Consistency constrained detection robot task allocation method

Abstract

The invention relates to the technical field of detection robots, in particular to a consistency-constrained detection robot task allocation method, which comprises the following steps: determining to-be-detected tasks and the number of detection robots; the to-be-detected tasks and the number of the detection robots are input into a task allocation model, task allocation results of the detection robots are obtained, and the task allocation model is constructed based on a consistency constraint algorithm; and driving the detection robots to perform detection work based on the task allocation result, updating the number of the to-be-detected tasks and the number of the detection robots in real time in the detection process, continuously inputting the to-be-detected tasks and the number of the detection robots into the task allocation model, and updating the task allocation result until the detection is finished. The detection robot task allocation method based on the consistency beam is more practical after the tasks are allocated, allocation of the detection tasks can be dynamically adjusted in the detection process, and resource waste is greatly reduced.

Description

Technical Field

[0001] The invention relates to the technical field of detection robots, and in particular to a detection robot task allocation method with consistency constraints. Background Art

[0002] With the complexity of the global security situation and the improvement of people's demand for public safety, the demand for the detection of contraband is increasing. Especially in places with dense traffic and a wide variety of items, the detection task of contraband has become increasingly complex and arduous. These places not only have a large flow of people, but also carry a wide variety of items, which may hide potential threats. Therefore, how to achieve efficient, comprehensive and accurate detection of contraband has become a major challenge facing security inspection work. In order to meet this challenge, the introduction of detection robots has become a major innovation in the field of security inspection. Detection robots can achieve efficient and accurate detection of contraband by integrating advanced sensor technology, computer vision, artificial intelligence algorithms, etc. They can autonomously navigate, identify targets, analyze data, and perform corresponding operations according to preset rules or algorithms. Compared with manual detection, detection robots have higher accuracy, efficiency and stability, which can greatly reduce the risk of missed detection and false detection, and improve the quality and efficiency of security inspection work.

[0003] However, with the increase in the number of robots, how to reasonably and effectively allocate tasks to ensure that each robot can fully play its role has become an urgent problem to be solved. Especially in large security inspection sites, the inspection tasks are heavy and complex, and a single inspection robot can no longer meet the requirements of efficient and comprehensive inspection. In the traditional inspection process, each inspection robot only performs one inspection task and then ends, which will undoubtedly cause a huge waste of resources. The existing robot task allocation methods have the defects of independent task processing, lack of coordination and poor adaptability to state changes. Many task allocation methods assume that each robot can complete the task independently, so these methods usually assign tasks to each robot independently without considering the cooperation between robots. As a result, each robot only focuses on the completion of its own task, and it is difficult to coordinate and solve tasks that require collaboration. In a multi-robot system, some robots may fail during the execution of the task, and the lack of adaptability cannot respond to these failures quickly, resulting in a decrease in system efficiency.

[0004] Therefore, how to achieve the collaborative work of multiple robots and optimize the task allocation strategy has become a hot topic and difficulty in the current security inspection field research. Therefore, the present invention proposes a consistency-constrained detection robot task allocation method. Summary of the invention

[0005] The purpose of the present invention is to provide a detection robot task allocation method with consistency constraints for the problem of task allocation when the detection robot detects contraband, realize the collaborative work of multiple detection robots, and optimize the task allocation strategy.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] The detection robot task allocation method based on consistency constraints includes:

[0008] Determine the tasks to be inspected and the number of inspection robots;

[0009] Inputting the to-be-detected tasks and the number of detection robots into a task allocation model to obtain a task allocation result of the detection robot, wherein the task allocation model is constructed based on a consistency constraint algorithm;

[0010] Based on the task allocation result, the detection robot is driven to perform detection work, and during the detection process, the tasks to be detected and the number of detection robots are updated in real time, and continuously input into the task allocation model to update the task allocation result until the detection is completed.

[0011] Optionally, the task allocation model includes a first task allocation model and a second task allocation model, wherein the first task allocation model is used to make optimal allocation based on the task execution time of the detection robots when the number of the detection robots is less than the tasks to be detected; and the second task allocation model is used to make optimal allocation based on the task execution benefits of the detection robots when the number of the detection robots is greater than or equal to the tasks to be detected.

[0012] Optionally, the objective function of the task allocation model is:

[0013]

[0014] Among them, N is the number of detection robots, M is the number of tasks to be detected, and x i is to detect the current state of robot i, p i is the list vector of tasks to be detected that the detection robot i is expected to perform, c ij (x i ,p i ) is the detection robot i execution list p i The profit after that, x ij is a binary decision variable, indicating the task execution of the detection robot i. ij When x is equal to 1, it means that the detection robot i is assigned to perform the detection task j; when x ij When it is equal to 0, it means that the task j to be inspected has not been assigned to the inspection robot i for execution.

[0015] Optionally, the constraints of the task allocation model are:

[0016]

[0017] Among them, m is the upper limit of the execution task of detection robot i, I is the set of detection robots, J is the set of tasks to be detected, and L min The minimum amount of tasks assigned to the detection robot.

[0018] Optionally, allocating tasks to the detection robot based on the task allocation model constructed by the consistency constraint algorithm includes:

[0019] Develop individual task sequences based on each inspection robot’s own conditions;

[0020] All detection robots are required to exchange information, and during the interaction process, they bid for the same task based on the personal task sequence. If the bid is successful, the task is retained in the personal task sequence. If the bid fails, the task and subsequent tasks are removed from the personal task sequence, and the personal task sequence is re-established.

[0021] Optionally, formulating a personal task sequence for the detection robot includes:

[0022] Step 1: Input the task sequence, task execution path, bidder bid set and bidder set of the detection robot;

[0023] Step 2: Determine whether the task sequence of the detection robot reaches the storage limit. If so, proceed to step 2; if not, proceed to step 3.

[0024] Step 3: Check whether the task to be detected exists in the task execution path. If it does, check the next task to be detected; if it does not exist, proceed to step 4;

[0025] Step 4: Determine whether the value of the task to be tested exceeds the single highest bid in the bid set of the winning bidder. If so, proceed to step 5; if not, proceed to step 6.

[0026] Step 5: Check whether the assigned robots of the tasks to be tested in the winning bidder set have reached the upper limit. If so, check the next task to be tested; if not, proceed to step 6;

[0027] Step 6: According to the principle of the highest benefit of the task execution path, insert the task to be detected into the best position in the task execution path, and record the current position information, the benefit value obtained at the current position and the task start time;

[0028] Step 7: Enter the next task to be detected, and go to step 2 until the current position information, the current position benefit value and the task start time of all tasks to be detected are obtained;

[0029] Step 8: According to the benefit values ​​obtained at the current positions of all tasks to be detected, the task to be detected with the largest benefit value is added to the task sequence, and the task execution path, the bidder bid set and the bidder set are updated according to the corresponding information;

[0030] Step 9, enter the next detection robot and go to step 1 until the traversal of all detection robots is completed, and output the updated task sequence, task execution path, bidder bid set and bidder set of the detection robot.

[0031] Optionally, in step 6, inserting the task to be detected into the optimal position in the task execution path according to the principle of maximum benefit of the task execution path includes:

[0032] Randomly determining an insertion position of the task to be detected in the task execution path;

[0033] The benefits at the insertion position and the position before the insertion position are calculated. If the benefit at the insertion position is greater than the benefit at the position before the insertion position, the insertion position is the best insertion position in the execution path. If the benefit at the insertion position is less than the benefit at the position before the insertion position, the insertion position is changed, and the benefits are calculated and compared again until the best insertion position is found.

[0034] The beneficial effects of the present invention are:

[0035] The present invention adopts a consistency constraint algorithm to allocate detection tasks, which better meets actual needs after allocation; during the execution of detection tasks, the allocation of detection tasks can be dynamically adjusted to greatly reduce resource waste; it has good versatility and can be applied to most task allocation problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 A flow chart of a method for assigning tasks to a detection robot with consistency constraints according to an embodiment of the present invention;

[0038] Figure 2 The detection robot task allocation result of scenario one of the embodiment of the present invention, wherein (a) is the detection robot path time diagram of scenario one, and (b) is the detection robot task timeline of scenario one;

[0039] Figure 3 The detection task allocation result of scenario 2 of the embodiment of the present invention, wherein (a) is the detection robot path time diagram of scenario 2, and (b) is the detection robot task timeline of scenario 2;

[0040] Figure 4 This is the detection task allocation result of scene three of the embodiment of the present invention, where (a) is the detection robot path time diagram of scene two, and (b) is the detection robot task timeline of scene two. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] This embodiment provides a method for assigning tasks to detection robots with consistency constraints, including:

[0044] Determine the tasks to be detected and the number of detection robots, input the tasks to be detected and the number of detection robots into a task allocation model, and obtain the task allocation results of the detection robots, wherein the task allocation model is constructed based on a consistency constraint algorithm;

[0045] Based on the task allocation results, the inspection robot is driven to perform inspection work, and the tasks to be inspected and the number of inspection robots are updated in real time during the inspection process. The tasks are continuously input into the task allocation model, and the task allocation results are updated until the inspection is completed.

[0046] Among them, the task allocation model includes a first task allocation model and a second task allocation model. The first task allocation model is used to make optimal allocation based on the task execution time of the detection robots when the number of detection robots is less than the tasks to be detected; the second task allocation model is used to make optimal allocation based on the task execution benefits of the detection robots when the number of detection robots is greater than or equal to the tasks to be detected.

[0047] Specifically, this embodiment uses a consistency algorithm to allocate detection tasks, which better meets actual needs after allocation; during the execution of detection tasks, the allocation of detection tasks can be dynamically adjusted to greatly reduce resource waste; it has good versatility and can be applied to most task allocation problems.

[0048] Furthermore, the objective function of the task allocation model is:

[0049]

[0050] Among them, N is the number of detection robots, M is the number of tasks to be detected, and x i is to detect the current state of robot i, p i is the list vector of tasks to be detected that the detection robot i is expected to perform, c ij (x i ,p i ) is the detection robot i execution list p i The profit after that, x ij is a binary decision variable, indicating the task execution of the detection robot i. ij When x is equal to 1, it means that the detection robot i is assigned to perform the detection task j; when x ij When it is equal to 0, it means that the task j to be inspected has not been assigned to the inspection robot i for execution.

[0051] Furthermore, the constraints of the task allocation model are:

[0052]

[0053] Among them, m is the upper limit of the execution task of detection robot i, I is the set of detection robots, J is the set of tasks to be detected, and L min The minimum amount of tasks assigned to the detection robot.

[0054] In this embodiment, the ultimate goal is to obtain the maximum benefit from executing the task. The consistency constraint algorithm iterates to obtain a conflict-free task allocation plan to maximize the overall benefit of the detection robot. The so-called conflict-free means ensuring that each detection robot executes one task at a time, and each task can only be assigned to one detection robot.

[0055] Furthermore, the task allocation model constructed based on the consistency constraint algorithm allocates tasks to the detection robots, including: formulating a personal task sequence based on the conditions of each detection robot, exchanging information among all the detection robots, bidding for the same task based on the personal task sequence during the interaction, retaining the task in the personal task sequence if the bid is successful, and removing the task and subsequent tasks from the personal task sequence if the bid fails, and re-formulating the personal task sequence.

[0056] Specifically, the individual task sequence for developing the inspection robot includes:

[0057] Step 1: Input the task sequence, task execution path, bidder bid set and bidder set of the detection robot;

[0058] Step 2: Determine whether the task sequence of the detection robot reaches the storage limit. If so, proceed to step 2; if not, proceed to step 3.

[0059] Step 3: Check whether the task to be detected exists in the task execution path. If so, check the next task to be detected; if not, proceed to step 4.

[0060] Step 4: Determine whether the value of the task to be tested exceeds the single highest bid in the bid set of the winning bidder. If it exceeds, proceed to step 5; if not, proceed to step 6;

[0061] Step 5: Check whether the assigned robots of the tasks to be tested in the winning bidder set have reached the upper limit. If so, check the next task to be tested; if not, proceed to step 6;

[0062] Step 6: According to the principle of the highest benefit of the task execution path, insert the task to be detected into the best position in the task execution path, and record the current position information, the benefit value obtained at the current position and the task start time;

[0063] Step 7: Enter the next task to be detected, and go to step 2 until the current position information, the current position benefit value and the task start time of all tasks to be detected are obtained;

[0064] Step 8: According to the benefit values ​​obtained at the current positions of all tasks to be detected, the task to be detected with the largest benefit value is added to the task sequence, and the task execution path, the bidder bid set and the bidder set are updated according to the corresponding information;

[0065] Step 9, enter the next detection robot and go to step 1 until the traversal of all detection robots is completed, and output the updated task sequence, task execution path, bidder bid set and bidder set of the detection robot.

[0066] Among them, in step 6, according to the principle of the highest benefit of the task execution path, the best position for inserting the task to be detected into the task execution path includes:

[0067] The insertion position of the task to be detected is randomly determined in the task execution path;

[0068] Calculate the benefits at the insertion position and the position before the insertion position. If the benefit at the insertion position is greater than the benefit at the position before the insertion position, the insertion position is the best insertion position in the execution path. If the benefit at the insertion position is less than the benefit at the position before the insertion position, change the insertion position, and calculate and compare the benefits again until the best insertion position is found.

[0069] Combine the following Figure 1-Figure 4The consistency constraint detection robot task allocation method proposed in this embodiment is specifically described and applied, including:

[0070] Step 1: Confirm the number of prohibited items to be detected and the number of detection robots, and determine the relationship between the two;

[0071] Step 2: If the number of robots is less than the number of contraband: Although the task execution capabilities provided by the detection robots meet the existing operational requirements, the consistency bundle algorithm can allocate tasks based on the length of time each robot takes to perform the detection. That is, in the process, the detection robot selects the target with the greatest time benefit to complete the detection requirements;

[0072] Step 3: If the number of robots is equal to the number of contraband items: During the detection and allocation process of the consistency beam algorithm for contraband items, the detection robot can be selected to perform the detection task based on the information of contraband items during the task. The robot with the greatest benefit can be selected to complete the detection. For example, after detection robot 1 completes detection task 1, the distance to detection task 2 is closer than that of detection robot 2, so detection robot 1 can continue to complete detection task 2, and detection robot 2 can perform other detection tasks or not arrange detection;

[0073] Step 4: If the number of robots is greater than the number of contraband items: Same as step 3;

[0074] Step 5: The detection robot performs the contraband detection task according to the allocation plan of steps 2 to 4 until the task is completed.

[0075] The traditional method is that each inspection robot only performs one inspection task and then ends, which will undoubtedly cause a huge waste of resources. In comparison, according to the inspection robot task allocation method based on the consistency constraint algorithm, it is more practical to allocate tasks. The allocation of inspection tasks can be dynamically adjusted during the inspection process, greatly reducing resource waste.

[0076] Assuming that there are N detection robots, corresponding to M prohibited items, the consistency constraint algorithm iterates to obtain a conflict-free task allocation plan to maximize the overall benefit of the detection robots. The so-called conflict-free means ensuring that each detection robot performs one task at a time, and each task can only be assigned to one detection robot. Under the premise of meeting its own conditions, assuming that each detection robot performs at most m tasks, the consistency constraint algorithm will allocate at least L detection robots min =min{M, N×m} tasks.

[0077] The objective function of the algorithm is:

[0078]

[0079] The objective function is the overall benefit obtained by executing the current running task allocation scheme, where p i is the task list vector that robot i is expected to perform, c ij (x i ,p i ) is the list of tasks executed by robot i p i The profit after that, x ij It is a binary decision variable, which indicates the task execution status of robot i. When it is equal to 1, robot i is assigned to execute task j; when it is equal to 0, task j is not assigned to robot i.

[0080] The constraints are:

[0081]

[0082] Among them, I is the set of robots, and J is the set of tasks to be performed: I = {1, 2, ..., N}, J = {1, 2, ..., M}.

[0083] The consistency constraint algorithm consists of two stages. In the first stage, i.e., task formulation, each robot formulates its own task sequence based on its own conditions. In the second stage, i.e., conflict resolution or consensus, each robot establishes information exchange with surrounding robots to determine the ownership of the same bidding task. The following vectors also need to be established in the first stage algorithm:

[0084] (1) Task sequence vector b i (beam vector): This vector is defined as in is the task sequence currently won by the robot, n t That is, the number of tasks. The tasks in the task sequence are arranged according to priority. Indicates that robot i did not bid for the task.

[0085] (2) Task execution path vector p i :This vector is obtained by expanding the task list vector into a task path with a sequence, which contains the moving path between each task point and the related path optimization information, defined as p i ={p i1 ,p i2 ,……,p in},in Execute b for robot i running this path i The corresponding path vector.

[0086] (3) Time set τ i :The set is defined as τ i ={τ i1 ,τ i2 ,……,τin},in For execution time.

[0087] (4) The set of successful bidders z i :The set is defined as z i ={z i1 ,z i2 ,……,z in},in Which robot does robot i think is performing task n?

[0088] (5) The winning bidder’s bid set y i : The set is defined as y i ={y i1 ,y i2 ,……,y in},in For all robots to task n s The highest bid in the auction.

[0089] (6) Timestamps i : The set is defined as s i ={s i1 ,s i2 ,……,s in}, is the time when robot i receives the latest information of robot n from the communication network.

[0090] The following is a detailed introduction to the two phases of the consistency bundle algorithm, including:

[0091] (1) Task formulation phase: Each robot continuously adds and adjusts tasks in its own sequence to build the task sequence it is expected to perform.

[0092] Input: Task sequence b of robot i after the last iteration i 、Task execution path p i , winning bidder bid set y i and the set of winning bidders z i ;

[0093] Step 1: Robot i determines task sequence b i Whether the storage limit has been reached, if not, go to Step 2, if so, go to Step 9;

[0094] Step 2: Robot i checks whether task j exists in task path p i If it exists, then j=j+1, if it does not exist, then go to Step 3;

[0095] Step 3: Determine whether the value of task j exceeds the robot's highest single bid. If not, proceed to Step 5. If so, proceed to Step 4.

[0096] Step 4: Check z i Check whether the assigned robots of task j have reached the upper limit. If so, j=j+1. If not, go to Step 5.

[0097] Step 5: Insert task j into the best position in the sequence, and add task j to the task execution path of robot i. i , for position k=1:|p i |+1, calculate the benefit of the entire path after inserting to position k;

[0098] Step 6: Determine the benefits of position k and k-1. If k is greater than k-1, record the current position information, the benefit value obtained at the current position and the time when the task starts. If k is less than k-1, then k=k+1 and go to Step 5.

[0099] Step 7: j = j + 1, go to Step 2;

[0100] Step 8: Compare each target and insert p i The maximum benefit in b is added to the target with the maximum benefit i and update b i 、p i ,y i and z i ;

[0101] Step 9: i=i+1, then determine whether all members of the robot have been traversed. If yes, go to the second stage, otherwise, continue to Step 1.

[0102] Output: Task sequence b of robot i after this iteration i 、Task execution path p i , winning bidder bid set y i and the set of winning bidders z i .

[0103] (2) Consensus phase: Adjacent robots communicate with each other and obtain each other’s b i 、p i 、s i To determine whether the bid is successful, if the bid for task j is successful, all surrounding robots including those bidding for task j must be modified, j and tasks added later must be released, and the robot must be re-tasked.

[0104] Description of scene application and application effect:

[0105] Assuming that 10 prohibited items need to be detected, the number of detection robots is selected as 5, 10, and 18 as application scenarios.

[0106] Scenario 1: The number of detection robots is 5, which is less than the number of prohibited items. The result is as follows: Figure 2 As shown in Figure 1, (a) is the detection robot path time diagram for scenario 1, and (b) is the detection robot task timeline for scenario 1. Figure 2 It can be seen that all five detection robots followed the information preferential selection method, and there was no situation where a robot detected a prohibited item and then returned to the starting point as in the traditional method.

[0107] Scenario 2: The number of detection robots is 10, which is equal to the number of prohibited items. The results are as follows: Figure 3 As shown in Figure 2, (a) is the detection robot path time diagram for scenario 2, and (b) is the detection robot task timeline for scenario 2. Figure 3 It can be seen that all 10 detection robots followed the principle of information optimization, and there was no situation where a robot detected a contraband and then returned to the starting point as in the traditional method. According to the principle of dynamic information optimization, detection robots 4, 5, 9, and 10 were not assigned detection tasks.

[0108] Scenario 3: The number of detection robots is 18, which is greater than the number of prohibited items. The results are as follows: Figure 4 As shown in Figure 2, (a) is the detection robot path time diagram for scenario 2, and (b) is the detection robot task timeline for scenario 2. Figure 4 It can be seen that all 18 detection robots followed the principle of information optimization, and there was no situation where a robot detected a contraband and then returned to the starting point as in the traditional method. According to the principle of dynamic information optimization, detection robots 4, 5, 6, 7, 8, 9, 13-18 were not assigned detection tasks.

[0109] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for allocating tasks of a detection robot based on consistency constraints, characterized in that: include: Determine the tasks to be inspected and the number of inspection robots; Inputting the to-be-detected tasks and the number of detection robots into a task allocation model to obtain a task allocation result of the detection robot, wherein the task allocation model is constructed based on a consistency constraint algorithm; Based on the task allocation result, the detection robot is driven to perform detection work, and during the detection process, the tasks to be detected and the number of detection robots are updated in real time, and continuously input into the task allocation model to update the task allocation result until the detection is completed.

2. The method for assigning tasks to detection robots based on consistency constraints according to claim 1, characterized in that: The task allocation model includes a first task allocation model and a second task allocation model, wherein the first task allocation model is used to make optimal allocation based on the task execution time of the detection robots when the number of the detection robots is less than the tasks to be detected; and the second task allocation model is used to make optimal allocation based on the task execution benefits of the detection robots when the number of the detection robots is greater than or equal to the tasks to be detected.

3. The method for assigning tasks to detection robots based on consistency constraints according to claim 1, characterized in that: The objective function of the task allocation model is: Among them, N is the number of detection robots, M is the number of tasks to be detected, and x i is to detect the current state of robot i, p i is the list vector of tasks to be detected that the detection robot i is expected to perform, c ij (x i ,p i ) is the detection robot i execution list p i The profit after that, x ij is a binary decision variable, indicating the task execution of the detection robot i. ij When x is equal to 1, it means that the detection robot i is assigned to perform the detection task j; when x ij When it is equal to 0, it means that the task j to be inspected has not been assigned to the inspection robot i for execution.

4. The method for assigning tasks to a consistency-constrained detection robot according to claim 3, characterized in that: The constraints of the task allocation model are: Among them, m is the upper limit of the execution task of detection robot i, I is the set of detection robots, J is the set of tasks to be detected, and L min The minimum amount of tasks assigned to the detection robot.

5. The method for assigning tasks to detection robots based on consistency constraints according to claim 1, characterized in that: The task allocation model constructed based on the consistency constraint algorithm allocates tasks to the detection robot, including: Develop individual task sequences based on each inspection robot’s own conditions; All detection robots are required to exchange information, and during the interaction process, they bid for the same task based on the personal task sequence. If the bid is successful, the task is retained in the personal task sequence. If the bid fails, the task and subsequent tasks are removed from the personal task sequence, and the personal task sequence is re-established.

6. The method for assigning tasks to a consistency-constrained detection robot according to claim 5, characterized in that: Formulating the individual task sequence of the inspection robot includes: Step 1: Input the task sequence, task execution path, bidder bid set and bidder set of the detection robot; Step 2: Determine whether the task sequence of the detection robot reaches the storage limit. If so, proceed to step 2; if not, proceed to step 3. Step 3: Check whether the task to be detected exists in the task execution path. If it does, check the next task to be detected; if it does not exist, proceed to step 4; Step 4: Determine whether the value of the task to be tested exceeds the single highest bid in the bid set of the winning bidder. If so, proceed to step 5; if not, proceed to step 6. Step 5: Check whether the assigned robots of the tasks to be tested in the winning bidder set have reached the upper limit. If so, check the next task to be tested; if not, proceed to step 6; Step 6: According to the principle of the highest benefit of the task execution path, insert the task to be detected into the best position in the task execution path, and record the current position information, the benefit value obtained at the current position and the task start time; Step 7: Enter the next task to be detected, and go to step 2 until the current position information, the current position benefit value and the task start time of all tasks to be detected are obtained; Step 8: According to the benefit values ​​obtained at the current positions of all tasks to be detected, the task to be detected with the largest benefit value is added to the task sequence, and the task execution path, the bidder bid set and the bidder set are updated according to the corresponding information; Step 9, enter the next detection robot and go to step 1 until the traversal of all detection robots is completed, and output the updated task sequence, task execution path, bidder bid set and bidder set of the detection robot.

7. The method for assigning tasks to detection robots based on consistency constraints according to claim 6, characterized in that: In step 6, according to the principle of the highest benefit of the task execution path, inserting the task to be detected into the optimal position of the task execution path includes: Randomly determining an insertion position of the task to be detected in the task execution path; The benefits at the insertion position and the position before the insertion position are calculated. If the benefit at the insertion position is greater than the benefit at the position before the insertion position, the insertion position is the best insertion position in the execution path. If the benefit at the insertion position is less than the benefit at the position before the insertion position, the insertion position is changed, and the benefits are calculated and compared again until the best insertion position is found.

Images