructing a local congestion repulsion potential energy field, and calculating the reverse resistance vector pointing towards the outside of the temporary storage area on the robot attempting to enter the temporary storage area based on the local congestion repulsion potential energy field; calculating the dynamic passage penetration rate of the robot based on the consumed time of the dispensing task and the real-time robot density in the temporary storage area; using the dynamic passage penetration rate as an attenuation factor to correct the reverse resistance vector; synthesizing the corrected reverse resistance vector with the basic virtual gravity vector to obtain the final navigation resultant force vector; and controlling the robot's driving direction and speed according to the final navigation resultant force vector to complete the dispensing task.

[0006] This invention constructs a time-gravity gain coefficient related to the remaining effective time, dynamically enhancing the basic virtual gravity vector pointing towards the feeding port over time, ensuring the robot obtains high-priority kinetic energy when the task is about to expire. It also constructs a local congestion repulsion potential energy field based on robot distribution and calculates the reverse impedance vector, maintaining the micro-circulation flow of the feeding temporary storage area through a physical-level rejection mechanism, preventing physical deadlock. Furthermore, it uses the dynamic passage penetration rate, which changes with the elapsed time and real-time robot density, as an attenuation factor to correct the reverse impedance vector, achieving dynamic adjustment of environmental impedance properties. This significantly attenuates the reverse impedance vector experienced by urgent tasks nearing expiration, allowing them to ignore congestion repulsion and achieve penetrating entry under the guidance of the final navigation resultant force vector. Ordinary tasks continue to follow queuing rules. This mechanism prevents congestion from spreading while selectively releasing urgent feeding tasks, effectively avoiding feeding timeouts and material waste caused by queuing, thus improving the overall safety and intelligence of the system.

[0007] Preferably, the time-gravitational gain coefficient satisfies the expression: In the formula, This represents the time-gravitational gain coefficient at the current moment. Indicates the remaining valid time for the material feeding task; This represents the preset gravitational reference constant; To prevent extremely small positive numbers with a denominator of 0.

[0008] This invention establishes a gain mechanism that dynamically adjusts with the urgency of time, so that the time gravity gain coefficient increases as the remaining time decreases, providing dynamic weights for the subsequent calculation of the basic virtual gravity vector. This ensures that the system can assign higher priority weights to the robot during critical time windows such as when the task is about to exceed the time limit, providing a parameter basis for avoiding material feeding timeouts.

[0009] Preferably, the basic virtual gravity vector satisfies the expression: In the formula, This represents the basic virtual gravitational vector acting on the robot at its current position, pointing towards the feed inlet; This represents the time-gravitational gain coefficient at the current moment. Indicates the robot's current coordinates; Indicates the target coordinates of the feeding port; This represents the vector pointing from the robot's current coordinates to the target coordinates of the feeding port; This represents the Euclidean distance between the robot's current coordinates and the target coordinates of the feeding port.

[0010] This invention constructs a basic virtual gravity vector that dynamically increases with the urgency of the task, so that the traction force on the robot is no longer constant, but increases dramatically as the deadline approaches. This gives the robot performing the urgent task a much higher tendency to approach the field kinetic energy, ensuring that it has the driving ability to overcome environmental resistance and quickly move to the target point.

[0011] Preferably, the construction of the local congestion repulsion potential energy field includes: treating each existing robot in the feeding temporary storage area as a Gaussian repulsion source, superimposing the virtual potential energy generated by all existing robots in the temporary storage area to construct a local congestion repulsion potential energy field; and measuring the local congestion repulsion potential energy experienced by a robot attempting to enter the temporary storage area at its current coordinates within the local congestion repulsion potential energy field. Satisfying the expression: , This indicates the total number of robots currently in the temporary storage area; This represents the maximum potential energy amplitude generated by a single robot; This indicates the current coordinates of the robot attempting to enter the temporary storage area; Indicates the number of items in the temporary storage area. The coordinates of an existing robot; This indicates that the robot attempting to enter the temporary storage area is in conflict with the robot already in the temporary storage area. Euclidean distance between existing robots; Indicates the radius of the repulsive force; This represents an exponential function with the natural constant as its base.

[0012] This invention constructs a local congestion repulsion potential energy field by superimposing the virtual potential energy generated by existing robots in the temporary storage area. This can transform the discrete robot distribution into a continuous spatial congestion repulsion potential energy distribution, providing a basis for the subsequent calculation of the reverse resistance force vector and preventing physical deadlock caused by excessive density.

[0013] Preferably, the step of calculating the reverse resistance vector pointing outward from the temporary storage area experienced by the robot attempting to enter the temporary storage area based on the local congestion repulsion potential energy field includes: calculating the negative gradient of the local congestion repulsion potential energy field to obtain the reverse resistance vector pointing outward from the temporary storage area experienced by the robot attempting to enter the temporary storage area.

[0014] This invention simulates the pressure repulsion effect in fluid mechanics, generating a reverse resistance vector pointing outward from the congested area. This reverse resistance vector can naturally block external robots from entering based on the degree of congestion, thereby forming a density-based rejection mechanism at the physical level, maintaining the micro-circulation flow within the temporary storage area, and preventing traffic paralysis caused by disorderly influx.

[0015] Preferably, the dynamic passage penetration rate satisfies the expression: In the formula, This indicates the current dynamic penetration rate of the robot. This indicates the time elapsed since the material was dispatched from the warehouse. This indicates the maximum time limit specified for the material feeding task; It is a positive smoothing factor; This indicates the current real-time robot density in the temporary storage area; This indicates the maximum design capacity density of the temporary storage area; This represents the natural logarithm function.

[0016] This invention introduces a dynamic passage penetration rate that grows non-linearly with the time elapsed in the task. By combining the time dimension with the spatial density dimension, it measures the robot's ability to traverse congested areas, causing the dynamic passage penetration rate to rise sharply as the task approaches its maximum time limit. This provides a calculable dynamic indicator for distinguishing between ordinary and emergency tasks, laying the algorithmic foundation for implementing differentiated passage strategies.

[0017] Preferably, the step of using the dynamic passage permeability as an attenuation factor to correct the reverse resistance vector includes: multiplying the reciprocal of the dynamic passage permeability by the reverse resistance vector to obtain the corrected reverse resistance vector.

[0018] Preferably, the final navigation resultant force vector satisfies the expression: In the formula, This represents the final resultant force vector for navigation, driving the robot's wheels to steer and accelerate. This represents the basic virtual gravity vector, pointing towards the feed inlet; This represents the reverse resistance force vector, pointing outwards from the temporary storage area; This indicates the dynamic penetration rate.

[0019] This invention constructs a final navigation resultant force vector that includes target-driven and environmental constraints by synthesizing a basic virtual gravity vector pointing towards the feeding port and a modified reverse impedance force vector. This allows the robot to follow the queuing rules of avoiding congestion in normal conditions, while in emergency conditions, strong gravity dominates and the weakened impedance force is ignored, enabling penetrating entry. This effectively solves the problem of feeding timeout in emergency tasks while ensuring the overall logistics order.

[0020] Preferably, the method for obtaining the current real-time robot density of the temporary storage area is as follows: calculate the total area of ​​the temporary storage area based on the geometric boundary information of the feeding temporary storage area; count the total number of robots currently located within the geometric boundary of the temporary storage area through the control system; calculate the ratio of the total number of robots to the total area of ​​the temporary storage area to obtain the current real-time robot density of the temporary storage area.

[0021] Secondly, the present invention provides a robot-based intelligent control system for pharmaceutical feeding, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned robot-based intelligent control method for pharmaceutical feeding is implemented.

[0022] By adopting the above technical solution, a computer program for the robot-based intelligent control method for pharmaceutical feeding is generated and stored in a memory for loading and execution by a processor. This allows for the creation of terminal devices based on the memory and processor, facilitating their use.

[0023] The beneficial effects of this invention are as follows: Firstly, by constructing a time-gravity gain coefficient related to the remaining effective time, the basic virtual gravity vector pointing towards the feeding port dynamically increases over time, ensuring that the robot can obtain high-priority kinetic energy when the task is about to expire. Secondly, by constructing a local congestion repulsion potential energy field based on robot distribution and calculating the reverse impedance vector, this invention maintains the micro-circulation fluidity of the feeding temporary storage area using a physical-level rejection mechanism, preventing physical deadlock. Thirdly, this invention uses the dynamic passage penetration rate, which changes with the consumed time and real-time robot density, as an attenuation factor to correct the reverse impedance vector, achieving dynamic adjustment of environmental impedance properties. This allows the reverse impedance vector experienced by urgent tasks nearing expiration to be significantly attenuated, thus enabling them to ignore congestion repulsion and achieve penetrating entry under the guidance of the final navigation resultant force vector. Ordinary tasks continue to follow queuing rules. This mechanism prevents congestion from spreading while selectively releasing urgent feeding tasks, effectively avoiding feeding timeouts and material waste caused by queuing, and improving the overall safety and intelligence level of the system. Attached Figure Description

[0024] Figure 1 This is a schematic flowchart illustrating a robot-based intelligent control method for pharmaceutical feeding according to the present invention.

[0025] Figure 2 This is a comparison diagram of the basic virtual gravitational strength and the corrected reverse resistance strength of robot A.

[0026] Figure 3 This is a schematic diagram of the final navigation resultant force field and robot path trajectory for robot A.

[0027] Figure 4 This is a comparison chart of the basic virtual gravitational strength and the corrected reverse resistance strength of robot B.

[0028] Figure 5 This is a schematic diagram of the final navigation force field and robot path trajectory for robot B. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0031] This invention discloses a robot-based intelligent control method for pharmaceutical feeding, referring to... Figure 1 This includes steps S1-S4.

[0032] S1. Obtain the remaining effective time of the feeding task and the target coordinates of the feeding port. Determine the time gravity gain coefficient at the current moment based on the remaining effective time, and calculate the basic virtual gravity vector pointing to the feeding port on the robot based on the time gravity gain coefficient.

[0033] It should be noted that in the pharmaceutical manufacturing process, the process from material leaving the warehouse to being added to the reactor has a strict time window, for example, it must be completed within 20 minutes, otherwise it will lead to waste of the pharmaceutical solution or quality accidents. However, existing automated guided vehicle (AGV) scheduling systems usually calculate gravity based on static path distance, ignoring the rapidly increasing urgency of the task over time. As a result, when the task is about to exceed the time limit, the robot's movement speed is still limited by conventional settings, which cannot meet the needs of emergency transportation. Therefore, this invention establishes a mapping relationship between the remaining time of the task and the intensity of virtual gravity. By constructing a virtual gravity field centered on the feeding port, the intensity of gravity dynamically increases as the remaining time of the task decreases. This gives the robot a much higher acceleration tendency than usual when the task is about to exceed the time limit, ensuring that it has high-priority kinetic energy to reach the target point.

[0034] Specifically, the warehouse control system obtains the remaining effective time of the feeding task and the target coordinates of the feeding port corresponding to the feeding task in real time.

[0035] To measure the urgency gain resulting from the passage of time, this invention obtains a time-gravitational gain coefficient that dynamically changes over time:

[0036] ;

[0037] In the formula, This represents the time-gravitational gain coefficient at the current moment. Indicates the remaining valid time for the material feeding task; This represents a preset gravitational reference constant used to adjust the fundamental amplitude of global gravity. Its dimension is momentum, and its unit is ______. , The scale is set by the implementers based on the workshop map scale, for example, set to 100. In other embodiments, the implementers can set the gravitational reference constant according to the actual implementation situation. To prevent extremely small positive numbers with a denominator of 0, such as 0.01, the remaining effective time... The decrease, The numerical value decreases, resulting in a decrease in the time-gravitational gain coefficient. Increase this, thus providing a greater gain value when time is tight.

[0038] Furthermore, calculate the fundamental virtual gravitational vector acting on the robot's current position:

[0039] ;

[0040] In the formula, This represents the basic virtual gravitational vector acting on the robot at its current position, pointing towards the feed inlet; This represents the time-gravitational gain coefficient at the current moment. Indicates the robot's current coordinates; Indicates the target coordinates of the feeding port; This represents the vector pointing from the robot's current coordinates to the target coordinates of the feeding port; The Euclidean distance between the robot's current coordinates and the target coordinates of the feeding port is represented. In this invention, when the robot's current coordinates coincide with the target coordinates of the feeding port, i.e., the Euclidean distance is 0, in order to avoid the denominator being 0 and causing calculation singularity, the basic virtual gravity vector is directly defined as the zero vector. This is the unit direction vector pointing to the target coordinates of the feeding port. The magnitude of the basic virtual gravity vector is directly affected by the time-gravitational gain coefficient. Modulation, when the mission is extremely urgent, the time-gravity gain coefficient The surge generates tremendous traction, giving the robot a strong driving tendency to point towards the feeding port.

[0041] S2. Monitor the distribution of robots in the feeding temporary storage area in real time, construct a local congestion repulsion potential energy field, and calculate the reverse resistance force vector pointing out of the temporary storage area to the robot attempting to enter the temporary storage area based on the local congestion repulsion potential energy field.

[0042] It should be noted that the material storage area in a pharmaceutical workshop is narrow and involves frequent interactions. Allowing an unlimited influx of robots would lead to physical deadlock. This invention constructs a congestion repulsion field based on the robot distribution density within a local space, simulating the pressure repulsion effect in fluid mechanics. When the robot density in the storage area increases, a virtual repulsive force pointing outward is generated, pushing away external robots attempting to enter, thereby maintaining the micro-circulation fluidity within the storage area and preventing excessive accumulation.

[0043] Specifically, the distribution of robots within the material storage area is monitored in real time. Each existing robot in the storage area is considered a Gaussian repulsive force source. The virtual potential energy generated by all existing robots in the storage area is superimposed to construct a local congestion repulsion potential energy field. The local congestion repulsion potential energy experienced by a robot attempting to enter the storage area at its current coordinates within this field satisfies the following expression:

[0044] ;

[0045] In the formula, This indicates that the robot attempting to enter the temporary storage area is experiencing localized congestion and repulsion at its current coordinates. This indicates the total number of robots currently in the temporary storage area; This represents the maximum potential energy amplitude generated by a single robot. Its dimension is energy, and its unit is 1000 kilometres. , The maximum potential energy amplitude is set by the implementer according to the obstacle avoidance safety requirements, for example, it is set to 50. In other embodiments, the implementer can set the maximum potential energy amplitude according to the actual implementation situation. This indicates the current coordinates of the robot attempting to enter the temporary storage area; Indicates the number of items in the temporary storage area. The coordinates of an existing robot; This indicates that the robot attempting to enter the temporary storage area is in conflict with the robot already in the temporary storage area. Euclidean distance between existing robots; This indicates the radius of the repulsive force, which corresponds to the robot's safety envelope. For example, it is set to 1.2 meters. In other embodiments, the implementer can set the radius of the repulsive force according to the actual implementation situation. This represents an exponential function with the natural constant as its base.

[0046] By taking the negative gradient of the repulsive potential field of the local congestion, we obtain the reverse resistance force vector generated by the congested area on the external robot:

[0047] ;

[0048] In the formula, This represents the reverse resistance force vector generated from the temporary storage area to the external robot, with its direction pointing outwards from the temporary storage area; This indicates the total number of robots currently in the temporary storage area; This represents the maximum potential energy amplitude generated by a single robot; Indicates the radius of the repulsive force; This indicates the current coordinates of the robot attempting to enter the temporary storage area; Indicates the number of items in the temporary storage area. The coordinates of existing robots are given. When the robots are more densely packed or closer together in the temporary storage area, the superimposed outward repulsive force is stronger, which physically manifests as blocking external robots from entering.

[0049] S3. Calculate the dynamic passage penetration rate of the robot based on the time consumed by the feeding task and the real-time robot density in the temporary storage area.

[0050] It should be noted that traditional distance-based repulsive fields have the drawback of indiscriminate rejection. That is, regardless of the urgency of the task being performed by the external robot, it will be blocked by the same repulsive force from the congested area. This prevents critical tasks that are about to expire from entering the feeding area, leading to serious production accidents. To resolve this contradiction, this invention introduces a dynamic passage penetration rate that increases non-linearly with the urgency of the task. This rate reflects the process of the environment's resistance to the robot decreasing. For tasks nearing their expiration date, the dynamic passage penetration rate increases sharply, making the original repulsive wall highly conductive for the robot, achieving a penetration-like entry effect. For ordinary tasks, the original blocking effect is maintained.

[0051] Specifically, for each robot attempting to enter the temporary storage area, calculate its corresponding dynamic passage penetration rate:

[0052] ;

[0053] In the formula, This indicates the robot's current dynamic passage penetration rate; the higher the value, the stronger its ability to penetrate congestion. This indicates the time elapsed since the material was dispatched from the warehouse. This indicates the maximum time limit specified for the feeding task, such as 20 minutes. In other embodiments, the implementer can set the maximum time limit according to the actual implementation situation. This represents a non-negative constraint operation on the remaining time, when the time already consumed... Less than the maximum time limit When the time is set to the actual remaining time, the value is the time already consumed. Greater than or equal to the maximum time limit When the value is 0, it is set to 0 to prevent the denominator from being negative or zero, which could lead to calculation errors, and to ensure that the permeability remains at its peak under the timeout condition; A positive smoothing factor is used to prevent the denominator from being zero and to limit the numerical overflow of the function at its limits. For example, 0.5 minutes. In other embodiments, implementers can set the smoothing factor according to the actual implementation situation. This indicates the current real-time robot density in the temporary storage area, and its value is the ratio of the total number of robots in the temporary storage area to the total area of ​​the temporary storage area. This indicates the maximum design capacity density of the temporary storage area, which is set by the implementers based on the site area, for example, 0.5 vehicles / square meter; Represents the natural logarithm function, used to calculate the natural logarithm. A nonlinear mapping is applied to smoothly adjust the gain effect of congestion on dynamic traffic penetration. This is done when time has elapsed. Approaching the maximum time limit When the denominator approaches This makes the gain term A sharp rise; Ensure basic penetration rate compensation is provided during periods of high congestion to prevent deadlock.

[0054] S4. The reverse resistance force vector is corrected by using the dynamic passage penetration rate as an attenuation factor. The corrected reverse resistance force vector is combined with the basic virtual gravity vector to obtain the final navigation resultant force vector. The robot's driving direction and speed are controlled according to the final navigation resultant force vector to complete the feeding task.

[0055] It should be noted that, in order to achieve differentiated impedance for tasks of different urgency levels in congested areas, this invention uses the calculated dynamic passage penetration rate as an attenuation factor for the reverse impedance force vector to modify the physical field. This significantly weakens or even ignores the repulsive force perceived by the emergency robot at the algorithm level, thus enabling it to pass through congested areas primarily by gravity.

[0056] Specifically, in the robot's navigation controller, the final resultant force vector for navigation is calculated:

[0057] ;

[0058] In the formula, This represents the final resultant force vector for navigation, driving the robot's wheels to steer and accelerate. This represents the basic virtual gravity vector, pointing towards the feed inlet, and its magnitude depends on the distance and remaining mission time. This represents the reverse resistance vector, which points outward from the temporary storage area, and its magnitude depends on the degree of congestion in the temporary storage area. This indicates the dynamic penetration rate.

[0059] This represents the corrected reverse resistance vector. When the task is in a normal state, the time consumed is relatively short, and the dynamic passage penetration rate is... The value is a constant close to 1, at which point the corrected reverse resistance force vector... When the congestion is significant, robots are subject to normal repulsive forces and are blocked outside the temporary storage area according to the first-come, first-served rule; when the task is in a critical emergency state, the time already consumed is close to the maximum time limit, and the dynamic passage penetration rate... The value is much greater than 1, resulting in a corrected reverse resistance force vector. The value is significantly attenuated to near zero. At this point, regardless of the reverse resistance vector generated by congestion in the temporary storage area... How large is it? For this particular robot, the net repulsive force it experiences is extremely small, and the final navigation resultant force vector is... Almost entirely composed of the basic virtual gravity vector pointing towards the feed port This allows the robot to ignore congestion and prioritize passing through congested areas to enter the feeding port.

[0060] The robot's direction and speed are controlled by the final navigation force vector to complete the feeding task.

[0061] For example, the target coordinates of the feeding port are located at The temporary storage area is located within a rectangular area bounded by X coordinates of 13 to 16 meters and Y coordinates of 7 to 14 meters, with a total area of ​​21 square meters. Currently, there are 8 other robots waiting or working in the temporary storage area, forming a high-density physical congestion. The current real-time robot density in the temporary storage area is... Maximum time limit for material feeding tasks Set to 20 minutes, gravitational reference constant Maximum design capacity density of the temporary storage area Smoothing factor .

[0062] Suppose that robot A has only just begun performing its task and has already consumed time. Minutes remaining, ample time, task in normal state, robot A's time gravity gain coefficient. At this point, the basic virtual gravitational force vector pointing towards the feed inlet is relatively small, maintaining normal cruising state; when robot A approaches the congestion boundary of the temporary storage area, the reverse resistance vector... The module is relatively long and points in the opposite direction to robot A, blocking its entry and limiting its dynamic passage penetration rate. The environmental resistance to robot A was hardly reduced, and the final resultant force vector for navigation was synthesized. The comparison diagram of the basic virtual gravitational strength and the corrected reverse resistance strength of robot A is shown below. Figure 2 As shown, within the congested area of ​​13 to 16 meters on the X-coordinate of the workshop, the reverse resistance strength curve is significantly higher than the basic virtual attraction strength curve, forming a hindrance zone. This means that the repulsive force dominates the direction of the final navigation resultant force vector. Robot A is forced to decelerate and stop or detour around the congested area, following the first-come, first-served queuing rule, and does not forcibly enter the temporary storage area, effectively avoiding physical deadlock. The final navigation resultant force field of robot A and the robot's path trajectory are shown in the diagram below. Figure 3 As shown, when robot A encounters congestion, its path diverges, and it cannot pass through in a straight line.

[0063] Suppose that due to delays in the preceding steps, robot B has already consumed time by the time it reaches the same position as robot A. Minutes are up, only 0.5 minutes remain before timeout, the mission is in a critical emergency state, and robot B's time gravity gain coefficient is... At this point, the basic virtual gravity vector is approximately 30 times higher than that of robot A, giving robot B a significant tendency to gain entry kinetic energy. With the physical environment unchanged, robot B and robot A have the same reverse resistance vector, resulting in a dynamic passage penetration rate. The environmental impedance becomes extremely transparent to robot B, and the final navigation resultant force vector is synthesized. The comparison diagram of the basic virtual gravitational strength and the corrected reverse resistance strength of robot B is shown below. Figure 4 As shown, at this point, the originally powerful reverse resistance vector is divided by The final navigation resultant force vector is almost zero, and the final navigation resultant force vector is completely dominated by the enhanced basic virtual gravity vector, forming a passage zone that runs through the congestion storage area. The final navigation resultant force field of robot B and the robot's path trajectory are shown in the diagram below. Figure 5 As shown, the final navigation resultant force vector points straight towards the feeding port. Robot B ignores the congestion signal in the temporary storage area and inserts into the gap in the traffic flow with the shortest straight path, achieving penetrating entry and ensuring that feeding is completed within the last 0.5 minutes to avoid waste of the liquid medicine.

[0064] It should be noted that this invention pertains to the top-level path planning layer, which is only responsible for generating the resultant force vector for navigation. In practical industrial applications, robots are equipped with independent local obstacle avoidance layers at the bottom level, such as LiDAR-based emergency stop mechanisms and safe distance detection modules. When the actual distance between the robot and an obstacle is less than a preset safety threshold, the bottom-level obstacle avoidance system immediately interrupts the upper-level planning instructions, triggering emergency braking to ensure physical safety. This layered control architecture ensures that even under high dynamic traffic penetration, the robot can remain physically safe, avoiding the risk of actual collisions.

[0065] This invention also discloses a robot-based intelligent control system for pharmaceutical feeding, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement a robot-based intelligent control method for pharmaceutical feeding according to this invention.

[0066] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

Claims

1. A robot-based intelligent control method for pharmaceutical feeding, characterized in that, include: Obtain the remaining effective time for the feeding task and the target coordinates of the feeding port; The time-gravitational gain coefficient for the current moment is determined based on the remaining effective time, satisfying: , This represents the time-gravitational gain coefficient at the current moment. Indicates the remaining valid time for the material feeding task. This represents the preset gravitational reference constant. To prevent extremely small positive numbers with a denominator of 0; The basic virtual gravitational vector pointing towards the feed port of the robot is calculated based on the time-gravity gain coefficient. The distribution of robots in the feeding temporary storage area is monitored in real time, a local congestion repulsion potential energy field is constructed, and the reverse resistance force vector pointing out of the temporary storage area is calculated based on the local congestion repulsion potential energy field. Based on the consumed time of the material feeding task and the real-time robot density in the temporary storage area, calculate the dynamic passage penetration rate of the robots, satisfying the following: , This indicates the current dynamic penetration rate of the robot. This indicates the time elapsed since the material was dispatched from the point of delivery. This indicates the maximum time limit specified for the material feeding task. As a positive smoothing factor, This indicates the current real-time robot density in the temporary storage area. This indicates the maximum design capacity density of the temporary storage area. Represents the natural logarithm function; The reverse resistance vector is corrected by using dynamic passage permeability as an attenuation factor, including multiplying the reciprocal of dynamic passage permeability with the reverse resistance vector as the corrected reverse resistance vector. The corrected reverse resistance force vector is combined with the basic virtual gravity vector to obtain the final navigation resultant force vector; the robot's direction and speed are controlled according to the final navigation resultant force vector to complete the feeding task.

2. The intelligent control method for pharmaceutical feeding based on a robot according to claim 1, characterized in that, The fundamental virtual gravity vector satisfies the expression: ； In the formula, This represents the basic virtual gravitational vector acting on the robot at its current position, pointing towards the feed inlet; This represents the time-gravitational gain coefficient at the current moment. Indicates the robot's current coordinates; Indicates the target coordinates of the feeding port; This represents the vector pointing from the robot's current coordinates to the target coordinates of the feeding port; This represents the Euclidean distance between the robot's current coordinates and the target coordinates of the feeding port.

3. The intelligent control method for pharmaceutical feeding based on a robot according to claim 1, characterized in that, The construction of the local congestion repulsion potential energy field includes: Each existing robot in the material storage area is considered a Gaussian repulsive force source. The virtual potential energy generated by all existing robots in the storage area is superimposed to construct a local congestion repulsion potential energy field. The local congestion repulsion potential energy experienced by a robot attempting to enter the storage area at its current coordinates within this local congestion repulsion potential energy field is... Satisfying the expression: , This indicates the total number of robots currently in the temporary storage area; This represents the maximum potential energy amplitude generated by a single robot; This indicates the current coordinates of the robot attempting to enter the temporary storage area; Indicates the number of items in the temporary storage area. The coordinates of an existing robot; This indicates that the robot attempting to enter the temporary storage area is in conflict with the robot already in the temporary storage area. Euclidean distance between existing robots; Indicates the radius of the repulsive force; This represents an exponential function with the natural constant as its base.

4. The intelligent control method for pharmaceutical feeding based on a robot according to claim 3, characterized in that, The calculation of the reverse resistance force vector pointing outward from the temporary storage area experienced by the robot attempting to enter the temporary storage area, based on the local congestion repulsion potential energy field, includes: By taking the negative gradient of the local congestion repulsion potential field, we obtain the reverse resistance force vector pointing outward from the temporary storage area experienced by the robot attempting to enter the temporary storage area.

5. The intelligent control method for pharmaceutical feeding based on a robot according to claim 1, characterized in that, The final navigation resultant force vector satisfies the expression: ； In the formula, This represents the final resultant force vector for navigation, driving the robot's wheels to steer and accelerate. This represents the basic virtual gravity vector, pointing towards the feed inlet; This represents the reverse resistance force vector, pointing outwards from the temporary storage area; This indicates the dynamic penetration rate.

6. The intelligent control method for pharmaceutical feeding based on a robot according to claim 5, characterized in that, The method for obtaining the current real-time robot density in the temporary storage area is as follows: The total area of ​​the temporary storage area is calculated based on the geometric boundary information of the material feeding temporary storage area; the total number of robots currently located within the geometric boundary of the temporary storage area is counted by the control system; the ratio of the total number of robots to the total area of ​​the temporary storage area is calculated to obtain the current real-time robot density of the temporary storage area.

7. A robot-based intelligent control system for pharmaceutical material dispensing, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a robot-based intelligent control method for pharmaceutical feeding according to any one of claims 1-6.

Images