A cleaning robot task scheduling method and system

Abstract

This invention belongs to the field of robot scheduling technology, specifically relating to a method and system for scheduling cleaning robot tasks. The method includes: determining the global competitiveness of a cleaning task based on its area and the distance between the cleaning robot and the task; determining an exclusionary tendency based on the global competitiveness and the distance between the cleaning robot and the task; determining a task allocation inhibition factor for the cleaning robot based on the exclusionary tendency of other cleaning robots; determining the total social attraction vector acting on the cleaning robot based on the task allocation inhibition factor and the distance between the cleaning robot and the task; superimposing this vector with the repulsion vector to obtain the total driving force of the cleaning robot; generating motion commands; and scheduling the cleaning robot to perform cleaning. This invention achieves intelligent and balanced allocation of cleaning tasks among multiple cleaning robots, improving the operational efficiency of the tipper system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot scheduling. More particularly, the present application relates to a cleaning robot work task scheduling method and system. BACKGROUND

[0002] In the bulk material transfer scene of mines, ports and the like, the dumper system as the key equipment to realize efficient unloading of the train compartment undertakes an important material transfer task. After the compartment is turned over for unloading, a part of the residual coal blocks will often adhere or freeze in the interior, especially in the corners and wallboards. These residual materials not only affect the subsequent loading efficiency, but also can cause the compartment to be overloaded or affect the safety of train operation. In order to ensure the complete emptying of the compartment and improve the operation efficiency of the dumper system, multiple cleaning robots are usually used to cooperatively enter the interior of the compartment for automatic cleaning operation.

[0003] In the traditional multi-robot cooperative path planning method, the artificial potential field method is widely used due to its simple calculation and good real-time performance. In the artificial potential field method, the movement of the robot in the workspace is virtually a movement in an abstract force field. The target point generates an attractive force on the robot, while the obstacle generates a repulsive force. The robot finally moves along the direction of the combined force of the attractive force and the repulsive force. In practical applications, the artificial potential field method is used to guide the movement of the cleaning robot in the interior of the compartment, so that it can avoid obstacles and move towards the area that needs to be cleaned.

[0004] However, the artificial potential field method mainly focuses on the physical distance relationship between the robot and the target point, and lacks consideration of the task competition characteristics in the multi-robot system. In the artificial potential field method, the attractive force is only related to the target attribute and the distance, which causes all cleaning robots to be attracted by the residual coal block area with the closest distance or the largest area, thereby causing the cleaning robots to rush to the same task area, resulting in local congestion. This congestion not only causes the cleaning robots to generate a strong repulsive force due to the close distance between them, which interferes with each other's cleaning path and even causes the movement to stop, but also causes the cleaning tasks in other areas of the compartment to be unattended, affecting the cleaning efficiency. SUMMARY

[0005] To solve the technical problems of uneven distribution of cleaning tasks and low efficiency caused by the clustering of multiple cleaning robots in the traditional artificial potential field method, the present application provides solutions in the following aspects.

[0006] In a first aspect, the present application provides a cleaning robot work task scheduling method, comprising:

[0007] Obtain the cleaning task information of the car interior; determine the global competitiveness of the cleaning task according to the area size of the cleaning task and the distance of each cleaning robot to the cleaning task; determine the exclusive tendency of each cleaning robot to the cleaning task according to the global competitiveness of the cleaning task and the distance of each cleaning robot to the cleaning task; determine the task allocation inhibition factor of the cleaning robot to the cleaning task according to the exclusive tendency of other cleaning robots to the cleaning task; determine the total social attractive force vector received by the cleaning robot according to the task allocation inhibition factor and the distance of the cleaning robot to the cleaning task; determine the repulsive force vector received by the cleaning robot from the obstacle, superimpose the total social attractive force vector and the repulsive force vector to obtain the total driving force of the cleaning robot, generate a motion instruction according to the total driving force, and dispatch the cleaning robot to clean.

[0008] The application identifies the position and size of each residual coal block area in the car, determines the global competitiveness of the cleaning task based on the area size of the cleaning task and the distance relationship, can identify the popular task that is easy to cause congestion, calculates the exclusive tendency of the cleaning robot to the cleaning task and the task allocation inhibition factor, integrates the social relationship between the cleaning robots into the path planning, automatically reduces the attraction of the cleaning task to the specific cleaning robot when the cleaning task is concerned by multiple cleaning robots, promotes the cleaning robot to turn to the cleaning task with less competition, and avoids the phenomenon that all cleaning robots rush to the same cleaning task in the traditional method; the application combines the social attractive force field and the repulsive force field, enables the cleaning robot to autonomously avoid collision while pursuing the cleaning task target, ensures the orderly distribution in the limited car space, enables multiple cleaning robots to efficiently cooperate in different areas of the car, improves the cleaning efficiency, and shortens the overall operation time.

[0009] Preferably, it further comprises: when the cleaning task is not selected within a continuous preset period, reducing the task allocation inhibition factor of all cleaning robots to the cleaning task to enhance the attraction of the cleaning task to the cleaning robot.

[0010] The application effectively solves the task deadlock problem that may occur in the cooperative cleaning of multiple cleaning robots by dynamically adjusting the task allocation inhibition factor, automatically reduces the task allocation inhibition factor of all cleaning robots to the task when the cleaning task is not selected by any cleaning robot within a continuous multiple period, ensures that even the most remote or smallest task can obtain sufficient attention, promotes the cleaning robot that may ignore the cleaning task to reconsider and select to execute the cleaning task, avoids the phenomenon that the popular task is excessively concerned and the unpopular task is ignored for a long time in the traditional multiple cleaning robot system, and guarantees the balanced allocation and timely completion of all cleaning tasks.

[0011] Preferably, the step of obtaining the cleaning task information inside the carriage includes: using a sensor array installed above the tipper structure to scan the inside of the carriage after it has been tipped over and returned to its original position, identifying the residual coal areas that need to be cleaned, obtaining the geometric center coordinates and area size of each residual coal area, and establishing a cleaning task for each residual coal area.

[0012] Preferably, the global competitiveness satisfies the expression: In the formula, For the first The overall competitiveness of a cleaning task; For the first The size of the area to be cleaned; The size of the largest cleaning task among all cleaning tasks; For the first The geometric center position vector of each cleaning task; For the first The real-time position vector of the cleaning robot; For vectors The modulus length; This represents the total number of cleaning robots; This is the length of the diagonal of the carriage; This indicates normalization.

[0013] This invention assesses the size of the cleaning task and the average distance from the task to all cleaning robots, giving larger cleaning tasks a higher base attractiveness, while considering cleaning tasks located closer to the center of the cleaning robot cluster as more competitive and popular tasks. This makes large-area cleaning tasks located in the center of the carriage accurately identified as highly competitive cleaning tasks.

[0014] Preferably, the exclusivity tendency satisfies the expression: In the formula, For the first Taiwan cleaning robot for the first Exclusivity of a cleaning task; For the first The geometric center position vector of each cleaning task; For the first The real-time position vector of the cleaning robot; For vectors The modulus length; This is the length of the diagonal of the carriage; For the first The overall competitiveness of a cleaning task; For hyperparameters; For normalization.

[0015] This invention assesses the distance relationship between cleaning robots and cleaning tasks, assigning higher weight to the correlation between nearby cleaning robots and tasks. Simultaneously, it considers the global competitiveness of cleaning tasks, giving higher exclusivity to tasks with lower global competitiveness. This allows the robot to gain priority in subsequent scheduling, automatically guiding it to avoid popular cleaning tasks already being targeted by multiple robots and instead focus on less competitive, closer tasks. This solves the localized congestion problem caused by all robots simultaneously rushing to the same cleaning task area in traditional multi-robot systems.

[0016] Preferably, the task allocation inhibition factor satisfies the expression: In the formula, For the first Taiwan cleaning robot for the first Task allocation inhibition factor for cleaning tasks; To exclude the first The serial numbers of all cleaning robots other than the one from Taiwan. Indicates the first Taiwan cleaning robot for the first Exclusivity of a cleaning task; This represents the total number of cleaning robots.

[0017] This invention assesses the level of attention other cleaning robots pay to a cleaning task, classifying highly-attracted tasks as popular tasks. Through a task allocation inhibition mechanism, the attractiveness of popular tasks to specific cleaning robots is automatically reduced. When multiple cleaning robots are simultaneously focused on a task, the system dynamically senses and strengthens its inhibition effect, prompting robots that might otherwise choose that task to turn to other, less competitive cleaning task areas. Simultaneously, for less popular cleaning tasks with low attention, the system correspondingly weakens its inhibition effect, increasing their attractiveness to nearby cleaning robots and guiding them to perform these neglected tasks. This avoids the localized congestion caused by all cleaning robots simultaneously rushing to the same cleaning task area in traditional multi-robot systems, enabling multiple cleaning robots to work efficiently and collaboratively in different areas of the vehicle.

[0018] Preferably, the total social gravitational vector satisfies the expression: In the formula, For the first The total social gravitational vector acting on the Taiwan cleaning robot; For the first Taiwan cleaning robot for the first Task allocation inhibition factor for cleaning tasks; For the first an area size of the i-th cleaning task; an area size of the cleaning task with the largest area among all cleaning tasks; a total number of cleaning tasks; a geometric center position vector of the i-th cleaning task, a real-time position vector of the i-th cleaning robot, a real-time position vector of the i-th cleaning robot, a module of the vector a module of the vector a module of the vector a gravitational gain coefficient.

[0019] The present application multiplies the task allocation inhibition factor complement with the cleaning task area normalized value, so that the cleaning task with lower competition degree and larger area obtains higher basic attraction, combines the attraction with the unit directional vector pointing to the cleaning task, forms social attraction with clear directionality, guides the cleaning robot to move towards the target area, so that when one cleaning task is focused on by multiple cleaning robots, the task allocation inhibition factor increases, resulting in weakening of the social attraction of the task to the specific cleaning robot, prompting the cleaning robot to automatically turn to other cleaning tasks with smaller competition, at the same time, the larger area cleaning task can obtain appropriate attention due to its stronger basic attraction, avoiding being completely ignored, realizing intelligent allocation of cleaning tasks among multiple cleaning robots, ensuring sufficient processing of large-area cleaning tasks, avoiding local congestion caused by simultaneous rush of all cleaning robots to the same cleaning task, improving the overall collaborative efficiency of the multi-robot system, shortening the total cleaning time, and ensuring the efficiency and stability of the multi-robot collaborative cleaning process.

[0020] Preferably, the repulsive force vector satisfies the expression: ; in the formula, a repulsive force vector of the i-th cleaning robot from the j-th obstacle; a repulsive force vector of the i-th cleaning robot from the j-th obstacle; a repulsive force gain coefficient; an Euclidean distance between the i-th cleaning robot and the j-th obstacle; a repulsive force action range; a real-time position vector of the i-th cleaning robot; a real-time position vector of the i-th cleaning robot; a position vector of the j-th obstacle, a module of the vector a module of the vector a module of the vector a module of the vector a module of the vector a module of the vector

[0021] Preferably, further comprising: when the module value of the total driving force of the cleaning robot exceeds the maximum driving force of the cleaning robot, scaling the module value of the total driving force to within the range of the actual power limit of the cleaning robot.

[0022] In a second aspect, the present application provides a cleaning robot work task scheduling system, comprising a processor and a memory, the memory stores computer program instructions, when the computer program instructions are executed by the processor, the above-mentioned cleaning robot work task scheduling method is realized.

[0023] By adopting the above technical scheme, the above-mentioned cleaning robot work task scheduling method is generated into a computer program and stored in the memory, so as to be loaded and executed by the processor, so as to manufacture a terminal device according to the memory and the processor, and facilitate use.

[0024] The beneficial effects of the present application are as follows:

[0025] The present application identifies the position and size of each residual coal block area in the car body, determines the global competitiveness of the cleaning task based on the size of the cleaning task area and the distance relationship of the cleaning robot, and can identify the popular task that is easy to cause congestion. The present application calculates the exclusive tendency of the cleaning robot to the cleaning task and the task allocation inhibition factor, integrates the social relationship between the cleaning robots into the path planning, when a cleaning task is concerned by multiple cleaning robots, the attraction of the cleaning task to a specific cleaning robot is automatically reduced, the cleaning robot is prompted to turn to a smaller cleaning task, and the piling phenomenon caused by all cleaning robots rushing to the same cleaning task in the traditional method is avoided. The present application combines the social attractive field and the repulsive field, so that the cleaning robot can autonomously avoid collision while pursuing the cleaning task target, ensures the orderly distribution in the limited car body space, and makes multiple cleaning robots can efficiently cooperate in different areas of the car body, improves the cleaning efficiency, shortens the overall operation time, realizes the intelligent balanced distribution of the cleaning task among multiple cleaning robots, and improves the operation efficiency and reliability of the car dumper system. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a flow chart schematically showing a cleaning robot work task scheduling method in the present application;

[0027] Figure 2 is a schematic diagram of an initial state of a simulated car dumper cleaning task allocation;

[0028] Figure 3 is a schematic diagram of an intermediate execution state of a simulated car dumper cleaning task allocation;

[0029] Figure 4 is a schematic diagram of a final completion state of a simulated car dumper cleaning task allocation. DETAILED DESCRIPTION

[0030] 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.

[0031] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0032] This invention discloses a method for scheduling the work tasks of a cleaning robot, referring to... Figure 1 This includes steps S1-S3:

[0033] S1. Obtain cleaning task information inside the carriage.

[0034] It should be noted that after the tippler system completes the tipping and unloading of the wagon and returns to its upright position, some coal will remain inside the wagon. These residual areas constitute a task that needs to be cleaned. In actual tippler operation, due to the adhesive properties of coal and the complexity of the wagon structure, the residual coal is usually distributed unevenly in the bottom, side walls, and corners of the wagon. To effectively allocate tasks and plan paths, it is first necessary to accurately obtain information on all areas inside the wagon that need to be cleaned, including the location and size of each area. This invention identifies the areas of residual coal inside the wagon by using a sensor array installed on the tippler structure.

[0035] Specifically, by installing a sensor array above the tipper structure, the interior of the truck is scanned after the truck has been tipped over and returned to its upright position to identify areas of residual coal that need to be cleaned. For each area of ​​residual coal, its geometric center coordinates and area are obtained, and a cleaning task is created for each area of ​​residual coal. At the same time, the real-time position of all cleaning robots in the unified coordinate system of the truck is obtained.

[0036] S2. Determine the overall competitiveness of the cleaning task based on the area size of the cleaning task and the distance of each cleaning robot to the cleaning task. Determine the exclusive tendency of each cleaning robot to the cleaning task based on the overall competitiveness of the cleaning task and the distance of each cleaning robot to the cleaning task. Determine the task allocation inhibition factor of the cleaning robot to the cleaning task based on the exclusive tendency of other cleaning robots to the cleaning task.

[0037] It should be noted that in the scenario of cleaning the tipper truck compartment, due to the limited space and regular shape of the compartment, path conflicts and task competition are prone to occur when multiple cleaning robots work simultaneously. For example, when multiple cleaning robots are attracted to the same cleaning task at the same time, it will not only cause local congestion, but also generate additional motion consumption due to mutual avoidance, reducing the overall cleaning efficiency. The traditional artificial potential field method only considers the distance relationship between the cleaning robot and the cleaning task, ignoring the dynamic characteristics of task competition in a multi-cleaning robot system, resulting in uneven distribution of cleaning tasks. Therefore, this invention introduces a cleaning task competition perception mechanism. By acquiring the global competitiveness of the cleaning task, the exclusive tendency of the cleaning robot to the cleaning task, and the task allocation inhibition factor, a social gravity field that can reflect the group's cooperative relationship is constructed. In this gravity field, the attraction of the cleaning task to the cleaning robot will be dynamically adjusted according to the real-time competitive situation. When a cleaning task is being watched by multiple cleaning robots, its attraction to a specific cleaning robot will be weakened accordingly, thereby prompting the cleaning robot to turn to other cleaning tasks with less competition, achieving a dynamic and balanced distribution of cleaning tasks.

[0038] Specifically, for any given cleaning task, the overall competitiveness of the cleaning task is determined based on the size of the area to be cleaned and the distance of each cleaning robot to the task.

[0039] ;

[0040] In the formula, For the first The overall competitiveness of a cleaning task; For the first The size of the area to be cleaned; The size of the largest cleaning task among all cleaning tasks, used for... Perform normalization; For the first The geometric center position vector of each cleaning task; For the first The real-time position vector of the cleaning robot; For vectors The modulus length reflects the first The cleaning task and the first The Euclidean distance between the cleaning robots; The total number of cleaning robots, This indicates that all cleaning robots have reached the [number]th [number]. The average distance of each cleaning task; The length of the diagonal of the carriage. Used for Perform normalization; This indicates normalization; the present invention employs linear normalization.

[0041] When the area of ​​the cleaning task The larger the area, the larger the task; large areas of coal residue are more attractive to the cleaning robot, thus increasing its overall competitiveness in the cleaning task. When the cleaning task is closer to the center of the robot cluster, the average distance... A smaller overall competitiveness leads to greater competition among robot vacuums, resulting in a higher overall competitiveness for the cleaning task. When the value is close to 1, it indicates that the cleaning task is highly likely to become a target of competition among multiple cleaning robots, resulting in a very high inherent risk of congestion. When the overall competitiveness... When the value is close to 0, it indicates that the cleaning task is not competitive and the cleaning robot pays little attention to the cleaning task.

[0042] Furthermore, for any given cleaning task, based on the overall competitiveness of the task and the distance of each cleaning robot to that task, the exclusivity preference of each cleaning robot for that task is determined:

[0043] ;

[0044] In the formula, For the first Taiwan cleaning robot for the first Exclusivity of a cleaning task; For the first The geometric center position vector of each cleaning task; For the first The real-time position vector of the cleaning robot; For vectors The modulus length reflects the first The cleaning task and the first The Euclidean distance between the cleaning robots; The length of the diagonal of the carriage. Used to eliminate Dimensions; For the first The overall competitiveness of a cleaning task; These are hyperparameters used to prevent When the value is 0, the denominator is 0. This invention In other embodiments, implementers may set the appropriate parameters according to the actual implementation situation. However, in order to reduce the impact on the degree of exclusionary tendency, Must meet ; For normalization, this invention employs linear normalization.

[0045] When the Taiwan cleaning robot approaches the first When assigning a cleaning task, I prefer to assign it to the first one. The cleaning machine performs the first A cleaning task, at this time Decrease Increase, so that the first Taiwan cleaning robot for the first The exclusivity tendency for the first cleaning task increases, but if the second task is more exclusive... The cleaning task is a popular cleaning task, and its overall competitiveness is... Very high, even if the cleaning task is far from the first Even when the cleaning robot is nearby, its tendency to exclude other cleaning robots remains low. Conversely, for remote or minor, less popular cleaning tasks, its overall competitiveness... The initial sensitivity is low; once a cleaning robot approaches, the robot's exclusivity towards that cleaning task increases dramatically.

[0046] Furthermore, for any given cleaning robot, based on the exclusivity of other cleaning robots towards the cleaning task, a task allocation inhibition factor for that cleaning robot is determined:

[0047] ;

[0048] In the formula, For the first Taiwan cleaning robot for the first Task allocation inhibition factor for cleaning tasks; To exclude the first The serial numbers of all cleaning robots other than the one from Taiwan. Indicates the first Taiwan cleaning robot for the first Exclusivity of a cleaning task; This represents the total number of cleaning robots.

[0049] In the formula, To exclude the first All cleaning robots other than the first cleaning robot are for the first The sum of the exclusivity tendencies for each cleaning task reflects the... Taiwan's cleaning robot competitors are the first The sum of intentions for each cleaning task; when the positions of other cleaning robots or the overall competitiveness of the cleaning task changes, causing other cleaning robots to affect the first cleaning task... When the sum of the exclusivity of each cleaning task increases, the task allocation inhibition factor... Increase, thus affecting the first Taiwan cleaning robot goes to the first The motivation for a cleaning task produces a stronger inhibitory effect; conversely, when When the value is small, the task allocation inhibition factor is low. Decrease, for the first Taiwan cleaning robot goes to the first The weaker the inhibition of a cleaning task, the better. This dynamic inhibition mechanism can prevent congestion caused by multiple cleaning robots rushing to the same cleaning task at the same time.

[0050] S3. Determine the total social gravitational force vector acting on the cleaning robot based on the task allocation inhibition factor and the distance from the cleaning robot to the cleaning task. Determine the repulsive force vector acting on the cleaning robot from the obstacle. Superimpose the total social gravitational force vector and the repulsive force vector to obtain the total driving force of the cleaning robot. Generate motion commands based on the total driving force and schedule the cleaning robot to perform cleaning.

[0051] It should be noted that in the actual working environment of cleaning the tipper truck compartment, the cleaning robots need to work efficiently and collaboratively within a limited space. They need to complete the cleaning task quickly while avoiding collisions and path conflicts. Traditional path planning methods often only focus on task completion efficiency and ignore the cooperative relationship between cleaning robots, resulting in frequent clustering of cleaning robots in actual operations. Therefore, this invention combines social gravitational fields with traditional repulsive fields, enabling cleaning robots to autonomously avoid collisions or congestion with other cleaning robots while pursuing task objectives. This achieves efficient and safe collaborative operation, ensuring that multiple cleaning robots are orderly distributed within a limited space, maximizing parallel operation efficiency, and avoiding motion stagnation caused by mutual interference.

[0052] Specifically, based on the task allocation inhibition factor and the distance between the cleaning robot and the cleaning task, the total social gravitational force vector acting on the cleaning robot is determined:

[0053] ;

[0054] In the formula, For the first The total social gravitational vector acting on the Taiwan cleaning robot; For the first Taiwan cleaning robot for the first Task allocation inhibition factor for cleaning tasks; For the first The size of the area to be cleaned; The size of the largest cleaning task among all cleaning tasks, used for... Perform normalization; The total number of cleaning tasks; For the first The geometric center position vector of each cleaning task. For the first The real-time position vector of the robot vacuum cleaner. For vectors The length of the mold, Indicates the first Taiwan cleaning robot points to the first The unit direction vector for each cleaning task; This is the gravitational gain coefficient, used to control the strength of social gravity.

[0055] In the formula, This is the gravitational gain coefficient, used to control the strength of social gravity. As a proportionality coefficient, it can adjust the magnitude of social gravity, directly affecting the strength of the driving force that moves the cleaning robot toward the cleaning task. The higher the value, the stronger the social gravitational pull, and the faster the cleaning robot tends to its cleaning task. However, this may increase the risk of system oscillation or collisions. The smaller the value, the weaker the social gravity, and the more conservative the robot's movements, but this may lead to a decrease in the efficiency of completing the cleaning task. To ensure that the system has sufficient driving force to complete the task while maintaining stable movement, The value range is [0.1, 1.0]. In this invention, [the value is...]. Setting it to 0.5 balances task completion efficiency and system stability, preventing the cleaning robot from moving too aggressively or too conservatively.

[0056] This invention dynamically modulates the fixed gravity in the traditional artificial potential field method according to the task competition situation, enabling the cleaning robot to sense the degree of competition for cleaning tasks. Taiwan cleaning robot for the first Task allocation inhibition factor for cleaning tasks When the force is relatively small, the total social gravitational vector acting on the cleaning robot will be biased towards the first... The number of cleaning tasks indicates that cleaning tasks with less competition are more attractive to cleaning robots; when When the magnitude is large, the total social gravitational vector acting on the cleaning robot will be biased towards the first... The number of cleaning tasks indicates that large-area cleaning tasks have a stronger fundamental appeal. By calculating the total social gravity vector, the cleaning robot can automatically avoid highly competitive popular cleaning tasks and turn to less competitive but equally important less competitive cleaning tasks, achieving a balanced distribution of cleaning tasks. In particular, when When the time is right, it means that the cleaning task has been completely completed and no longer generates gravity.

[0057] Furthermore, determine the first Taiwan's cleaning robots received from the first The repulsive force vector of each obstacle:

[0058] ;

[0059] In the formula, is the repulsive force vector of the first obstacle to the second cleaning robot; is the repulsive force gain coefficient, used to control the strength of the repulsive force; is the Euclidean distance between the first obstacle and the second cleaning robot, which is calculated based on the coordinate positions of the geometric center of the cleaning robot and the coordinate positions of the geometric center of the obstacle; is the repulsive force action range; is the real-time position vector of the first is the position vector of the first is the length of the vector is the unit directional vector of the first obstacle pointing to the second cleaning robot.

[0060] The obstacles include the car body wall and other cleaning robots, and the size of the repulsive force of the first obstacle to the second cleaning robot is inversely proportional to the distance between the robot and the obstacle. When the robot is close to the car body wall or close to other cleaning robots, the distance decreases, and when the distance is less than the repulsive force action range , the repulsive force is generated, and the smaller the distance, the greater the repulsive force, which pushes the cleaning robot away from the car body wall or other cleaning robots. When the distance is greater than the repulsive force action range , the repulsive force is zero.

[0061] In the formula, is the repulsive force gain coefficient, used to control the strength of the repulsive force field, reflecting the sensitivity of the system to obstacles, representing the aggressiveness of the safety obstacle avoidance behavior, as a proportional coefficient, which can adjust the size of the repulsive force and directly affect the behavior intensity of the cleaning robot avoiding obstacles and each other. The greater the value, the stronger the repulsive force, and the more aggressive the cleaning robot's obstacle avoidance behavior, but it may cause the motion trajectory to be too tortuous and increase the total path length; The smaller the value, the weaker the repulsive force, and the more conservative the cleaning robot's obstacle avoidance behavior, but it may increase the risk of collision. To ensure that the system can effectively avoid collision and not excessively affect the task execution efficiency, the value range of is in the present application,​​​​​​Setting it to 0.05 balances obstacle avoidance safety and task execution efficiency, ensuring that the cleaning robot can work safely and collaboratively in a limited space.

[0062] In the formula, The range of repulsive force is used to determine the distance threshold at which the repulsive force begins to take effect, when the cleaning robot is at a distance from the obstacle. Less than At that time, the repulsive force begins to act, when the distance... Greater than At that time, the repulsive force is zero. It reflects the system's perception range of obstacles and represents the size of the safety buffer zone. The larger the force, the wider the repulsive force range, and the earlier the cleaning robot will start to avoid it, but this may cause the cleaning task to detour excessively. The smaller the value, the narrower the repulsive force's range, and the more delayed the cleaning robot's avoidance behavior. However, this may lead to collisions due to untimely avoidance. To ensure the system has sufficient safety buffer without excessively impacting task execution efficiency, It needs to be larger than the maximum feature size of the end effector of the cleaning robot, and less than twice the maximum feature size of the end effector of the cleaning robot. In this invention, The size is set to 1.5 times the maximum feature size of the end effector of the cleaning robot to ensure safe and collaborative operation of the cleaning robot in a confined space.

[0063] Furthermore, by superimposing the total social gravitational force vector acting on the cleaning robot with the repulsive force vectors acting on the cleaning robot from all obstacles, we obtain the total driving force of the cleaning robot:

[0064] ;

[0065] In the formula, For the first The total driving force of the cleaning robot; For the first The total social gravitational vector acting on the Taiwan cleaning robot; For the first Taiwan's cleaning robots received from the first The repulsive force vector of each obstacle; Let be the total number of obstacles. The total driving force is the vector sum of social attraction and all repulsive forces, which together determine the direction and speed of the cleaning robot's movement. The total social gravitational vector acting on the Taiwan cleaning robot increases At that time, total driving force It tends to point in the direction of the cleaning task; when the cleaning robot approaches an obstacle or another cleaning robot... Increase, total driving force It may deviate from the direction of obstacles. Through this combined force calculation, the cleaning robot can maintain a safe distance from obstacles and other cleaning robots while pursuing its task objectives, ensuring safe and smooth movement. It is important to note that when... The modulus value exceeds the maximum driving force of the cleaning robot. At that time, The modulus is scaled proportionally to Within the specified range, to comply with the actual power limitations of the cleaning robot.

[0066] Furthermore, based on the total driving force Size and orientation, generate the pair of the first The robot vacuum cleaner's underlying motion controller sends speed and angular velocity commands to drive the first... A cleaning robot is moving around.

[0067] When the system detects that any cleaning robot has arrived and completed its cleaning task, it removes that cleaning task from the set of tasks to be cleaned. In the next control cycle, based on the updated set of tasks to be cleaned, all operations in steps S2 and S3 are repeated to recalculate the social gravitational field for all cleaning robots and generate new motion commands, until the set of tasks to be cleaned is empty and the iteration stops.

[0068] Specifically, to prevent deadlock situations where cleaning tasks are ignored, when a cleaning task is continuously... If the task is not selected by any cleaning robot within a control cycle, the task allocation inhibition factor for that cleaning task for all cleaning robots is multiplied by the attenuation coefficient. ( This reduces the degree of inhibition, thereby increasing the attractiveness of the cleaning task to the cleaning robot and ensuring that even the most remote or smallest cleaning task can be assigned and executed in an appropriate time, avoiding deadlock.

[0069] in, This is the number of deadlock detection cycles, used to determine the criteria for ignoring a cleaning task. When a cleaning task is continuously... If a robot is not selected by any cleaning robot within a control cycle, the system will activate the deadlock prevention mechanism. This represents the maximum duration for which cleaning tasks can be reasonably ignored. The larger the value, the higher the system's tolerance for uneven distribution of cleaning tasks, which may result in remote or minor cleaning tasks being ignored for a long time. The smaller the value, the more sensitive the system is to uneven distribution of cleaning tasks, which may lead to frequent triggering of deadlock prevention mechanisms and interfere with normal cleaning task allocation. To ensure that the system can promptly detect ignored tasks without excessively interfering with normal task allocation, The range of values ​​is a control cycle. In the present application, the is set to 5 control cycles, ensuring that even the most remote or smallest task can be allocated and executed within a reasonable time.

[0070] wherein, is a decay coefficient, used to adjust the strength of the deadlock prevention mechanism, in the deadlock prevention mechanism, when a certain cleaning task is determined to be possibly deadlocked, the task allocation inhibition factor of all cleaning robots to the cleaning task is multiplied by to reduce the degree of inhibition and enhance the task attractiveness. represents the degree of aggressiveness of the deadlock prevention mechanism, The greater, the smaller the degree of decay of the inhibition factor, the slower the improvement of the cleaning task attractiveness, and the deadlock prevention mechanism is more conservative. The smaller, the greater the degree of decay of the inhibition factor, the faster the improvement of the cleaning task attractiveness, and the deadlock prevention mechanism is more aggressive. To ensure that the system can effectively prevent deadlock and not interfere with normal task allocation, the value range of is [0.8, 0.95]. In the present application, the is set to 0.9, ensuring that the ignored task is re-allocated within a reasonable time, while avoiding excessive interference with normal task allocation.

[0071] Exemplarily, Figure 2 is a schematic diagram of the initial state of the simulation of the tipping machine cleaning task allocation, Figure 3 is a schematic diagram of the intermediate execution state of the simulation of the tipping machine cleaning task allocation, Figure 4 is a schematic diagram of the final completion state of the simulation of the tipping machine cleaning task allocation.

[0072] The embodiment of the present application also discloses a cleaning robot work task scheduling system, comprising a processor and a memory, the memory stores computer program instructions, when the computer program instructions are executed by the processor, a cleaning robot work task scheduling method according to the present application is realized.

[0073] The above system also includes communication bus and communication interface and other components familiar to those skilled in the art, the setting and function of which are known in the art, therefore, will not be repeated here.

Claims

1. A method for scheduling the work tasks of a cleaning robot, characterized in that, The method comprises: obtaining cleaning task information of the interior of the car body; determining global competitiveness of the cleaning task according to the area size of the cleaning task and the distance of each cleaning robot to the cleaning task, determining exclusive tendency of each cleaning robot to the cleaning task according to the global competitiveness of the cleaning task and the distance of each cleaning robot to the cleaning task, and determining a task allocation inhibition factor of the cleaning robot to the cleaning task according to the exclusive tendency of other cleaning robots to the cleaning task; determining a total social attractive force vector received by the cleaning robot according to the task allocation inhibition factor and the distance of the cleaning robot to the cleaning task, determining a repulsive force vector received by the cleaning robot from an obstacle, superimposing the total social attractive force vector and the repulsive force vector to obtain a total driving force of the cleaning robot, generating a motion instruction according to the total driving force, and dispatching the cleaning robot to clean.

2. The method of claim 1, wherein, The method further comprises: when the cleaning task is not selected within a continuous preset period, reducing the task allocation inhibition factor of all cleaning robots to the cleaning task to enhance the attraction of the cleaning task to the cleaning robots.

3. The method of claim 1, wherein, The method further comprises: scanning the interior of the car body after the car body is returned to the normal position by a car dumper structure through a sensor array installed above the car dumper structure, identifying residual coal block areas that need to be cleaned, obtaining the geometric center coordinates and the area size of each residual coal block area, and establishing a cleaning task for each residual coal block area.

4. The method of claim 1, wherein, The global competitiveness satisfies the expression: ； wherein, is the global competitiveness of the i-th cleaning task; is the global competitiveness of the i-th cleaning task; is the area size of the i-th cleaning task; is the area size of the i-th cleaning task; is the area size of the largest area among all cleaning tasks; is the geometric center position vector of the i-th cleaning task; is the geometric center position vector of the i-th cleaning task; is the real-time position vector of the j-th cleaning robot; is the real-time position vector of the j-th cleaning robot; is the length of the vector is the length of the vector is the total number of cleaning robots; is the length of the diagonal line of the carriage; denotes normalization.

5. The method of claim 1, wherein, The exclusive tendency satisfies the expression: ； wherein, is the exclusive tendency of the i-th cleaning robot to the j-th cleaning task; is the geometric center position vector of the j-th cleaning task; is the real-time position vector of the i-th cleaning robot; is the modulus of the vector is the diagonal length of the vehicle cabin; is the global competitiveness of the j-th cleaning task; is the normalization.

6. The method of claim 1, wherein, The task allocation inhibition factor satisfies the expression: ； In the formula, For the first Taiwan cleaning robot for the first Task allocation inhibition factor for cleaning tasks; To exclude the first The serial numbers of all cleaning robots other than the one from Taiwan. Indicates the first Taiwan cleaning robot for the first Exclusivity of a cleaning task; This represents the total number of cleaning robots.

7. The method of claim 1, wherein, The total social attractive force vector satisfies the expression: ； In the formula, For the first The total social gravitational vector acting on the Taiwan cleaning robot; For the first Taiwan cleaning robot for the first Task allocation inhibition factor for cleaning tasks; For the first The size of the area to be cleaned; The size of the largest cleaning task among all cleaning tasks; The total number of cleaning tasks; For the first The geometric center position vector of each cleaning task. For the first The real-time position vector of the robot vacuum cleaner. For vectors The modulus length; This is the gravitational gain coefficient.

8. The method of claim 1, wherein, The repulsive force vector satisfies the expression: ； wherein, is the repulsive force vector from the first obstacle to the cleaning robot; is the repulsive force gain coefficient; is the Euclidean distance between the first obstacle and the cleaning robot; is the repulsive force action range; is the real-time position vector of the first cleaning robot; is the position vector of the first obstacle, is the module of the vector .​​ 9. The method of claim 1, wherein, The method further comprises: When the modulus of the total driving force of the cleaning robot exceeds the maximum driving force of the cleaning robot, the modulus of the total driving force is scaled to be within the range of 0 to 100% of the maximum driving force of the cleaning robot to comply with the actual power limit of the cleaning robot. of the maximum driving force of the cleaning robot to comply with the actual power limit of the cleaning robot.

10. A cleaning robot work task scheduling system characterized by, The method comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a cleaning robot work task scheduling method according to any one of claims 1-9 is realized.

Images