A sanitation robot formation control method

Abstract

This invention discloses a method for controlling sanitation robot platoons. The method involves: information collection; platoon construction; prediction of the position and working status of each sanitation robot using a neural network; initial platoon position and path planning; planning a safe path; continuous optimization of platoon operations; and completion of regional cluster platoon operations. Based on path deviation and fault alarm information, the method controls sanitation robots that have deviated from their paths to change their driving status. Simultaneously, it controls sanitation robots that have triggered alarms to automatically return to charging or water refilling positions according to cloud-based algorithm planning. Based on the latest product information for the work area, the method re-schedules sanitation robot platoons, replacing those charging, water refilling, or emptying garbage. This invention matches various platoon shapes to different road environments, achieving high platoon shape coverage and adaptability to various road conditions, greatly improving platoon operation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of sanitation robots performing cluster formation joint operations under a networked system, and in particular discloses a sanitation robot formation control method. Background Art

[0002] The sanitation robot fleet includes 5G (5th Generation Mobile Communication Technology) unmanned sanitation cleaning robots, 5G unmanned sanitation cleaning robots, 5G unmanned road sweeping robots, 5G unmanned road cleaning robots, 5G unmanned garbage collection and transportation robots and 5G unmanned integrated cleaning and flushing sanitation robots.

[0003] The sanitation robot fleet is based on the 5G+ sanitation robot AI (Artificial Intelligence) brain, which has super intelligence and security, super fast information response and data interaction capabilities, and super high cluster collaborative operation efficiency. It can autonomously plan routes, intelligently identify garbage, automatically pick up and clean, and has the ability to warn of obstacles ahead and "consciously" avoid danger and detour, achieving efficient operation in complex environments. However, the existing sanitation robot fleet has the following main defects:

[0004] 1. The previous sanitation robot formations were single in shape and could not be operated in an efficient formation in complex road environments;

[0005] 2. When forming a team for operation, equipment costs are not considered, which is inconsistent with the focus on cost in actual sanitation operations;

[0006] 3. When the sanitation robot is operating, the obstacle avoidance fails to comprehensively consider the actual size of the robot and dynamic and static obstacle avoidance, resulting in poor obstacle avoidance effect;

[0007] 4. When the sanitation robots are operating in formation, equipment replacement is not taken into consideration, and changes in factors such as power, water volume, and garbage volume cannot be predicted in real time. When the power of a certain device is too low, it will not be able to operate, thus affecting the stability of the formation's operation, and the timeliness of replacement is also critical.

[0008] Therefore, the above-mentioned defects of the existing sanitation robot fleet are technical problems that need to be solved urgently. Summary of the invention

[0009] The present invention provides a sanitation robot formation control method, aiming to solve at least one of the above-mentioned defects of the existing sanitation robot formation.

[0010] The present invention relates to a sanitation robot formation control method, comprising the following steps:

[0011] Information collection: Collect information about sanitation operation areas and sanitation robots;

[0012] Operation team building: Based on the collected sanitation operation area and sanitation robot information, the sanitation robot products and operation areas participating in this cluster operation are configured, and the operation team information of the sanitation robot is automatically allocated;

[0013] Position and task demand prediction: Use neural networks to predict the position and working status information of each sanitation robot;

[0014] Initial formation position and path planning: Use spider optimization algorithm to establish initial formation position and path planning;

[0015] Planning safe paths: Using artificial potential field algorithm to plan safe paths;

[0016] Continuous optimization of formation operation: According to the planned safe path, the sanitation robots are controlled to start cluster operation; the operation status information of the sanitation robots is monitored in real time during cluster operation, and the deviation between the actual path of the equipment and the planned path is calculated using cloud algorithms based on the operation status information of the sanitation robots, which is used to continuously correct the driving path; at the same time, the product information and fault alarm information of the sanitation robots are monitored in real time. When the water and electricity level of the sanitation robots is detected to be lower than the preset water and electricity level threshold, the corresponding sanitation robots are controlled to issue fault alarms;

[0017] Complete regional cluster formation operations: Based on the path deviation information and fault alarm information, control the sanitation robots that have deviated from the path to change the product driving status. At the same time, control the sanitation robots that have issued an alarm to automatically return to charge or add water according to the cloud algorithm planning; based on the latest product information in the operation area, re-dispatch the sanitation machine formation, and re-dispatch new sanitation machines to replace the sanitation robots that are charging, adding water or dumping garbage.

[0018] The steps to build a job team include:

[0019] According to the collected sanitation operation area and sanitation robot information, configure the type and operation area of ​​the sanitation robot participating in this cluster operation;

[0020] According to the type of sanitation robots configured to participate in this cluster operation and the operating area, several sanitation robots are allocated to form a corresponding operating team.

[0021] Furthermore, the operation formation includes a diamond formation, a trapezoidal formation, a circular formation, a triangular formation, a horizontal linear formation, and a vertical linear formation. According to the type of sanitation robots configured to participate in the cluster operation and the operation area, the steps of allocating a number of sanitation robots to form a corresponding operation formation include:

[0022] If the number of sanitation robots participating in this cluster operation is less than or equal to a preset first number threshold and the width and length of the road section in the operation area participating in this cluster operation are similar, the sanitation robots participating in this cluster operation are allocated to form a diamond formation;

[0023] If the number of sanitation robots participating in this cluster operation is greater than a preset first number threshold and the width and length of the road section in the operation area participating in this cluster operation are similar, the sanitation robots participating in this cluster operation are allocated to form a circular formation;

[0024] If the length of the road section in the operation area participating in this cluster operation is significantly greater than the width of the road section and the number of sanitation robots participating in this cluster operation is within the preset second number threshold range, the sanitation robots participating in this cluster operation are allocated to form a ladder formation;

[0025] If the number of sanitation robots participating in this cluster operation is less than or equal to the preset third number threshold and the width of the road section in the operation area participating in this cluster operation is narrow, the sanitation robots participating in this cluster operation are allocated to form a triangle formation;

[0026] If the length of the road section in the operation area participating in this cluster operation is less than the road section width and the length of the road section in the operation area participating in this cluster operation is relatively small, the sanitation robots participating in this cluster operation are allocated to form a horizontal linear formation;

[0027] If the length of the road section in the operation area participating in this cluster operation is greater than the width of the road section and the width of the road section in the operation area participating in this cluster operation is relatively small, the sanitation robots participating in this cluster operation will be allocated to form a vertical linear formation.

[0028] Furthermore, in the step of information collection, a set of complete evaluation functions for operation formations is designed to adapt to the design of formation schemes for different cities and sanitation operation scenarios. The complete evaluation function for operation formations is a key tool for measuring the efficiency and cost-effectiveness of formation configuration. The complete evaluation function for operation formations is calculated by the following formula:

[0029] in, represents the complete evaluation function of the operation team, , , are the number of cleaning robots, washing robots and garbage transfer robots, respectively. , , These are the prices of the cleaning robot, washing robot, and garbage transfer robot; is the total cost weight, To cover the uneven weight, is the equipment efficiency weight, is the operating capability weight, is the area covered by each robot; represents the coverage area per dollar cost, The working width of the cleaning robot. To clean the robot's operating width, is the operating speed of all robots.

[0030] Furthermore, in the step of location and task demand prediction, the estimated distance is calculated as follows:

[0031] in, For robots With robots The distance between robots The position coordinates are ( , ),robot The position coordinates are ( , ),robot The current position coordinates are ( , ),robot The current position coordinates are ( , );

[0032] The estimated time is calculated as follows:

[0033] in, To predict from the robot To the robot time, For robots With robots The distance between is the speed of neural network prediction.

[0034] Furthermore, in the steps of initial formation position and path planning, the spider bee optimization algorithm adopts a position update reference model, which is:

[0035] in, and They are the optimal positions for the new and old respectively; and is a random number, and are the local and global optimal positions respectively.

[0036] Furthermore, in the step of planning a safe path, a formation integrity parameter is added to the objective function.

[0037] in, For robots Minimize replacement time, For robots The replacement time, is the formation integrity parameter, and are the coefficients for adjusting the weights of the two, is the number of robots that need to be replaced.

[0038] Furthermore, in the step of performing continuous optimization of the formation operation, the latest replacement time is obtained as:

[0039] in, For robots Minimize replacement time, For robots The path time, It's a robot Fixed operation time, is the formation integrity parameter, and are the coefficients for adjusting the weights of the two, is the number of robots that need to be replaced.

[0040] The beneficial effects achieved by the present invention are:

[0041] The present invention provides a sanitation robot formation control method, which includes information collection: collecting sanitation operation area and sanitation robot information; operation formation construction: configuring sanitation robot products and operation areas participating in this cluster operation according to the collected sanitation operation area and sanitation robot information, and automatically allocating the sanitation robot operation formation information; position and task demand prediction: using a neural network to predict the position and working status information of each sanitation robot; initial formation position and path planning: using a spider optimization algorithm to establish the initial formation position and path planning; planning a safe path: using an artificial potential field algorithm to plan a safe path; performing continuous optimization of formation operations: controlling the sanitation robot to start cluster operation according to the planned safe path; and real-time monitoring of the sanitation robot operation during cluster operation. Status information, based on the operating status information of the sanitation robot, use the cloud algorithm to calculate the deviation between the actual path of the equipment and the planned path, and use it to continuously correct the driving path; at the same time, monitor the product information and fault alarm information of the sanitation robot in real time, and when it is detected that the water and electricity level of the sanitation robot is lower than the preset water and electricity level threshold, the corresponding sanitation robot is controlled to issue a fault alarm; complete regional cluster formation operations: according to the path deviation information and fault alarm information, control the sanitation robot that deviates from the path to change the product driving state, and at the same time control the sanitation robot that has an alarm to automatically return to charge or add water according to the cloud algorithm planning; according to the latest product information in the operating area, re-dispatch the sanitation machine formation, and re-dispatch the new sanitation machine to replace the sanitation robot that is charging, adding water or dumping garbage. The beneficial effects achieved by the sanitation robot formation control method provided by the present invention are specifically as follows:

[0042] 1. The previous sanitation robot formation shape is single, and it is not possible to adopt an efficient formation shape to operate in complex road environments. The present invention matches a variety of formation shapes according to different road environments, has a high formation shape coverage rate, can adapt to a variety of road environments, and greatly improves the efficiency of formation operations.

[0043] 2. In the past, when setting up sanitation teams, equipment costs were not considered, which is inconsistent with the focus on cost in actual sanitation operations. The present invention can optimize the configuration of robot teams according to the needs of different cities through economic evaluation functions. This flexible strategy allows cities to choose the most appropriate robot combination and operation mode based on their own economic conditions and cleaning needs, thereby controlling or optimizing costs while ensuring cleaning effects.

[0044] 3. When the sanitation robot is operating, the improved artificial potential field algorithm takes into account the actual size of the robot and the dynamic and static obstacle avoidance, and the obstacle avoidance effect is better. Through real-time adjustment of strategies and dynamic potential field calculation, the robot can effectively avoid obstacles and safely navigate to the target location. The optimized artificial potential field algorithm has become more robust and efficient, and can effectively handle complex and changing environments, ensuring that the path planning of the sanitation robot is both safe and efficient, and is suitable for robot navigation and path planning in complex urban environments.

[0045] 4. The introduction of spider bee optimization algorithm and neural network for equipment replacement of sanitation robots can predict changes in factors such as power, water, and garbage volume in real time, and can replace early warning equipment in a timely manner, and ensure that the replacement time is minimal. The spider bee algorithm ensures that the robot formation can maintain its formation and efficiency when performing tasks through global and local leader decision-making mechanisms, especially when facing complex environments or emergencies, it can quickly reorganize the formation to ensure the continuity of operations. As data accumulates, the neural network continuously optimizes the prediction model and improves the accuracy of the prediction. This learning ability enables the robot scheduling system to continue to evolve. Through continuous performance evaluation and strategy adjustment, the system can self-optimize the decision-making process, adapt to the environment more and more, and respond to strategies more and more efficiently. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of an embodiment of a sanitation robot formation control method of the present invention. DETAILED DESCRIPTION

[0047] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0048] like Figure 1 As shown, the first embodiment of the present invention proposes a sanitation robot formation control method, comprising the following steps:

[0049] Step S100, information collection: collecting information on sanitation operation areas and sanitation robots.

[0050] The sanitation operation area includes the sanitation operation length and sanitation operation width, and the sanitation robot information includes the flushing robot product information, the road sweeping robot product information, the robot operation speed V, and the garbage transfer robot price T3. The collected information such as the sanitation operation area (length L, width W), flushing robot product information (flushing width L1, price T1), road sweeping robot product information (sweeping width L2, price T2), robot operation speed V, and garbage transfer robot price T3 are entered into the cloud platform.

[0051] Step S200, work formation construction: according to the collected sanitation work area and sanitation robot information, the sanitation robot products and work areas participating in this cluster operation are configured, and the work formation information of the sanitation robot is automatically allocated.

[0052] On the mobile terminal, based on the robot product information and sanitation operation area obtained from the cloud server, select the flushing robots, sweeping robots, garbage transfer robots products participating in this cluster operation and select the operation area, and then upload the above configuration information to the cloud server. The cloud server automatically assigns several sanitation robots to form various operation formations.

[0053] Step S300, position and task requirement prediction: using a neural network to predict the position and working status information of each sanitation robot.

[0054] By adopting the neural network prediction model, in the process of equipment matching, the neural network can be used to assist in judging and predicting the status and behavior of each sanitation robot, such as predicting the power consumption rate, water consumption rate, garbage generation rate, etc. This can more accurately determine the replacement time and optimal replacement plan of the sanitation robot.

[0055] 1. Neural network prediction model function:

[0056] Predict the power, water, and garbage volume status of the sanitation robot; output prediction values ​​are used to determine the warning level and response strategy.

[0057] 2. Neural network prediction model structure:

[0058] Input layer: current status parameters (power, water, garbage volume, location).

[0059] Hidden layer: Multilayer Perceptron (MLP).

[0060] Output layer: predicted future state (power, water, garbage volume).

[0061] 3. Optimization objectives:

[0062] Minimize the total replacement time of the sanitation robot while minimizing the path distance.

[0063] 4. System requirements and parameter definition

[0064] 1) Define the number of the sanitation robots in operation

[0065] Robot number and position: Each robot lie in( , ).

[0066] Sweeper No.

[0067] Cleaning machine number

[0068] Waste transfer machine number .

[0069] 2) Record robot status parameters

[0070] Power :robot The percentage of charge at time t.

[0071] Water volume : The percentage of water in the washing machine's tank.

[0072] Trash bin capacity) : The percentage of the waste bin volume of the waste transfer machine.

[0073] 3) Real-time prediction and status update

[0074] Use a neural network to predict the state of each robot at the next time step:

[0075] enter: , , , ,

[0076] Output: , , , (Recommended speed).

[0077] 4) Dynamic location update and path planning

[0078] Setting up the robot The position is updated by the current velocity )Decide:

[0079]

[0080]

[0081] in, It's a robot In time The direction of movement, is the time step.

[0082] 5) Distance and time calculation

[0083] The estimated distance is calculated as:

[0084] (1)

[0085] In formula (1), For robots With robots The distance between robots The position coordinates are ( , ),robot The position coordinates are ( , ),robot The current position coordinates are ( , ),robot The current position coordinates are ( , ).

[0086] The estimated time is calculated as follows:

[0087] (2)

[0088] In formula (2), To predict from the robot To the robot time, For robots With robots The distance between is the speed of neural network prediction.

[0089] 6) Early warning and backup robots

[0090] Early warning robot set: P = { ∣Warning conditions met}.

[0091] Backup robot set: S={ ∣In good condition and can be replaced}.

[0092] Step S400, initial formation position and path planning: using the spider optimization algorithm to establish the initial formation position and path planning.

[0093] Detailed design of spider bee optimization algorithm:

[0094] 1. Initialization

[0095] A group of spider bees are randomly initialized, each of which represents a possible robot scheduling solution.

[0096] 2. Local and global update rules

[0097] Local update: Based on the position of the local leader, use the position update formula to find a better solution.

[0098] Global update: All spider bees update their positions according to the best solution (global leader).

[0099] 3. Update rules

[0100] Location update reference model:

[0101] The spider bee optimization algorithm adopts a position update reference model, which is:

[0102] (3)

[0103] In formula (3), and They are the optimal positions for the new and old respectively; and is a random number, and are the local and global optimal positions respectively.

[0104] Step S500, planning a safe path: using an artificial potential field algorithm to plan a safe path.

[0105] 1. The following is a complete description of the artificial potential field algorithm:

[0106] 1. Improved gravity formula (increase equity coefficient):

[0107] (4)

[0108] In formula (4), is the gravitational field constant, is the gravitational weight coefficient, is a small constant that avoids infinite gravity at the target point, is the robot's current position, is the robot's target position, It is the distance threshold at which gravity is reduced when approaching the target point.

[0109] 2. Repulsion field optimization (considering robot volume and dynamic and static obstacles)

[0110] The repulsion field is adjusted based on the robot's size and the obstacle type (dynamic or static):

[0111] Static obstacle repulsion formula:

[0112] (5)

[0113] In formula (5), is the repulsive field constant, is the repulsion weight coefficient, is the distance from the robot center to the nearest static obstacle, is the safe distance determined by the robot volume, usually set to max(L, W) / 2max(L, W) / 2; is the maximum distance over which the repulsive force acts.

[0114] Dynamic obstacle repulsion formula:

[0115] (6)

[0116] In formula (6), and It is the repulsive field constant and weight coefficient set specifically for dynamic obstacles. is the distance from the center of the robot to the nearest dynamic obstacle. This distance is the time Function of is the safe distance from dynamic obstacles, usually greater than big, It is the maximum distance of repulsion in a dynamic obstacle environment.

[0117] Calculation of combined force and adjustment force:

[0118] The resultant force is the vector sum of the gravitational force and the repulsive force. A regulating force is added to ensure that the resultant force is not zero, thus preventing the robot from stagnating:

[0119] Resultant force formula:

[0120] (7)

[0121] In formula (7), is the total potential energy, including the sum of attractive and repulsive forces; It is a regulating force, a small random vector or a fine-tuning force determined based on the path planning strategy, which is used to deal with potential stagnation and avoid the device from stagnating due to the resultant force being zero.

[0122] 3. Complete application of artificial potential field algorithm

[0123] In practical applications, the robot's control system needs to calculate the resultant force in real time according to the above potential field formula and adjust the robot's moving direction and speed. The system's integrated sensors (such as radar and vision system) provide the necessary obstacle information (position, speed, etc.) and dynamically update the repulsive field parameters to achieve effective navigation in complex environments.

[0124] The optimized artificial potential field algorithm has become more robust and efficient, and can effectively handle complex and changing environments, ensuring that the robot's path planning is both safe and efficient, and is suitable for robot navigation and path planning in complex urban environments. By adjusting strategies in real time and calculating dynamic potential fields, the robot can effectively avoid obstacles and safely navigate to the target location.

[0125] 2. Equipment Scheduling Objective Function Design

[0126] The goal of equipment scheduling is to minimize the total equipment replacement time, that is, to find a replacement device as soon as possible and complete the replacement when the device has an early warning. Taking into account factors such as power, water, and trash bin capacity, the goal is to minimize the total equipment replacement time. , which includes the total time of finding a replacement robot and executing the replacement. Specifically, the objective function can be expressed as:

[0127] (8)

[0128] In formula (8), It is The replacement time of the robots that need to be replaced, is the number of robots that need to be replaced.

[0129] Add formation integrity parameter to the objective function ):

[0130] (9)

[0131] In formula (9), For robots Minimize replacement time, For robots The replacement time, is the formation integrity parameter, and are the coefficients for adjusting the weights of the two, is the number of robots that need to be replaced.

[0132] The formation integrity parameters of each sanitation robot reflect its importance and urgency in maintaining the formation structure, such as calculations based on its power level, task completion, etc.

[0133] Step S600, perform continuous optimization of formation operations: control the sanitation robots to start cluster operations according to the planned safe path; monitor the operating status information of the sanitation robots in real time during cluster operations, and use cloud-based algorithms to calculate the deviation between the actual path of the equipment and the planned path based on the operating status information of the sanitation robots, so as to continuously correct the driving path; at the same time, monitor the product information and fault alarm information of the sanitation robots in real time, and when it is detected that the water and electricity level of the sanitation robots is lower than the preset water and electricity level threshold, control the corresponding sanitation robots to issue a fault alarm.

[0134] 1. Warning thresholds, priorities, and response strategies

[0135] Alert priority setting

[0136] 1. Power warning (highest priority):

[0137] Threshold: 30%.

[0138] Action: If the robot's battery level is less than 30%, an alert is issued and a robot with a battery level above 80% is sought for replacement.

[0139] 2. Water volume warning (second priority):

[0140] Threshold: The setting is determined based on the actual capacity, such as 20% full capacity.

[0141] Action: When the water level is lower than the threshold, the cleaning robot automatically plans a path to the nearest water inlet to add water.

[0142] 3. Trash bin volume warning (third priority):

[0143] Threshold: such as 75% full load.

[0144] Action: When the volume of the garbage bin of the garbage transfer robot exceeds the threshold, it automatically plans a path to the nearest garbage station to dump the garbage.

[0145] Optimization goal:

[0146] Minimize the total robot change time while minimizing the path distance.

[0147] Specific implementation steps and formulas

[0148] Monitor robot status:

[0149] Collecting power , water volume , garbage volume ,Location( , ).

[0150] Early warning and dynamic scheduling:

[0151] Use neural networks to predict future states and determine whether an early warning will be triggered.

[0152] According to the warning level, the Spider Bee optimization algorithm calculates the optimal replacement path.

[0153] Calculation of replacement time and path time:

[0154] 1. Path time for:

[0155] = (10)

[0156] In formula (10), is the distance between two points, is the robot speed.

[0157] 2. Replacement time for:

[0158] = (11)

[0159] In formula (11), It is the fixed operation time of a single robot, such as the time to add water or dump garbage.

[0160] 2. The latest replacement time is:

[0161] (12)

[0162] In formula (12), For robots Minimize replacement time, For robots The path time for the robot The path time, It's a robot Fixed operation time, is the formation integrity parameter, and are the coefficients for adjusting the weights of the two, is the number of robots that need to be replaced.

[0163] Through the combination of spider bee optimization algorithm and neural network, the sanitation robot dispatching system can not only respond to various warning states quickly and accurately, but also intelligently optimize the dispatching strategy to ensure the continuity of operation and the integrity of the formation, greatly improving the efficiency and quality of sanitation operations. This method effectively reduces the robot replacement time and improves the overall operation efficiency. It has high practical value and promotion prospects. This highly automated and intelligent system is the trend of future urban sanitation management.

[0164] Step S700, complete the regional cluster formation operation: according to the path deviation information and fault alarm information, control the sanitation robot that deviates from the path to change the product driving state, and at the same time control the sanitation robot that has an alarm to automatically return to charge or add water according to the cloud algorithm planning; according to the latest product information in the operation area, re-dispatch the sanitation machine formation, and re-dispatch the new sanitation machine to replace the sanitation robot that is charging, adding water or dumping garbage.

[0165] In order to better achieve the requirements of the operation equipment allocation and matching in step S700, when the power of any sanitation robot, the water volume of the cleaning robot, and the volume of the garbage bin of the garbage transfer robot are warned, the power priority is set to the highest. When the power of all robots is less than 30%, the robot issues a warning and adopts an algorithm to match other equipment with a power greater than 80% to replace the operation. The water volume of the cleaning robot is the second priority. When the water volume alarm is triggered, it is necessary to automatically match the nearest water inlet and drive to the nearby water inlet to add water. The volume of the garbage bin of the garbage transfer robot is the third priority. When the volume of the garbage bin is greater than a certain threshold, the alarm is triggered, and the equipment needs to drive to the nearby garbage station to dump garbage. For any sanitation robot, a warning must be issued a certain time in advance. When leaving the operation area, the replacement sanitation robot must fill in the position within a certain time to ensure that the sanitation robot formation completes the equipment replacement as quickly as possible. At the same time, when the sanitation robot is replaced, the operation integrity of the formation must be considered to keep the formation intact as soon as possible.

[0166] Preferably, see Figure 1 In the sanitation robot formation control method provided in this embodiment, step S200 includes:

[0167] Step S210: According to the collected information about the sanitation operation area and the sanitation robot, the type and operation area of ​​the sanitation robot participating in the cluster operation are configured.

[0168] The mobile terminal configures the type and operation area of ​​the sanitation robot participating in this cluster operation based on the robot product information and sanitation operation area obtained from the cloud server.

[0169] Step S220: According to the types of sanitation robots configured to participate in this cluster operation and the operation areas, a number of sanitation robots are allocated to form a corresponding operation formation.

[0170] The cloud server allocates several sanitation robots to form corresponding operation formations according to the types of sanitation robots and operation areas that are configured to participate in this cluster operation. The operation formations may include diamond formations, ladder formations, circular formations, triangle formations, horizontal linear formations, and vertical linear formations, all of which are within the scope of protection of this patent.

[0171] Preferably, in the sanitation robot formation control method provided in this embodiment, step S220 includes:

[0172] Step S221: If the number of sanitation robots participating in this cluster operation is less than or equal to a preset first number threshold and the width and length of the road section in the operation area participating in this cluster operation are similar, the sanitation robots participating in this cluster operation are allocated to form a diamond formation.

[0173] When the number of devices is small (e.g. ≤4), and when the width and length of the road section are similar (i.e., W≈L), choosing a diamond formation can effectively cover the entire area.

[0174] Applicable scenarios: Suitable for almost square areas, which can evenly distribute the equipment's workload.

[0175] Step S222: If the number of sanitation robots participating in this cluster operation is greater than a preset first number threshold and the width and length of the road section in the operation area participating in this cluster operation are similar, the sanitation robots participating in this cluster operation are allocated to form a circular formation.

[0176] When the width and length of the road section are relatively uniform and the number of equipment is large ( =6), the circular formation can maximize the coverage area and reduce the omission of edge operations.

[0177] Applicable scenarios: Suitable for areas close to square, especially where the center needs to be focused.

[0178] Step S223: If the length of the road section in the operation area participating in this cluster operation is significantly greater than the road section width and the number of sanitation robots participating in this cluster operation is within a preset second number threshold range, the sanitation robots participating in this cluster operation are allocated to form a ladder formation.

[0179] When the length of the road section is significantly greater than its width (L>W), and the number of equipment is moderate (3≤ ≤6), the use of echelon formation is conducive to effective operation along the length direction.

[0180] Applicable scenarios: Suitable for long road sections, especially when more equipment needs to be concentrated at one end of the road section for key operations.

[0181] Step S224: If the number of sanitation robots participating in this cluster operation is less than or equal to a preset third number threshold and the width of the road section in the operation area participating in this cluster operation is narrow, the sanitation robots participating in this cluster operation are allocated to form a triangle formation.

[0182] When the number of devices is small ( ≤3) and when the road width is narrow, the triangular formation helps to concentrate forces and increase operation density.

[0183] Applicable scenarios: Suitable for narrow and long roads or work areas where high-efficiency coverage is required.

[0184] Step S225: If the length of the road section in the operation area participating in this cluster operation is less than the road section width and the length of the road section in the operation area participating in this cluster operation is relatively small, the sanitation robots participating in this cluster operation are allocated to form a horizontal linear formation.

[0185] When the road section is short, wide, and the operation requirements are concentrated (W>L and Relatively small), choosing a horizontal straight line formation can maximize edge operation efficiency.

[0186] Applicable scenarios: Suitable for wide roads that require concentrated working force along the width direction.

[0187] Step S226: If the length of the road section in the operation area participating in this cluster operation is greater than the road section width and the road section width in the operation area participating in this cluster operation is relatively small, the sanitation robots participating in this cluster operation are allocated to form a vertical linear formation.

[0188] When the road section is short in width and long in length, and the work is concentrated along the length direction (L>W and Relatively small), choosing a vertical straight formation can maximize the efficiency of coverage length.

[0189] Applicable scenarios: Suitable for long and narrow road sections, especially where it is necessary to concentrate working forces along the length.

[0190] Further, see Figure 1 In the sanitation robot formation control method provided in this embodiment, in step S100, in order to better achieve the purpose of the invention, a set of operation formation complete evaluation function is designed , in order to adapt to the formation scheme design of different cities and sanitation operation scenarios. Complete evaluation function of the operation formation It is a key tool for measuring the efficiency and cost-effectiveness of fleet configuration. This function combines four evaluation indicators: total cost, coverage unevenness, equipment utilization efficiency, and operational capacity to optimize fleet decisions. The following is a detailed description and calculation method of these indicators and evaluation functions, specifically:

[0191] 1. Definition of evaluation function indicators

[0192] 1. Total cost

[0193] Definition: The sum of the costs of all selected equipment.

[0194] (13)

[0195] In formula (13), , , are the number of cleaning robots, washing robots and garbage transfer robots, respectively. , , These are the prices of cleaning robot, washing robot and garbage transfer robot respectively.

[0196] The sum of the costs of the equipment described, taking into account the purchase or operating cost of each piece of equipment.

[0197] 2. Uneven coverage

[0198] Definition: It is measured based on the standard deviation of the coverage of the device. Ideally, we want each device to evenly cover its working area, assuming that each robot has the same coverage efficiency.

[0199] If we assume that the area covered by each robot is , the total coverage area is: .

[0200] Coverage non-uniformity can be calculated as the deviation of the actual coverage area from the ideal coverage area (assuming that each robot is evenly distributed).

[0201] By calculating the standard deviation of the coverage area of ​​each device, the larger the standard deviation, the higher the coverage unevenness.

[0202] 3. Equipment efficiency

[0203] Definition: The ratio of equipment coverage area to equipment cost, characterizing cost effectiveness. Overall utilization efficiency is the weighted average of these ratios, and equipment utilization efficiency reflects the coverage area generated per unit cost.

[0204]

[0205] here Indicates the coverage area per dollar cost. This value is different for different types of robots.

[0206] 4. Operational ability

[0207] Definition of robot types and operating capabilities:

[0208] Cleaning robot working width , cleaning robot working width ,in .

[0209] Garbage transfer robot:

[0210] The operating capacity is equivalent to that of a cleaning robot. .

[0211] Operational capacity calculation:

[0212] Assuming that the operating speed of all robots is V (unit: m / s), the operating capacity of each robot (unit: m2 / s), i.e. the operating area, is given by the following formula:

[0213] Cleaning robot operation capabilities

[0214] Cleaning robot operation capability

[0215] Transfer robot operation capability .

[0216] 2. Evaluation function formula

[0217] Considering that the importance of each indicator may be different in actual operation, the weight coefficients α, β, γ, and δ are introduced to adjust the influence of each indicator. The complete evaluation function of the operation formation is calculated by the following formula:

[0218] (14)

[0219] In formula (14), represents the complete evaluation function of the operation team, , , are the number of cleaning robots, washing robots and garbage transfer robots, respectively. , , These are the prices of the cleaning robot, washing robot, and garbage transfer robot; is the total cost weight, To cover the uneven weight, is the equipment efficiency weight, is the operating capability weight, is the area covered by each robot; represents the coverage area per dollar cost, The working width of the cleaning robot. To clean the robot's operating width, is the operating speed of all robots. Used to reflect the importance of cost. Should be set larger. Used to reflect the importance of uniform coverage. Should be larger. Used to reflect the coverage area per dollar cost, when more attention is paid to cost-effectiveness Should be larger. This is a newly added weight that measures the importance of work capacity (area / time). If the task focuses on high-efficiency work, Should be larger.

[0220] The sanitation robot formation control method provided in this embodiment has the following beneficial effects:

[0221] 1. This embodiment matches various formation shapes such as diamond, trapezoid, circle, etc. according to different road environments. The formation shape coverage rate is high, which can adapt to various road environments and greatly improve the efficiency of formation operation;

[0222] 2. This embodiment introduces an economic evaluation function , considering factors such as robot parameters, operating area, and customer input costs, the configuration of robot fleets can be optimized according to the needs of different cities. This flexible strategy allows cities to choose the most appropriate robot combination and operation mode based on their own economic conditions and cleaning needs, thereby ensuring cleaning results while controlling or optimizing costs;

[0223] 3. This embodiment adopts improved artificial comprehensive consideration of the actual volume size of the robot and dynamic and static obstacle avoidance, which has better obstacle avoidance effect and avoids the shortcomings of local optimum and minimum. Through real-time adjustment strategy and dynamic potential field calculation, the robot can effectively avoid obstacles and safely navigate to the target position;

[0224] 4. This embodiment introduces the spider bee optimization algorithm and neural network, which can predict the changes of factors such as power, water, and garbage volume in real time, so as to timely replace the sanitation robot and ensure the minimum replacement time and the stability of the formation. The spider bee algorithm ensures that the robot formation can maintain its formation and efficiency when performing tasks through the global and local leader decision-making mechanism, especially when facing complex environments or emergencies, it can quickly reorganize the formation to ensure the continuity of operations. The neural network continuously optimizes the prediction model as data accumulates, improving the accuracy and response speed of the prediction.

[0225] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A sanitation robot formation control method, characterized in that: The following steps are involved: Information collection: Collect information about sanitation operation areas and sanitation robots; Operation team construction: according to the collected sanitation operation area and sanitation robot information, the sanitation robot products and operation areas participating in this cluster operation are configured, and the operation team information of the sanitation robot is automatically allocated; Position and task demand prediction: Use neural networks to predict the position and working status information of each sanitation robot; Initial formation position and path planning: The spider bee optimization algorithm is used to establish the initial formation position and path planning; Planning safe paths: Using artificial potential field algorithm to plan safe paths; Continuous optimization of formation operation: according to the planned safe path, control the sanitation robot to start cluster operation; monitor the operation status information of the sanitation robot in real time during cluster operation, and use the cloud algorithm to calculate the deviation between the actual path of the device and the planned path based on the operation status information of the sanitation robot, so as to continuously correct the driving path; at the same time, monitor the product information and fault alarm information of the sanitation robot in real time, and when it is detected that the water / electricity of the sanitation robot is lower than the preset water / electricity threshold, control the corresponding sanitation robot to issue a fault alarm; Complete regional cluster formation operations: According to the path deviation information and fault alarm information, control the sanitation robot that deviates from the path to change the product driving status, and at the same time control the sanitation robot that has an alarm to automatically return to charge or add water according to the cloud algorithm planning; according to the latest product information in the operation area, re-dispatch the sanitation machine formation, and re-dispatch new sanitation machines to replace the sanitation robots that are charging, adding water or dumping garbage; the product information includes flushing robot product information, sweeping robot product information and garbage transfer robot product information.

2. The sanitation robot formation control method according to claim 1, characterized in that: The steps of constructing the task team include: According to the collected sanitation operation area and sanitation robot information, configure the type and operation area of ​​the sanitation robot participating in this cluster operation; According to the type of sanitation robots configured to participate in this cluster operation and the operating area, several sanitation robots are allocated to form a corresponding operating team.

3. The sanitation robot formation control method according to claim 2, characterized in that: The working formation includes a diamond formation, a trapezoidal formation, a circular formation, a triangular formation, a horizontal linear formation and a vertical linear formation. The step of allocating a number of sanitation robots to form a corresponding working formation according to the type of sanitation robots configured to participate in the cluster operation and the working area includes: If the number of sanitation robots participating in this cluster operation is less than or equal to a preset first number threshold and the width and length of the road section in the operation area participating in this cluster operation are similar, the sanitation robots participating in this cluster operation are allocated to form a diamond formation; If the number of sanitation robots participating in this cluster operation is greater than a preset first number threshold and the width and length of the road section in the operation area participating in this cluster operation are similar, the sanitation robots participating in this cluster operation are allocated to form a circular formation; If the length of the road section in the operation area participating in this cluster operation is significantly greater than the width of the road section and the number of sanitation robots participating in this cluster operation is within the preset second number threshold range, the sanitation robots participating in this cluster operation are allocated to form a ladder formation; If the number of sanitation robots participating in this cluster operation is less than or equal to the preset third number threshold and the width of the road section in the operation area participating in this cluster operation is narrow, the sanitation robots participating in this cluster operation are allocated to form a triangle formation; If the length of the road section in the operation area participating in this cluster operation is less than the road section width and the length of the road section in the operation area participating in this cluster operation is relatively small, the sanitation robots participating in this cluster operation are allocated to form a horizontal linear formation; If the length of the road section in the operation area participating in this cluster operation is greater than the width of the road section and the width of the road section in the operation area participating in this cluster operation is relatively small, the sanitation robots participating in this cluster operation will be allocated to form a vertical linear formation.

4. The sanitation robot formation control method according to claim 1, characterized in that: In the step of collecting information, a set of complete evaluation functions for operation formations is designed to adapt to the design of formation schemes for different cities and sanitation operation scenarios. The complete evaluation function for operation formations is a key tool for measuring the efficiency and cost-effectiveness of formation configuration. The operation capability of the garbage transfer robot is equivalent to that of the cleaning robot. Assuming W3≈W2, the complete evaluation function for the operation formation is calculated by the following formula: in, represents the complete evaluation function of the operation team, , , are the number of cleaning robots, washing robots and garbage transfer robots, respectively. , , These are the prices of the cleaning robot, washing robot, and garbage transfer robot; is the total cost weight, To cover the uneven weight, is the equipment efficiency weight, is the operational capability weight, is the area covered by each robot; represents the coverage area per dollar cost, The working width of the cleaning robot. To clean the robot's working width, is the operating speed of all robots.

5. The sanitation robot formation control method according to claim 4, characterized in that: In the step of predicting the position and task requirements, the estimated distance is calculated as follows: in, For robots With robots The distance between robots The position coordinates are ( , ),robot The position coordinates are ( , ),robot The current position coordinates are ( , ),robot The current position coordinates are ( , ); The estimated time is calculated as follows: in, To predict from the robot To the robot time, For robots With robots The distance between is the speed of neural network prediction.

6. The sanitation robot formation control method according to claim 5, characterized in that: In the steps of initial formation position and path planning, the spider bee optimization algorithm adopts a position update reference model, which is: in, and They are the new and old optimal positions respectively; and is a random number, and are the local and global optimal positions respectively.

7. The sanitation robot formation control method according to claim 6, characterized in that: In the step of planning a safe path, a formation integrity parameter is added to the objective function. in, For robots Minimize replacement time, For robots The replacement time, is the formation integrity parameter, and are the coefficients for adjusting the weights of the two, is the number of robots that need to be replaced.

8. The sanitation robot formation control method according to claim 7, characterized in that: In the step of performing continuous optimization of the formation operation, the latest replacement time is obtained as: in, For robots Minimize replacement time, For robots The path time, It's a robot Fixed operation time, is the formation integrity parameter, and are the coefficients for adjusting the weights of the two, is the number of robots that need to be replaced.

Images