Unmanned cooperative task allocation method and system applied to power plant protection equipment
By adopting the capability matching bid and sales contract mechanism in the inspection of power plant protection equipment, the problems of unbalanced task allocation and high computational complexity are solved, and the efficient completion of inspection tasks and optimal resource allocation are achieved.
Patent Information
- Application Number
- CN202410910375.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-09
- Publication Date
- 2024-11-26
AI Technical Summary
The existing technology has problems such as unbalanced task allocation, fixed inspection routes, high task complexity and poor real-time performance in the inspection of power plant protection equipment, resulting in inefficiency and waste of resources.
The capability matching bid mechanism is adopted, and the capabilities, cost and time constraints of the inspection robot are comprehensively considered. Task allocation and resource allocation are carried out through the sales contract mechanism and the exchange contract mechanism to ensure flexible scheduling and efficient completion of tasks.
It improves the completion efficiency and quality of inspection tasks, ensures full utilization of resources and smooth completion of tasks, reduces calculation complexity and parameter sensitivity, and is suitable for fast response scenarios.
Smart Images

Figure CN119026826A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent agent task allocation, and in particular to an unmanned collaborative task allocation method and system applied to power plant protection equipment. Background Art
[0002] In the operation of power plants, the normal operation of protection equipment is crucial to the safety and stability of the power system. In order to promptly detect and deal with potential faults or problems, regular inspections of power plant protection equipment are essential. However, traditional manual inspections have problems such as low efficiency and high labor intensity, and there are safety risks under extreme weather conditions. In recent years, with the development of unmanned technology and artificial intelligence, inspection robots have become a new inspection method, which can effectively improve inspection efficiency, reduce labor costs, and replace manual inspections under extreme conditions, significantly improving the safety and stability of inspections.
[0003] The unmanned collaborative task allocation of inspection robots is an important part of the safety control of power plants. At present, there are still some challenges and problems in the task allocation for power plant protection equipment: First, the current task allocation mechanism may lead to uneven distribution of inspection tasks. Some areas may be inspected repeatedly while other areas are ignored, resulting in waste of resources and low task efficiency; second, existing inspection robots often inspect along pre-set fixed paths and cannot dynamically adjust the inspection route according to real-time conditions, resulting in low inspection efficiency; at the same time, when multiple inspection robots perform tasks at the same time, task conflicts and competition may occur, affecting the efficiency and accuracy of task execution; in addition, the existing task allocation system cannot flexibly adjust the task allocation of inspection robots according to environmental changes and task requirements, resulting in uneven distribution of inspection tasks, greatly reducing inspection efficiency and causing waste of resources.
[0004] For example, Chinese patent CN113011780A discloses a task allocation method for collaborative power inspection of multiple UAVs, including constructing a task allocation model for collaborative power inspection of multiple UAVs, which uses a genetic algorithm to solve the task allocation model, and can obtain an overall optimal solution that achieves the optimization goal and satisfies the constraints; for another example, Chinese patent CN109872010A discloses a task allocation method for a substation inspection robot, including constructing a task point set and setting the number of security points of the inspection robot, which solves the site selection optimization model to obtain the security point location and security point task allocation; for another example, Chinese patent CN116307648A discloses a method and system for assigning and traversal optimization of inspection tasks for multiple inspection robots in a substation, including using the mobile energy consumption and mobile time of the inspection robot to establish a target model for multi-robot inspection task allocation and traversal optimization, which can obtain the individual with the highest fitness after mutation and iteration of the genetic algorithm. However, the above-mentioned existing technologies all have the following shortcomings in the process of specific application: the existing technologies are too dependent on environmental data, which may lead to inaccurate inspection strategies; the genetic algorithm and particle swarm algorithm used have high computational complexity when processing large-scale tasks, resulting in too long solution time; in addition, the existing technologies are sensitive to parameter settings and have poor real-time performance, and are not suitable for scenarios that require rapid response.
[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention
[0006] In response to the problems in the related technology, the present invention proposes an unmanned collaborative task allocation method and system for power plant protection equipment, which comprehensively considers factors such as the capability, cost, task adaptability and time constraints of the inspection robot, matches the bids of the inspection robots that collaboratively perform tasks by means of capability matching bids, and uses capability satisfaction to make full use of inspection resources to improve the efficiency and quality of task completion, thereby solving the problems in the prior art of over-reliance on environmental data, the high computational complexity of the genetic algorithm and particle swarm algorithm used when processing large-scale tasks, the long solution time, the sensitivity to parameter settings, and the poor real-time performance.
[0007] To this end, the specific technical solution adopted by the present invention is as follows:
[0008] According to one aspect of the present invention, there is provided an unmanned collaborative task allocation method applied to power plant protection equipment, the unmanned collaborative task allocation method applied to power plant protection equipment comprising the following steps:
[0009] S1. Based on objective and subjective evaluation indicators, comprehensively evaluate the utility of each inspection robot and generate a bid. The objective evaluation indicators include time factors and cost factors, and the subjective evaluation indicators include task adaptability and capability satisfaction.
[0010] S2. According to the current task, the inspection robots are divided into tenderers and bidders, and the purchasing power of the bidders is judged by the contract rules and the bids are submitted;
[0011] S3. Based on the minimum capability satisfaction threshold, evaluate the capability satisfaction of the bidders who submitted the tender documents, and dispatch additional inspection robots to form a collaborative task group to deliver the tender documents of the collaborative task group;
[0012] S4. Compare the utility of the submitted bids and select the winning bidder, use the winning bidder to sign a contract with the tenderer, and perform the assigned tasks.
[0013] Furthermore, based on the objective evaluation index and the subjective evaluation index, the utility of each inspection robot is comprehensively evaluated and the bid is generated, including the following steps:
[0014] S11. Calculate objective evaluation indicators of the inspection robot based on time factors and cost factors;
[0015] S12. Calculate the subjective evaluation index of the inspection robot based on task suitability and capability satisfaction;
[0016] S13. Use the utility formula to integrate the objective evaluation index and the subjective evaluation index to obtain the bid document.
[0017] Furthermore, the expressions for calculating the objective evaluation index of the inspection robot based on time factors and cost factors are:
[0018]
[0019] Where Time i represents the time factor for executing task i, WT i Indicates the waiting time for executing task i, NT iw represents the time required for the inspection robot w to perform task i, and n represents the number of inspection robots participating in the task;
[0020]
[0021] Where, Value i represents the cost of the inspection robot participating in task i, Price w represents the price of inspection robot w, and n represents the number of inspection robots participating in task i.
[0022] Furthermore, the expressions for calculating the subjective evaluation index of the inspection robot based on task adaptability and capability satisfaction are:
[0023]
[0024] In the formula, Adaptationi represents the suitability of the task, TComplexity represents the average autonomy of the inspection robots participating in the i-th task. i Indicates the complexity of the task;
[0025]
[0026] In the formula, CS i Indicates ability satisfaction, equal to task T i The average value of the sum of the various ability satisfactions; j Represents task T i The requirement for the jth capability of the inspection robot to perform the task, Indicates the execution of task T i The size of the j-th capability of the inspection robot, m i Represents task T i The number of capabilities for which the capability requirement is not 0.
[0027] Furthermore, the expression of integrating objective evaluation index and subjective evaluation index by using utility formula is:
[0028] U i =ω i1 CS i +ω i2 Adaptation i +ω i3 Time i -ω i4 Value i ;
[0029] In the formula, CS i Indicates ability satisfaction, Adaptation i Indicates task suitability, Time i Indicates the time factor for executing task i, Value i represents the cost of the inspection robot participating in task i, w i1 、w i2 、w i3 and i4 are weight parameters, which respectively represent the importance of task capability satisfaction, task adaptation, time factor and cost factor.
[0030] Furthermore, the inspection robots are divided into tenderers and bidders according to the current tasks, and the purchasing power of the bidders is judged by using the contract rules and the bids are submitted, including the following steps:
[0031] S21. Divide the inspection robot into tenderers and bidders according to current task requirements, and use the tenderer to publish task information to the bidders;
[0032] S22, based on the received task information, using the sales contract rules to determine the purchasing power of each bidder, and sending the bidding information to the tenderer according to the determination result;
[0033] S23. Use the exchange contract rules to determine the purchasing power of each bidder, and send bidding information to the bidder based on the determination result.
[0034] Further, based on the received task information, the purchasing power of each bidder is judged by using the sales contract rules, and the bidding information is sent to the tenderer according to the judgment result, including the following steps:
[0035] S221. Classify the purchasing power level of each bidder based on the task set of the inspection robot itself;
[0036] S222, determining the task requirement level based on the received task information;
[0037] S223. Traverse the purchasing power level of each bidder. If the purchasing power level of the bidder is greater than or equal to the task requirement level, send the bid to the tenderer. If the purchasing power level of the bidder is less than the task requirement level, execute step S23.
[0038] Further, using the exchange contract rules to determine the purchasing power of each bidder and sending bidding information to the bidder according to the determination result includes the following steps:
[0039] S231. The bidder pre-exchanges the tasks in the task set of its own equipment with the bidding tasks in sequence;
[0040] S232, judging whether the bidders after the pre-exchange meet both the distance constraint and the time constraint, and if not, giving up submitting the bid;
[0041] S233, calculating the utility of the two inspection robots after the pre-exchange, regenerating the bidder's bid and sending it to the tenderer, wherein the expression for calculating the utility of the two inspection robots after the pre-exchange is:
[0042]
[0043] In the formula, It represents the utility of the two inspection robots after the pre-exchange, T i Indicates the bidding task, represents the pth task in the task set of bidder j’s own equipment, U j represents the utility of bidder j performing the original task set, U z represents the utility of bidder z performing the original task set, Indicates that bidder j gives up the task in his task set Execute bidding tasks i The utility of Indicates that the bidder z performs task T i The latter effect.
[0044] Furthermore, based on the minimum capability satisfaction threshold, the capability satisfaction of the bidders who submitted the bid is evaluated, and additional inspection robots are dispatched to form a collaborative task group. The bid submission of the collaborative task group includes the following steps:
[0045] S31, traversing the bidders who have submitted bids, and comparing the capability satisfaction with the minimum capability satisfaction threshold;
[0046] S32. Based on the comparison result, if the bidder's capability satisfaction is less than the preset minimum capability satisfaction threshold, an additional inspection robot is dispatched to form a collaborative bidding group, and the utility of each collaborative bidding group is calculated;
[0047] S33. Compare the calculation results, mark the collaborative bidding group with the largest utility as the successful bidder, and release the bid failure information to the unsuccessful bidders.
[0048] According to another aspect of the present invention, there is also provided an unmanned collaborative task allocation system applied to power plant protection equipment, the unmanned collaborative task allocation system applied to power plant protection equipment comprising:
[0049] Comprehensive evaluation module, which is used to comprehensively evaluate the utility of each inspection robot based on objective evaluation indicators and subjective evaluation indicators, and generate bids. The objective evaluation indicators include time factors and cost factors, and the subjective evaluation indicators include task adaptability and capability satisfaction.
[0050] The sales contract module is used to divide the inspection robots into tenderers and bidders according to the current tasks, and use the contract rules to judge the purchasing power of the bidders and submit the bids;
[0051] The collaborative task module is used to evaluate the capability satisfaction of the bidders who submitted the tender documents based on the minimum capability satisfaction threshold, and to dispatch additional inspection robots to form a collaborative task group to deliver the tender documents of the collaborative task group;
[0052] The bid matching module is used to compare the utility of the submitted bids and select the winning bidder, use the winning bidder to sign a contract with the tenderer, and execute the assigned tasks.
[0053] The beneficial effects of the present invention are:
[0054] The present invention adopts a sales contract mechanism that introduces an exchange contract to distribute tasks, and uses capability satisfaction to match bids for inspection robots that collaboratively perform tasks, thereby realizing collaborative work tasks between inspection robots; at the same time, the present invention comprehensively considers multiple factors such as capability, time, task adaptability, and inspection robot cost, and effectively distributes tasks to inspection robots; secondly, the exchange contract is introduced to improve the sales contract mechanism, and flexible task scheduling and resource allocation are achieved by setting a minimum capability satisfaction threshold to ensure smooth completion of tasks; in addition, according to the different capabilities of inspection robots and the differences in requirements for task types, a capability matching bid method is adopted to accurately distribute tasks to suitable robots, thereby improving the efficiency and quality of task completion. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] 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.
[0056] Figure 1 is a flow chart of an unmanned collaborative task allocation method applied to power plant protection equipment according to an embodiment of the present invention;
[0057] Figure 2 is a specific implementation diagram of the unmanned collaborative task allocation method applied to power plant protection equipment according to an embodiment of the present invention;
[0058] Figure 3 It is a principle block diagram of an unmanned collaborative task allocation system applied to power plant protection equipment according to an embodiment of the present invention. DETAILED DESCRIPTION
[0059] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.
[0060] According to an embodiment of the present invention, an unmanned collaborative task allocation method and system applied to power plant protection equipment are provided.
[0061] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1-Figure 2As shown, according to one embodiment of the present invention, an unmanned collaborative task allocation method applied to power plant protection equipment is provided, and the unmanned collaborative task allocation method applied to power plant protection equipment includes the following steps:
[0062] S1. Based on objective and subjective evaluation indicators, comprehensively evaluate the utility of each inspection robot and generate a bid. The objective evaluation indicators include time factors and cost factors, and the subjective evaluation indicators include task adaptability and capability satisfaction.
[0063] S2. According to the current task, the inspection robots are divided into tenderers and bidders, and the purchasing power of the bidders is judged by the contract rules and the bids are submitted;
[0064] S3. Based on the minimum capability satisfaction threshold, evaluate the capability satisfaction of the bidders who submitted the tender documents, and dispatch additional inspection robots to form a collaborative task group to deliver the tender documents of the collaborative task group;
[0065] S4. Compare the utility of the submitted bids and select the winning bidder, use the winning bidder to sign a contract with the tenderer, and perform the assigned tasks.
[0066] Specifically, in this embodiment, when allocating unmanned collaborative tasks, the main factors to be considered are capability, time, task adaptability and inspection robot cost. The utility U obtained by integrating the above four factors according to the formula is used as the bid for the inspection robot. The capability factor is the calculated capability satisfaction, and the time factor refers to the acceptable waiting time for a task to be performed, which is determined by each inspection robot based on its own task set and the time required to reach the location of the task to be performed.
[0067] Specifically, in this embodiment, the main steps of unmanned collaborative task allocation are as follows: Step 1, the tenderer publishes the tender information; Step 2, the bidder's ability assessment, determines whether to submit a bid based on its own status information, if the bidder has the purchasing power, considers bidding for the purchase and sale contract, if the bidder does not have the purchasing power, determines whether to bid for the exchange contract; Step 3, if the ability satisfaction is lower than the minimum threshold, it will lead to a lower task success rate, consider adding a patrol robot to perform the task in collaboration with the patrol robot; Step 4, calculates the utility of the group of bids, the utility generated by the patrol robot's collaborative execution of the task, that is, the patrol robot's bid for task i, and integrates the objective evaluation index and the subjective evaluation index calculation through the utility formula; Step 5, selects a group of bids with the largest total utility and notifies the bidding results.
[0068] In one embodiment, based on objective evaluation indicators and subjective evaluation indicators, comprehensively evaluating the utility of each inspection robot and generating a bid document includes the following steps:
[0069] S11. Calculate objective evaluation indicators of the inspection robot based on time factors and cost factors;
[0070] S12. Calculate the subjective evaluation index of the inspection robot based on task suitability and capability satisfaction;
[0071] S13. Use the utility formula to integrate the objective evaluation index and the subjective evaluation index to obtain the bid document.
[0072] In one embodiment, the expressions for calculating the objective evaluation index of the inspection robot based on the time factor and the cost factor are:
[0073]
[0074] Where Time i represents the time factor for executing task i, WT i Indicates the waiting time for executing task i, NT iw represents the time required for the inspection robot w to perform task i, and n represents the number of inspection robots participating in the task;
[0075]
[0076] Where, Value i represents the cost of the inspection robot participating in task i, Price w represents the price of inspection robot w, and n represents the number of inspection robots participating in task i.
[0077] Specifically, in this embodiment, the waiting time for executing task i is equal to the latest start time set for the task based on the urgency of the task and experience. The time required for the inspection robot w to execute task i is equal to the distance between the current position of the inspection robot and the task point divided by the speed of the inspection robot. The price of the inspection robot w is obtained directly from the manufacturer.
[0078] In one embodiment, the expressions for calculating the subjective evaluation index of the inspection robot based on task suitability and capability satisfaction are:
[0079]
[0080] In the formula, Adaptation i represents the suitability of the task, TComplexity represents the average autonomy of the inspection robots participating in the i-th task. i Indicates the complexity of the task;
[0081]
[0082]
[0083] In the formula, CS i Indicates ability satisfaction, equal to task T i The average value of the sum of the various ability satisfactions; j Represents task T i The requirement for the jth capability of the inspection robot to perform the task, Indicates the execution of task T i The size of the j-th capability of the inspection robot, m i Represents task T i The number of capabilities for which the capability requirement is not 0.
[0084] Specifically, task suitability is an indicator used to measure the rationality of unmanned collaborative task allocation, that is, to judge which tasks are suitable for allocation to the inspection robot based on the performance of different inspection robots; to determine the required capability requirements based on the task type. In this embodiment, the inspection task requires the inspection robot to have autonomous navigation, data collection and analysis, and security monitoring capabilities. The size of the inspection robot's capabilities is obtained from the various capability parameters of the inspection robot.
[0085] Specifically, in the above embodiment, the autonomy of the inspection robot is divided into levels 1-5 (basic, semi-interactive, interactive, autonomous, and adaptive) according to its autonomous capabilities. The average autonomy of the inspection robot is equal to the sum of the autonomies of the inspection robots participating in the task divided by the number of inspection robots. The task complexity is based on historical task situations and manual experience.
[0086] Specifically, in the above embodiment, the utility calculation of unmanned collaborative task allocation is subject to the constraint of capability satisfaction, and the capability satisfaction CS i Represents task T i The average value of the sum of the various capability satisfactions is equal to the sum of the proportions of the various capabilities provided by the inspection robots actually assigned to perform the task and the task capability requirements after the capabilities are synthesized, divided by m i ; Due to task T i The requirements for the various capabilities of the inspection robot are different. The combined value of the capabilities of each inspection robot divided by the capabilities required for the task can reflect the weight of each capability of the inspection robot in the task.
[0087] Specifically, when matching bids, first perform capability matching, and then calculate the capability vector of each set of matched joint bids through capability synthesis. The following conditions are met: The capabilities of any two bids meet when λ j ≠0,
[0088] In one embodiment, the expression for integrating the objective evaluation index and the subjective evaluation index using the utility formula is:
[0089] U i =ω i1 CS i +ω i2 Adaptation i +ω i3 Time i -ω i4 Value i ;
[0090] In the formula, CS i Indicates ability satisfaction, Adaptation i Indicates task suitability, Time i Indicates the time factor for executing task i, Value i represents the cost of the inspection robot participating in task i, w i1 、w i2 、w i3 and i4 are weight parameters, which respectively represent the importance of task capability satisfaction, task adaptation, time factor and cost factor.
[0091] Specifically, in this embodiment, the weight parameters of task fitness, time, cost and capability satisfaction are dynamically adjusted by introducing a reinforcement learning algorithm. The specific steps are as follows: Step 1 defines the current state of the inspection robot as a state vector S including the characteristics of task fitness, time factor, cost factor and capability satisfaction, defines the adjustment of the weight parameter as an action vector A, and defines the adjustment range of the weight parameter as [-0.1, 0.1], which represents the increase or decrease of each feature weight. The action vector A is the adjustment amount in four dimensions, corresponding to the adjustment of each feature weight; Step 2 selects the Q learning algorithm in reinforcement learning, establishes a Q learning model, and designs a reward R to measure the performance of the inspection robot after performing the task. After each task is executed, the action value Q is updated according to the reward R to learn the optimal weight adjustment strategy. The expression for updating the action value is:
[0092] Q(S,A)'=Q(S,A)+α×(R+γ×maxQ(S',a)-Q(S,A));
[0093] Wherein, Q(S, A)′ represents the updated action value, Q(S, A) represents the Q value of taking action A in state S, α represents the learning rate, which is used to determine the degree of update of the Q value by the newly learned information, R represents the immediate reward after taking action A, γ represents the discount factor, which is used to determine the importance of future rewards relative to immediate rewards; maxQ(S', a) represents the maximum Q value of taking a specific action a in the next state S'; Step 3 collects the state, action and reward data of the inspection robot when performing tasks, uses the collected data to train the Q learning model, deploys the trained model to the inspection robot, and before each task execution, uses the model to predict the best action A according to the current state S, and adjusts the weight parameters according to the action predicted by the model; Step 4 regularly evaluates the model performance, including task completion speed, cost savings and task success rate, and adjusts the reward coefficient and model parameters according to the performance evaluation results.
[0094] In one embodiment, the inspection robots are divided into tenderers and bidders according to the current tasks, and the purchasing power of the bidders is judged by using the contract rules and the bids are submitted, including the following steps:
[0095] S21. Divide the inspection robot into tenderers and bidders according to current task requirements, and use the tenderer to publish task information to the bidders;
[0096] S22, based on the received task information, using the sales contract rules to determine the purchasing power of each bidder, and sending the bidding information to the tenderer according to the determination result;
[0097] S23. Use the exchange contract rules to determine the purchasing power of each bidder, and send bidding information to the bidder based on the determination result.
[0098] In one embodiment, based on the received task information, the purchasing power of each bidder is judged by using the sales contract rules, and the bidding information is sent to the tenderer according to the judgment result, including the following steps:
[0099] S221. Classify the purchasing power level of each bidder based on the task set of the inspection robot itself;
[0100] S222, determining the task requirement level based on the received task information;
[0101] S223. Traverse the purchasing power level of each bidder. If the purchasing power level of the bidder is greater than or equal to the task requirement level, send the bid to the tenderer. If the purchasing power level of the bidder is less than the task requirement level, execute step S23.
[0102] Specifically, in this embodiment, the purchasing power level of each bidder is divided according to the autonomy of the inspection robot, and is divided into levels 1-5 (basic, semi-interactive, interactive, autonomous, and adaptive) according to the autonomy. If the inspection robot does not have a certain capability (such as anomaly detection, high-definition image recording, etc.), then the capability is 0. Then, according to the received task information, the task requirement level is determined based on manual experience.
[0103] Specifically, the sale and purchase contract rule is a method of allocating tasks through "buying and selling" between the tenderer and the bidder. There are four main steps in task allocation based on the sale and purchase contract: Step 1 Task release. When the existing resources of the inspection robot cannot complete the task or a new task is discovered, it will publish the information of the task to the surrounding inspection robots as a tenderer; Step 2 Bidding. The inspection robot that receives the task information will evaluate whether it is capable of completing the task and calculate the utility that can be obtained by completing the task, that is, the bid for the task. If it is capable of performing the task, it will submit a bid to the tenderer and participate in the task bidding; Step 3 Task winning. The tenderer will receive bidding information from other inspection robots, conduct a comprehensive evaluation of these bidding information, select the most suitable bidder, and send the winning result to the bidder, and publish the losing bid information to other bidding inspection robots; Step 4 Task execution. The inspection robot that receives the winning bid information signs a contract with the tenderer, adds the task to its own task set, and returns the task completion information to the tenderer within a certain period of time.
[0104] In one embodiment, using the exchange contract rules to determine the purchasing power of each bidder and sending bidding information to the bidder according to the determination result includes the following steps:
[0105] S231. The bidder pre-exchanges the tasks in the task set of its own equipment with the bidding tasks in sequence;
[0106] S232, judging whether the bidders after the pre-exchange meet both the distance constraint and the time constraint, and if not, giving up submitting the bid;
[0107] Specifically, when one inspection robot among the bidders gives up bidding for the task, there are other inspection robots that can bid for the task. If all inspection robots among the bidders do not bid for the task, it means that the current inspection robot has too many tasks to perform and cannot perform this task, and manual processing is required.
[0108] S233, calculating the utility of the two inspection robots after the pre-exchange, regenerating the bidder's bid and sending it to the tenderer, wherein the expression for calculating the utility of the two inspection robots after the pre-exchange is:
[0109]
[0110] In the formula, It represents the utility of the two inspection robots after the pre-exchange, T i Indicates the bidding task, represents the pth task in the task set of bidder j’s own equipment, U j represents the utility of bidder j performing the original task set, U z represents the utility of bidder z performing the original task set, Indicates that bidder j gives up the task in his task set Execute bidding tasks i The utility of Indicates that the bidder z performs task T i The latter effect.
[0111] Specifically, when the inspection robot participating in the bidding is unable to perform the task, it cannot sign a sales contract with the tenderer. Considering the new contract rule - exchange contract, the bidder will pre-exchange the tasks in its own task set with the bidding tasks in turn, calculate the increased utility after the pre-exchange, and then select the task with the largest increase in utility after the pre-exchange for task exchange based on the calculation results, so as to achieve the purpose of improving utility.
[0112] Specifically, the steps of the task allocation algorithm based on the exchange contract are as follows: Step 1 The tenderer publishes the tender information; Step 2 After receiving the tender information, the bidder uses the exchange contract to bid and assigns the tasks in its task set to the tenderer. Tendering Tasks i After the exchange, determine whether the distance constraint and time constraint are met after the two tasks are exchanged. If not, give up the bid exchange contract. The distance constraint is to determine whether the distance that the inspection robot can travel with the remaining power is greater than the distance to the task point; the time constraint is to determine whether the inspection robot can reach the task point within the specified time range; Step 3 calculates the utility improved by the two inspection robots, finds out the task with the largest utility improvement after the inspection robot task is exchanged, and goes to Step 4, otherwise give up the bid exchange contract; Step 4 if the inspection robot's improved utility is greater than the minimum contract threshold preset based on experience, submit the task exchange contract bid to the tenderer, otherwise give up the bid exchange contract; Step 5 the tenderer evaluates the received bids, selects the bidder with the highest bid in this round of bidding, and publishes the bid failure information to other bidders.
[0113] In one embodiment, based on the minimum capability satisfaction threshold, the capability satisfaction of the bidders who submitted the bid is evaluated, and additional inspection robots are dispatched to form a collaborative task group. Submitting the bid of the collaborative task group includes the following steps:
[0114] S31, traversing the bidders who have submitted bids, and comparing the capability satisfaction with the minimum capability satisfaction threshold;
[0115] S32. Based on the comparison result, if the bidder's capability satisfaction is less than the preset minimum capability satisfaction threshold, an additional inspection robot is dispatched to form a collaborative bidding group, and the utility of each collaborative bidding group is calculated;
[0116] S33. Compare the calculation results, mark the collaborative bidding group with the largest utility as the successful bidder, and release the bid failure information to the unsuccessful bidders.
[0117] Specifically, the utility of the current inspection robot and other inspection robots performing the task in collaboration is calculated according to the utility formula, and the inspection robot with the largest utility is selected to perform the task together; if the utility calculated for the collaborative execution of the task is lower than the utility of a single robot performing the task, the collaborative execution of the task is not considered, and the task is performed by a single robot.
[0118] In order to facilitate understanding of the above technical solution of the present invention, a specific description is given below taking a certain X as an example:
[0119] like Figure 3 As shown, according to another embodiment of the present invention, an unmanned collaborative task allocation system applied to power plant protection equipment is also provided, and the unmanned collaborative task allocation system applied to power plant protection equipment includes:
[0120] Comprehensive evaluation module 1 is used to comprehensively evaluate the utility of each inspection robot based on objective evaluation indicators and subjective evaluation indicators, and generate a bid, wherein the objective evaluation indicators include time factors and cost factors, and the subjective evaluation indicators include task adaptability and capability satisfaction;
[0121] The sales contract module 2 is used to divide the inspection robot into tenderers and bidders according to the current task, and use the contract rules to judge the purchasing power of the bidders and submit the bids;
[0122] Collaborative task module 3 is used to evaluate the capability satisfaction of bidders who submit bids based on the minimum capability satisfaction threshold, and to dispatch additional inspection robots to form a collaborative task group to deliver bids of the collaborative task group;
[0123] The bid matching module 4 is used to compare the utility of the submitted bids and select a successful bidder, use the successful bidder to sign a contract with the tenderer, and perform the assigned tasks.
[0124] To sum up, with the help of the above-mentioned technical scheme of the present invention, task allocation is carried out by introducing the sales contract mechanism of exchange contracts, and the capability satisfaction is used to match the bids of the inspection robots that collaboratively perform tasks, so as to realize the collaborative work tasks between the inspection robots; at the same time, the present invention comprehensively considers various factors such as capability, time, task adaptability and inspection robot cost, and effectively allocates tasks to the inspection robots; secondly, the exchange contract is introduced to improve the sales contract mechanism, and flexible task scheduling and resource allocation are realized by setting a minimum capability satisfaction threshold to ensure the smooth completion of the task; in addition, according to the different capabilities of the inspection robots and the differences in the requirements of the task types, the method of capability matching bids is adopted to accurately allocate tasks to suitable robots, thereby improving the efficiency and quality of task completion.
[0125] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. Unmanned collaborative task allocation method for power plant protection equipment, It is characterized in that The unmanned collaborative task allocation method applied to power plant protection equipment comprises the following steps: S1. Based on objective evaluation indicators and subjective evaluation indicators, comprehensively evaluate the utility of each inspection robot and generate a bid, wherein the objective evaluation indicators include time factors and cost factors, and the subjective evaluation indicators include task adaptability and capability satisfaction; S2. According to the current task, the inspection robots are divided into tenderers and bidders, and the purchasing power of the bidders is judged by the contract rules and the bids are submitted; S3. Based on the minimum capability satisfaction threshold, evaluate the capability satisfaction of the bidders who submitted the tender documents, and dispatch additional inspection robots to form a collaborative task group to deliver the tender documents of the collaborative task group; S4. Compare the utility of the submitted bids and select the winning bidder, use the winning bidder to sign a contract with the tenderer, and perform the assigned tasks.
2. The unmanned collaborative task allocation method for power plant protection equipment according to claim 1, It is characterized in that The method of comprehensively evaluating the utility of each inspection robot based on objective evaluation indicators and subjective evaluation indicators and generating a bid includes the following steps: S11. Calculate objective evaluation indicators of the inspection robot based on time factors and cost factors; S12. Calculate the subjective evaluation index of the inspection robot based on task suitability and capability satisfaction; S13. Use the utility formula to integrate the objective evaluation index and the subjective evaluation index to obtain the bid document.
3. The unmanned collaborative task allocation method for power plant protection equipment according to claim 2, It is characterized in that The expressions for calculating the objective evaluation index of the inspection robot based on the time factor and the cost factor are: Where Time i represents the time factor for executing task i, WT i Indicates the waiting time for executing task i, NT iw represents the time required for the inspection robot w to perform task i, and n represents the number of inspection robots participating in the task; Where, Value i represents the cost of the inspection robot participating in task i, Price w represents the price of inspection robot w, and n represents the number of inspection robots participating in task i.
4. The unmanned collaborative task allocation method for power plant protection equipment according to claim 3, It is characterized in that The expressions for calculating the subjective evaluation index of the inspection robot based on task adaptability and capability satisfaction are: In the formula, Adaptation i represents the suitability of the task, TComplexity represents the average autonomy of the inspection robots participating in the i-th task. i Indicates the complexity of the task; In the formula, CS i Indicates ability satisfaction, equal to task T i The average value of the sum of the various ability satisfactions; j Represents task T i The requirement for the jth capability of the inspection robot to perform the task, Indicates the execution of task T i The size of the j-th capability of the inspection robot, m i Represents task T i The number of capabilities for which the capability requirement is not 0.
5. The unmanned collaborative task allocation method for power plant protection equipment according to claim 4, It is characterized in that The expression for integrating the objective evaluation index and the subjective evaluation index using the utility formula is: U i =ω i1 CS i +oh i2 Adaptation i +oh i3 Time i -oh i4 Value i ; In the formula, CS i Indicates ability satisfaction, Adaptation i Indicates task suitability, Time i Indicates the time factor for executing task i, Value i represents the cost of the inspection robot participating in task i, w i1 、w i2 、w i3 and i4 are weight parameters, which respectively represent the importance of task capability satisfaction, task adaptation, time factor and cost factor.
6. The unmanned collaborative task allocation method for power plant protection equipment according to claim 1, It is characterized in that The method of dividing the inspection robots into tenderers and bidders according to the current tasks, judging the purchasing power of the bidders by using the contract rules and submitting the bids includes the following steps: S21. Divide the inspection robot into tenderers and bidders according to current task requirements, and use the tenderer to publish task information to the bidders; S22, based on the received task information, using the sales contract rules to determine the purchasing power of each bidder, and sending the bidding information to the tenderer according to the determination result; S23. Use the exchange contract rules to determine the purchasing power of each bidder, and send bidding information to the bidder based on the determination result.
7. The unmanned collaborative task allocation method for power plant protection equipment according to claim 6, It is characterized in that The method of judging the purchasing power of each bidder based on the received task information by using the sales contract rules and sending the bidding information to the tenderer according to the judgment result includes the following steps: S221. Classify the purchasing power level of each bidder based on the task set of the inspection robot itself; S222, determining the task requirement level based on the received task information; S223. Traverse the purchasing power level of each bidder. If the purchasing power level of the bidder is greater than or equal to the task requirement level, send the bid to the tenderer. If the purchasing power level of the bidder is less than the task requirement level, execute step S23.
8. The unmanned collaborative task allocation method for power plant protection equipment according to claim 7, It is characterized in that The method of using the exchange contract rules to judge the purchasing power of each bidder and sending the bidding information to the bidder according to the judgment result includes the following steps: S231. The bidder pre-exchanges the tasks in the task set of its own equipment with the bidding tasks in sequence; S232, judging whether the bidders after the pre-exchange meet both the distance constraint and the time constraint, and if not, giving up submitting the bid; S233, calculating the utility of the two inspection robots after the pre-exchange, regenerating the bid of the bidder and sending it to the tenderer, wherein the expression for calculating the utility of the two inspection robots after the pre-exchange is: In the formula, It represents the utility of the two inspection robots after the pre-exchange, T i Indicates the bidding task, represents the pth task in the task set of bidder j’s own equipment, U j represents the utility of bidder j performing the original task set, U z represents the utility of bidder z performing the original task set, Indicates that bidder j gives up the task in his task set Execute bidding tasks i The utility of Indicates that the bidder z performs task T i The latter effect.
9. The unmanned collaborative task allocation method for power plant protection equipment according to claim 1, It is characterized in that Based on the minimum capability satisfaction threshold, the capability satisfaction of the bidders who submitted the bid is evaluated, and additional inspection robots are dispatched to form a collaborative task group. The bid submission to the collaborative task group includes the following steps: S31, traversing the bidders who have submitted bids, and comparing the capability satisfaction with the minimum capability satisfaction threshold; S32. Based on the comparison result, if the bidder's capability satisfaction is less than the preset minimum capability satisfaction threshold, an additional inspection robot is dispatched to form a collaborative bidding group, and the utility of each collaborative bidding group is calculated; S33. Compare the calculation results, mark the collaborative bidding group with the largest utility as the successful bidder, and release the bid failure information to the unsuccessful bidders.
10. An unmanned collaborative task allocation system for power plant protection equipment, used to implement the unmanned collaborative task allocation method for power plant protection equipment according to any one of claims 1 to 9, It is characterized in that The unmanned collaborative task allocation system for power plant protection equipment includes: A comprehensive evaluation module, which is used to comprehensively evaluate the utility of each inspection robot based on objective evaluation indicators and subjective evaluation indicators, and generate a bid, wherein the objective evaluation indicators include time factors and cost factors, and the subjective evaluation indicators include task adaptability and capability satisfaction; The sales contract module is used to divide the inspection robots into tenderers and bidders according to the current tasks, and use the contract rules to judge the purchasing power of the bidders and submit the bids; The collaborative task module is used to evaluate the capability satisfaction of the bidders who submitted the tender documents based on the minimum capability satisfaction threshold, and to dispatch additional inspection robots to form a collaborative task group to deliver the tender documents of the collaborative task group; The bid matching module is used to compare the utility of the submitted bids and select the winning bidder, use the winning bidder to sign a contract with the tenderer, and execute the assigned tasks.
Citation Information
Cited By
Robot cluster management system and method
CN119310952A
Dynamic public toilet inspection method, device and equipment based on visitor flow rate and problem reporting
CN120598255A
Dynamic public toilet inspection method and device based on passenger flow and problem reporting, and equipment
CN120598255B
Low-electric-quantity inspection task processing system for inspection robot
CN120802803A
