CN120802803A - Low-power inspection task processing system of inspection robot - Google Patents
Low-power inspection task processing system of inspection robotInfo
- Publication number
- CN120802803A CN120802803A CN202511269360.6A CN202511269360A CN120802803A CN 120802803 A CN120802803 A CN 120802803A CN 202511269360 A CN202511269360 A CN 202511269360A CN 120802803 A CN120802803 A CN 120802803A
- Authority
- CN
- China
- Prior art keywords
- task
- inspection
- robot
- low
- power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a low-power inspection task processing system of an inspection robot, which comprises a central cooperative processing server and at least two inspection robots, wherein the central cooperative processing server and the at least two inspection robots are in interactive connection, the inspection robots are provided with a vehicle-mounted control unit, the vehicle-mounted control unit monitors the residual power of a battery in real time, a first node is preset, a low-power early warning signal is triggered when the power is reduced to the node for the first time, the central cooperative processing server comprises a state aggregation sensing module, a dynamic energy consumption prediction module and a task reconstruction decision module, the former acquires real-time state data of the aggregation robots, the dynamic energy consumption prediction module predicts task energy consumption, and the task reconstruction decision module selects to execute a task replacement protocol or a task exchange protocol according to the real-time state data and the predicted energy consumption after receiving early warning so as to ensure continuity of the inspection task, data integrity and robustness of system operation.
Description
Technical Field
The invention relates to the technical field of train inspection, in particular to a low-power inspection task processing system of an inspection robot.
Background
Along with the increasing demand of railway train inspection, the application of inspection robots in automatic detection and intelligent management gradually becomes a research hot spot. However, in the actual inspection task, the inspection robot cannot complete the whole inspection task due to insufficient electric quantity, and especially in the case of complex inspection environments or prolonged time consumption of data acquisition, the problem is particularly remarkable. Therefore, how to effectively process tasks of the inspection robot in a low-power state becomes a technical problem to be solved.
Through retrieval, a train inspection robot control system with the publication number of CN116476099B is disclosed, and the high-efficiency detection of parts at the bottom of a train is realized by means of real-time acquisition of road surface information, generation of positioning data, recognition of obstacles and the like. However, the technical scheme does not relate to a task allocation and processing mechanism of the inspection robot in a low-power state, which may cause the robot to interrupt inspection due to insufficient power when executing complex or time-consuming tasks, and affect the completion efficiency of the overall task. In addition, the scheme lacks a dynamic adjustment function for the priority of the inspection task, and the inspection path or task decomposition cannot be reasonably planned according to the electric quantity state, so that the risk of task failure can be increased.
Through the search, a transfer device and a transfer method of a train inspection robot are disclosed, wherein the publication number of the transfer device is CN116986232B, and the flexible transfer device is designed, so that the inspection robot can be quickly transferred to different positions, and the restrictions of basic construction transformation and fixed transfer positions are eliminated. However, the technical scheme mainly focuses on the problem of transferring the inspection robot before or after the task starts, and fails to fully consider the situation that the robot needs emergency treatment due to insufficient electric quantity in the task execution process. For example, when the power of the robot is insufficient during inspection, the solution cannot provide an effective emergency measure or task delivery mechanism, which may cause the inspection task to be forcefully interrupted or delayed. In addition, the scheme does not mention a task reassignment strategy when multiple robots work cooperatively, and is difficult to deal with the problem of overall task coordination when the electric quantity of a single robot is insufficient.
The above problems indicate that the existing inspection robot technology still has a certain defect in terms of task processing capability under a low-power state, and especially lacks a systematic solution in terms of task dynamic adjustment, path optimization, collaborative operation, emergency transportation and the like.
Disclosure of Invention
Aiming at the defects existing in the prior art, the invention aims to overcome the defects of task interruption, data loss and low multi-robot cooperation efficiency caused by contradiction between static task planning and dynamic energy consumption in the conventional inspection robot task execution system under the low-power working condition.
Therefore, the invention provides a system for processing the low-power inspection task of the inspection robot, which realizes the accurate quantification, prediction and management of the energy state of a robot cluster by constructing a closed-loop control system taking a prediction energy model as a core, and establishes a set of multi-threshold triggered collaborative processing protocol comprising the dynamic reconstruction of the task, intelligent succession and peer-to-peer exchange on the basis of the energy state, thereby maximally ensuring the continuity of the inspection task, the data integrity and the robustness of the whole operation of the system on the premise of ensuring the safe return of the single robot.
In order to achieve the above purpose, the present invention provides the following technical solutions:
the system is characterized in that the vehicle-mounted control unit monitors the residual electric quantity of a battery carried by the vehicle-mounted control unit in real time, and a first node is preset, and when the residual electric quantity is monitored to drop to the first node for the first time, a low-electric quantity early warning signal is triggered;
the central co-processing server comprises:
the state aggregation sensing module is used for acquiring and aggregating real-time state data reported by all the inspection robots, wherein the real-time state data at least comprises current position coordinates and residual electric quantity of each robot;
The dynamic energy consumption prediction module predicts the real-time state information and the task to be executed based on a preset dynamic energy consumption prediction model to obtain the energy consumption required by executing the task;
and the task reconstruction decision module is used for selecting and executing a cooperative processing protocol according to the real-time state data and the predicted energy consumption after acquiring the low-power early warning signal, wherein the cooperative processing protocol comprises the following steps:
The task taking over protocol screens out an optimal taking over robot from other inspection robots in a non-low-power state, and the incomplete task of the inspection robot triggering the low-power early-warning signal is allocated to the taking over robot in an instructive manner for execution;
and the task exchange protocol is to complete the peer-to-peer exchange of the remaining task packages of the inspection robot triggering the low-power early warning signal and the inspection robot in the other non-low-power state under the condition of meeting the preset energy exchange feasibility and the global time efficiency verification.
Further, the task taking over protocol includes a selection policy, where the selection policy includes defining that the inspection robot triggering the low-power early warning signal is a first robot, starting an optimal taking over robot screening program after receiving a task packet to be handed over sent by the first robot and including the task packet to be handed over that cannot be completed, traversing all other candidate robots in a non-low-power state currently by the screening program, performing quantization evaluation for each candidate robot based on a multi-dimensional task affinity scoring model, and finally selecting a candidate robot with the highest task affinity scoring as the optimal taking over robot.
Further, the dimension factors in the multi-dimension task affinity scoring model comprise a space adjacency factor, an energy source abundance factor, a path synergy index factor, a historical task quality scoring factor and a current task load factor, wherein the space adjacency factor reflects the reciprocal of Euclidean distance from the current position of the candidate robot to a first task point in a task package to be handed over, the energy source abundance factor reflects the difference between the current residual electric quantity of the candidate robot and the estimated total energy required for completing the task package to be handed over, the path synergy index factor reflects the increment amplitude of the total path length of the task package to be handed over after being inserted into the current task sequence of the candidate robot, the historical task quality scoring factor reflects the comprehensive index of the data quality of task points completed in the past of the candidate robot, and the current task load factor reflects the current residual task quantity or the estimated residual working time of the candidate robot.
Further, the vehicle-mounted control unit further comprises a step of calling a preset dynamic energy consumption prediction model, performing accessibility assessment on all task points in a residual task list one by one, wherein the accessibility reflects total estimated energy consumption required for sequentially completing the residual task points from a current position, comparing the total estimated energy consumption with a preset available operation energy, the available operation energy reflects an electric quantity difference value between a first node and a preset second node which is lower than the first node, deducting a return safety margin dynamically calculated according to the distance from the current position to a charging station, and if the assessment result is that the available operation energy meets at least one task point in the residual task list, determining a maximum number of task point sets which can be completed before the electric quantity is exhausted to the second node as a final execution sequence, packaging the residual task points which cannot be completed into a task package to be handed over, and sending the task package to a central cooperative processing server.
Further, the task exchange protocol includes a preset check condition, where the preset check condition includes:
The energy consumption interchange feasibility check is carried out, and the current residual capacity of the inspection robot triggering the low-electric quantity early warning signal is larger than the estimated energy consumption required by completing the residual task package of the inspection robot in another non-low-electric-quantity state;
The overall time efficiency is checked, and the sum of time consumption of the inspection robot triggering the low-power early warning signal and the inspection robot in the other non-low-power state from the current position to the task starting point of the other party is smaller than or equal to a preset time window threshold;
And the compatibility verification of the sensor capability triggers the sensor configuration carried by each of the inspection robot with the low-power early warning signal and the inspection robot with the other non-low-power state, so that the data acquisition of all task points in a task package of the other party can be satisfied.
Further, when it is determined that there are switching objects satisfying all the verification conditions, a task switching instruction is simultaneously sent to the inspection robot and the switching object triggering the low-power early warning signal in the central co-processing server, the task switching instruction includes the remaining task packet information of the other party and the optimal path planning for reaching the task starting point, and if there is a difference between the types or calibration parameters of the inspection robot and the switching object triggering the low-power early warning signal, the task switching instruction further includes a data acquisition parameter conversion matrix.
Furthermore, the central co-processing server also comprises an initial task allocation module,
The initial task distribution module is used for clustering three-dimensional coordinates of multiple task points to be executed based on a spatial clustering algorithm of density to form a plurality of task point clusters which are adjacent in space, calculating the total amount of tasks for each task point cluster, wherein the total amount of tasks reflects the sum of the estimated moving time of the shortest traversal path between the estimated value of the reference time consumption of all the task points in the cluster and the estimated moving time of the shortest traversal path between the task points in the cluster, and distributing the task point clusters to different inspection robots according to an optimal distribution algorithm so as to minimize the estimated residual electric quantity variance of all the robots after completing respective tasks, thereby forming respective initial task lists.
Further, the central co-processing server also comprises a data fusion and report generation module,
And the data fusion and report generation module accurately correlates all data acquired by different inspection robots at different times and through different collaborative processing protocols according to the global unique task point identifier attached to each data file, and generates a comprehensive inspection report.
Further, the input quantity of the dynamic energy consumption prediction model comprises inherent physical parameters, real-time kinematic parameters, real-time load parameters, real-time power consumption of the sensor system and real-time calculation load of the vehicle-mounted calculation unit of the inspection robot.
Further, the dynamic energy consumption prediction module calculates the instantaneous total power consumption of the robot in real time through weighted summation according to driving power, sensor system power and calculation unit power, wherein the driving power reflects a function of real-time kinematic parameters and real-time load parameters, the sensor system power is determined according to an activated sensor list and working states of the activated sensor list, and the calculation unit power reflects the real-time load rate of the vehicle-mounted calculation unit.
The invention has the beneficial effects that 1, the electric quantity is monitored in real time through the vehicle-mounted control unit, the low-electric quantity early warning node is preset, after the early warning is triggered, the task reconstruction decision module of the central cooperative processing server reasonably distributes the task which is not completed by the low-electric quantity robot to other robots through the task taking over or task exchanging protocol, the problems of task interruption and data deletion caused by insufficient electric quantity of the inspection robot in the prior art are effectively solved, the continuity of the inspection task is ensured, in addition, the carried dynamic energy consumption prediction model comprehensively considers multiple factors such as inherent physical parameters, real-time kinematic parameters, load parameters, sensor power consumption, calculation load and the like of the robot, calculates the instantaneous total power consumption in real time through weighted summation, combines with the real-time data continuous optimization model, greatly improves the accuracy of energy consumption prediction, and provides reliable quantification basis for task taking over, exchanging and accessibility assessment;
2. The task exchange protocol in the invention covers space adjacency, energy source adequacy, path cooperative property, historical task quality and current load through a multidimensional task affinity scoring model, screens optimal successor, ensures efficient execution of the successor task, reduces resource waste, realizes peer-to-peer exchange of tasks when meeting conditions through triple verification of energy consumption feasibility, time efficiency and sensor compatibility, fully utilizes the residual electric quantity and capacity of each robot, improves the overall inspection efficiency, and in addition, a reachability evaluation mechanism of a vehicle-mounted control unit determines the maximum completable task set by comparing available operation energy with residual task energy consumption when in low electric quantity, reserves return safety margin, ensures that the robot can safely return while maximally completing the task, and realizes balance of efficiency and safety.
Drawings
FIG. 1 is a general flow chart of the present invention;
FIG. 2 is a flow chart of task taking over protocol execution in the present invention;
FIG. 3 is a flow chart of the task exchange protocol execution in the present invention.
Detailed Description
The invention will now be described in further detail with reference to the drawings and examples. Wherein like parts are designated by like reference numerals. It should be noted that the words "front", "back", "left", "right", "upper" and "lower" used in the following description refer to directions in the drawings, and the words "bottom" and "top", "inner" and "outer" refer to directions toward or away from, respectively, the geometric center of a particular component.
Because the existing inspection robot technology still has a certain defect in the task processing capability under the low-power state, and particularly lacks a systematic solution in the aspects of task dynamic adjustment, path optimization, collaborative operation, emergency transfer and the like, the invention designs the inspection robot low-power inspection task processing system, which comprises a central collaborative processing server and at least two inspection robots, as shown in fig. 1, wherein the central collaborative processing server is in interactive connection with the inspection robots, and the inspection robots are provided with vehicle-mounted control units.
The central cooperative processing server is internally integrated with a task definition and decomposition module, a system state initialization module, a dynamic energy consumption prediction module, an initial task allocation module, a state aggregation sensing module, a task reconstruction decision module and a data fusion and report generation module, wherein the task definition and decomposition module is used for receiving and analyzing an externally input train maintenance instruction, the instruction comprises a unique identifier of a train to be detected, vehicle type data, accurate three-dimensional space parking pose information of the train to be detected in a maintenance track and a structured key detection point list, and each point in the list is analyzed and converted into a task point by the module. Each task point is assigned a globally unique identifier and contains a set of parameters necessary to execute the task unit, in particular the set of parameters comprising the three-dimensional spatial coordinates of the target pointAnd sensor modalities such as high resolution visual image acquisition, three-dimensional laser point cloud scanning or ultrasonic thickness measurement, and sensor parameter configuration, a particular task point data structure specification may be defined as ：{AIU_ID:1001,TargetCoord:{x:45.72m,y:1.25m,z:0.88m},SensorModality:'HighResVisual',SensorParams:{ExposureTime:1500μs,Gain:18dB,Focus:'auto'}, EstimatedDuration: 2.8s, Priority: 1}.
The system state initialization module is responsible for broadcasting state inquiry instructions to all inspection robots in a standby state at the beginning of task starting, and a vehicle-mounted control unit of each inspection robot immediately returns an initial state vector of the inspection robot after receiving the instructions, wherein the initial state vector comprises current accurate position coordinates of the robot, current residual electric quantity percentages and hardware health state codes of the robot, and the central server collects the initial state vectors of all robots and completes initial counting of system resources.
The dynamic energy consumption prediction module establishes and maintains an independent and continuously iterated dynamic energy consumption prediction model at a server side for each inspection robot, wherein the model is not a static table look-up or linear extrapolation model, but a composite function considering multiple physical and calculation factors, the input variables of the model comprise inherent physical parameters of the robot, such as the whole machine quality, a rated power and efficiency curve of a driving motor, a wheel train structure and a transmission ratio, real-time kinematic parameters of the robot, such as the current linear speed, the angular speed and the acceleration, a load current of the driving motor fed back in real time by a vehicle control unit, the current value is used for reversely calculating the rolling friction resistance coefficient of the robot on the current road surface, the real-time power consumption of a sensor system is dynamically calculated according to the type of the currently started sensor and the working mode thereof, and the real-time load rate of the vehicle control unit is used for quantifying the calculation energy consumption generated by calculation tasks such as path planning, obstacle avoidance and the like, and the model calculates the instantaneous total power consumption of the robot through a weighted summation real-time formulaThe method specifically comprises the following steps: Wherein For driving power, it is a complex function of speed, acceleration and motor current,The power of the sensor is obtained by looking up a table according to the activated sensor list and the working state thereof,To calculate the power consumption, the load rates of the CPU and GPU are linearly related,、、For balancing the contribution of energy consumption of each part as weight coefficientIs an integral calibration coefficient, is calibrated by carrying out complete discharge test on the robot under standard working conditions, has the value of usually 1.0 to 1.1 so as to compensate unmodeled factors such as battery aging, internal resistance change and the like, and is realized by carrying out instantaneous total powerThe system can predict the energy consumption required by executing any task segment with extremely high precision by carrying out time integration along a planned path or in a task point execution period, and the model can utilize real electric quantity consumption data reported by a robot in real time in the task execution process, and continuously model internal parameters, particularly rolling friction resistance coefficient and calibration coefficient, through a Kalman filtering algorithmAnd carrying out online updating and optimization, thereby realizing the self-adaptive iteration of the model.
The initial task allocation module performs first task division based on the spatial distribution characteristics of all task points and the initial state of each robot, firstly, the module calls a spatial clustering algorithm based on density, such as a DBSCAN algorithm, clusters three-dimensional coordinates of all task points to form a plurality of spatially adjacent task point clusters, then, the system calculates the total amount of tasks of each cluster, which is the sum of a reference time consumption estimated value of all task points in the cluster and the estimated movement time consumption of the shortest traversal path between the task points in the cluster, and finally, allocates the task point clusters to different robots through an allocation strategy, such as optimized allocation of a genetic algorithm, aiming at optimizing the distribution of the global energy efficiency, so as to form respective initial task lists, wherein the estimated residual electric quantity of all robots after completing respective tasks is as balanced as possible.
The state aggregation sensing module continuously receives real-time state update data packets from each robot vehicle-mounted control unit during task execution, the data packets are sent at fixed high frequency, the content of the data packets forms a complete snapshot of the current state of the robots, the complete state aggregation sensing module comprises high-precision real-time position coordinates, instantaneous speed, accurate residual electric quantity, currently executed task point identifiers, completed task point lists, residual task point lists, driving motor real-time current and loads of vehicle-mounted computing units, a central collaboration server gathers, analyzes and aligns data streams from different robots with time stamps, a global collaborative situation map is constructed and maintained in a shared memory database, the map visually presents position tracks, electric quantity decline curves and task progress of all robots, continuous and real input data is provided for a subsequent dynamic energy consumption prediction model, and a multi-threshold task reconstruction decision making module is provided.
The vehicle-mounted control unit of the inspection robot is used for executing instructions from the central server and feeding back the state of the inspection robot in real time, and a module is configured in the vehicle-mounted control unit to read real-time voltage, current and residual capacity data of the battery. The vehicle-mounted control unit is also provided with a local power threshold value, namely a first node and a second node, wherein the first node is set to be 20 percent, and the second node is set to be 10 percent.
When a vehicle-mounted control unit of a patrol robot monitors that the self electric quantity of the patrol robot is reduced for the first time and touches a first node, the task reconstruction decision module is activated, one of the following two logic branches is selected and executed according to the current state and the global cooperative situation of the robot, and the patrol robot triggering the low-electric quantity early warning signal is defined as the first robot.
The method comprises the steps of calculating total estimated energy consumption required by sequentially completing a residual task point sequence from a current position, comparing the total estimated energy consumption with operation energy consumption, comparing the total estimated energy consumption with current available residual energy, wherein the current available residual energy is an electric quantity difference value between the first node and the second node, deducting reserved return safety margin, starting the protocol if an evaluation result shows that the available residual energy of the first robot is enough to complete at least one task point in a residual task list, and firstly determining a maximum number of task point subsets which can be completed before the electric quantity is exhausted to the second receiving list by the vehicle-mounted control unit of the first robot, marking the subsets as final execution sequences, simultaneously, packaging the residual tasks and the tasks which cannot be completed into a wireless communication packet, and continuing to execute the task packets to a wireless communication packet by the central processing device, and continuing to execute the task packets in a wireless communication packet.
Meanwhile, after receiving the task package to be handed over, the central server immediately starts an optimal successor robot screening program, the program traverses all other robots currently executing tasks, except the first robot and other robots in a return or low-power state, and carries out quantitative evaluation on each candidate robot, which is defined as a second robot, and the evaluation algorithm is based on a multi-dimensional task affinity scoring model. The calculation formula of the model is that. Wherein, the The Euclidean distance from the current position of the second robot to the first task point in the task package to be handed over shows the space adjacency; As the current remaining power of the second robot, To complete the estimated total energy consumption required by the task package, the difference between the two represents an energy adequacy; the path synergy index quantifies the increase amplitude of the total path length of the task to be handed over after the task to be handed over is inserted into the existing task sequence of the second robot, and the smaller the increase amplitude is, the higher the synergy is; is a historical task quality score of the second robot, and is a comprehensive index based on the quality (such as image definition and point cloud integrity) of task points completed in the past; Is the current remaining task load of the second robot; 、、、 And The server calculates the scores of all candidate robots and selects the robot with the highest score as the final successor, then the server sends the task package to be handed over to the successor robot in an instruction mode, the latter integrates the task package to be handed over into a task queue of the server, a new optimal execution path is planned, and the successor task is executed after the original task of the server is completed or at an appropriate node of path optimization.
In the process that the first robot executes the final execution sequence, if the electric quantity of the first robot further drops and touches the second node, the vehicle-mounted control unit immediately stops all current routing inspection operations unconditionally and starts an autonomous return program, the return path is provided with a local path planning module by adopting an A algorithm, a shortest and collision-free path for returning the charging pile is planned on an environment map which is preloaded and contains permanent obstacles and dynamic obstacle information issued by a server, a coordinate point sequence of the return path is uploaded to a central server in real time, after the server receives the return path, the server immediately marks the return path as a reserved space-time channel with high priority in a global coordination situation map, and gives an early warning and avoiding instruction to other robots possibly conflicting along the path, absolute safety of the return process is ensured, and if two or more candidate robots all meet the conditions when screening the candidate robots, the system can preferably select the robot with higher historical task quality score and better path index in coordination except taking the consideration of the score.
As shown in fig. 3, the task exchange protocol is triggered by the fact that when the first robot touches the first node, the local reachability evaluation result shows that the available residual energy is insufficient to complete any task point in the residual task list, and in this case, the first robot broadcasts a task exchange request to the central server, and the request includes the complete state vector of the first robot and the incomplete residual task package a.
After receiving the request, the central server starts a peer exchange matching program, which traverses all other robots with non-low electric quantity as potential exchange objects, such as a second robot, and performs strict feasibility check on each pair of possible exchange combinations, wherein the check comprises three necessary conditions, namely, first energy exchange feasibility check, that is, the current residual electric quantity of the first robot must be greater than estimated energy consumption required for completing a residual task package B of the second robot, and the current residual electric quantity of the second robot must be greater than estimated energy consumption required for completing a residual task package A of the first robot, second global time efficiency check, that is, the sum of the time consumption of the first robot moving from the current position to the task starting point of the second robot and the time consumption of the second robot moving from the current position to the task starting point of the first robot must be less than or equal to a preset threshold, and the setting of the threshold is aimed at ensuring that the task exchange does not cause significant extension of the whole project, and third sensor capacity compatibility check, that is, that the sensor configuration data of the first robot and the second robot must meet the requirement of acquiring all task data of the task configuration packages of the second robot.
This exchange is judged to be possible only when a pair of robots simultaneously satisfies all three conditions described above. If there are multiple feasible exchange partners, the server will select the combination which can minimize the total path increment of the two robots after exchange, once the exchange object is determined, the central server will send out task exchange instructions to the first robot and the exchange object at the same time, the instructions for the first robot include all information of the remaining task packages of the exchange object and the optimal path planning reaching the starting point of the task packages, and similarly, the instructions for the exchange object include the remaining task packages A of the first robot, and in addition, if there is a difference between the sensor model or calibration parameters of the first robot and the exchange object, the instructions also include a data acquisition parameter conversion matrix which is automatically generated by the server according to the pre-stored sensor specifications to instruct the robots to adjust the sensor settings when executing the exchange task, thereby ensuring consistency and comparability of the acquired data, and the first robot and the exchange object immediately discard the original remaining tasks respectively after receiving the instructions, and execute the new task after the exchange.
The data fusion and report generation module is responsible for summarizing all the uploaded task point data after all the inspection tasks are completed (including an initial allocation task, a successive task and an exchange task), accurately correlates data acquired from different robots at different times to initial train key detection points through global unique task point identifiers attached to each data file, automatically performs quality check on the data by a system, for example, evaluates the definition and contrast of an image by using an image processing algorithm, checks the density and integrity of a point cloud analysis algorithm check point cloud, finally, generates a complete and seamless spliced comprehensive inspection report, wherein in the report, each piece of detection data is attached with detailed data tracing information, and clearly marks what robot is used for completing acquisition in what cooperative mode, so that a complete evidence chain is provided for subsequent quality tracing and operation and maintenance decision.
Examples:
A simulated train overhaul warehouse with the length of 150 meters and the width of 20 meters is provided, wherein a train of 8-section grouped motor train unit models are parked, and three inspection robots with the specifications consistent with the specifications are deployed in total, wherein the inspection robots are R01, R02 and R03 respectively. In total, 120 considered points are defined and distributed at the two sides and the bottom of the train, the initial electric quantity of the robot is 100% (40 Ah), the electric quantity threshold is set to be that a first node=20% (8 Ah), a second node=10% (4 Ah), and the return safety margin is set to be 2Ah.
After the initial task allocation module operates, similar task amounts are allocated to three robots, wherein R01 allocates 42 task points, R02 allocates 38 task points and R03 allocates 40 task points.
After the task starts to be executed, the three robots work according to the plan, when the task is carried out for about 1 hour and 48 minutes, the vehicle-mounted control unit of R01 monitors that the residual electric quantity of the robot first touches the first node, at this time, the R01 has completed 32 task points, and 10 task points remain to be executed.
The local energy consumption model of R01 immediately starts up the reachability evaluation, and the evaluation result shows that, starting from the current position, the first 3 (AIU-R01-33, AIU-R01-34, AIU-R01-35) of the remaining 10 task points are completed, which is expected to consume 1.8Ah of electric quantity, and its available operation energy is 8Ah-4Ah-2Ah (safety margin) =2ah, so that R01 determines its final execution sequence as the 3 task points, and packages the remaining 7 task points (AIU-R01-36 to AIU-R01-42) into a task packet to be handed over to the central server.
After receiving the task package, the central server immediately evaluates R02 and R03, and the real-time state at that time is as follows:
R02, the residual electric quantity 22Ah (55%), 8 residual tasks, 15m from the current position to the starting point of the handover task, and 92 scores of historical task quality.
And R03, the residual electric quantity is 25Ah (62.5%), the number of the residual tasks is 11, the current position is 48 meters away from the starting point of the handover task, and the quality score of the historical task is 88 minutes.
The server calculates the energy consumption required to complete 7 task points in the handover packet to be 6.5Ah.
Score calculation (weight)=0.35,=0.30,=0.15,=0.10,=0.10):
TAS_R02=0.35*(1/(1+15))+0.30*(22-6.5)+0.15*(0.92)+0.10*(92/100)-0.108=0.022+4.65+0.138+0.092-0.8=4.102;
TAS_R03=0.35(1/(1+48))+0.30*(25-6.5)+0.15*(0.85)+0.10*(88/100)-0.10*11=0.007+5.55+0.128+0.088-1.1=4.673;
The calculation results show that, although R02 is spatially closer, R03 has more abundant energy and lower current task load, which scores higher, so the server selects R03 as the successor and issues the task package to R03.
R01 continues to execute and completes the final 3 task points, then safely starts the return program when the electric quantity is reduced to the second node, R03 seamlessly links 7 task points from R01 after completing the task of the return program, and R02 completes all the tasks of the return program according to the original plan.
Finally, all 120 task points are successfully executed, the data acquisition is complete, all three robots safely return to the charging pile after the task is completed, and the whole inspection task takes 2 hours and 55 minutes.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above examples, and all technical solutions belonging to the concept of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to the present invention may occur to one skilled in the art without departing from the principles of the present invention and are intended to be within the scope of the present invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202511269360.6A CN120802803B (en) | 2025-09-08 | 2025-09-08 | Low-power inspection task processing system of inspection robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202511269360.6A CN120802803B (en) | 2025-09-08 | 2025-09-08 | Low-power inspection task processing system of inspection robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN120802803A true CN120802803A (en) | 2025-10-17 |
| CN120802803B CN120802803B (en) | 2025-11-21 |
Family
ID=97326010
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202511269360.6A Active CN120802803B (en) | 2025-09-08 | 2025-09-08 | Low-power inspection task processing system of inspection robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120802803B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170361456A1 (en) * | 2014-11-19 | 2017-12-21 | Positec Technology (China) Co., Ltd | Self-moving robot |
| CN110139286A (en) * | 2019-05-21 | 2019-08-16 | 西安邮电大学 | Wireless sensor network coverage enhancement method and system towards three-dimensional environment |
| CN113635302A (en) * | 2021-07-29 | 2021-11-12 | 深圳墨影科技有限公司 | Integrated mobile cooperative robot control system based on field bus |
| US20220164747A1 (en) * | 2020-11-20 | 2022-05-26 | Lyft, Inc. | Operations task creation, prioritization, and assignment |
| CN118741573A (en) * | 2024-07-26 | 2024-10-01 | 广州航海学院 | A fault-tolerant method to improve the robustness of underwater robot networking |
| CN118886652A (en) * | 2024-07-08 | 2024-11-01 | 邓州市斌迅新能源科技有限公司 | A method and system for optimizing control of electric power |
| CN119026826A (en) * | 2024-07-09 | 2024-11-26 | 国电南京自动化股份有限公司 | Unmanned collaborative task allocation method and system for power plant protection equipment |
| CN119536317A (en) * | 2025-01-17 | 2025-02-28 | 北京数字绿土科技股份有限公司 | A method and system for scheduling fixed-point patrol tasks of drone clusters in multi-task mode |
-
2025
- 2025-09-08 CN CN202511269360.6A patent/CN120802803B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170361456A1 (en) * | 2014-11-19 | 2017-12-21 | Positec Technology (China) Co., Ltd | Self-moving robot |
| CN110139286A (en) * | 2019-05-21 | 2019-08-16 | 西安邮电大学 | Wireless sensor network coverage enhancement method and system towards three-dimensional environment |
| US20220164747A1 (en) * | 2020-11-20 | 2022-05-26 | Lyft, Inc. | Operations task creation, prioritization, and assignment |
| CN113635302A (en) * | 2021-07-29 | 2021-11-12 | 深圳墨影科技有限公司 | Integrated mobile cooperative robot control system based on field bus |
| CN118886652A (en) * | 2024-07-08 | 2024-11-01 | 邓州市斌迅新能源科技有限公司 | A method and system for optimizing control of electric power |
| CN119026826A (en) * | 2024-07-09 | 2024-11-26 | 国电南京自动化股份有限公司 | Unmanned collaborative task allocation method and system for power plant protection equipment |
| CN118741573A (en) * | 2024-07-26 | 2024-10-01 | 广州航海学院 | A fault-tolerant method to improve the robustness of underwater robot networking |
| CN119536317A (en) * | 2025-01-17 | 2025-02-28 | 北京数字绿土科技股份有限公司 | A method and system for scheduling fixed-point patrol tasks of drone clusters in multi-task mode |
Also Published As
| Publication number | Publication date |
|---|---|
| CN120802803B (en) | 2025-11-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116611635B (en) | Sanitation robot vehicle dispatching method and system based on vehicle-road collaboration and reinforcement learning | |
| CN110210806B (en) | Cloud-based unmanned vehicle frame structure with 5G edge calculation function and control evaluation method thereof | |
| CN118376251B (en) | A route collaborative control method and system for multiple inspection robots | |
| CN116759355B (en) | Wafer transmission control method and system | |
| CN115115165B (en) | A system having at least one facility system | |
| CN119443427A (en) | Automatic dispatching method for road obstacle removal and rescue enterprises | |
| CN119476670A (en) | Power multi-agent dynamic collaborative inspection method and system | |
| CN110825112A (en) | Oil field dynamic invasion target tracking system and method based on multiple unmanned aerial vehicles | |
| CN117970932B (en) | Task allocation method for collaborative inspection of multiple robots of rail train | |
| CN119105503A (en) | Multi-robot collaborative scheduling method based on edge-end collaboration | |
| CN120652942B (en) | Intelligent storage multi-AGV scheduling method and system | |
| CN120802803B (en) | Low-power inspection task processing system of inspection robot | |
| CN120742810A (en) | Intelligent control system of shuttle for pallet conveying | |
| CN120313614B (en) | Path planning method, device, equipment and storage medium for omnidirectional AGV | |
| CN119624066A (en) | A robot ground station intelligent monitoring system and method | |
| Zhao et al. | When autonomous vehicles meet accidents: A DT-enabled post-accident maintenance scheme | |
| Cao et al. | A migration strategy based on cluster collaboration predictions for mobile edge computing-enabled smart rail system | |
| CN118655897A (en) | An automatic docking system for seamless docking between logistics vehicles and delivery vehicles | |
| CN110046851A (en) | Unmanned vehicle logistics method for allocating tasks based on Multi-Paxos | |
| CN116341880A (en) | A Distributed Scheduling Method for Train Inspection Robot Based on Finite State Machine | |
| CN119623804B (en) | Mine truck unloading method and system | |
| CN120663328B (en) | A method and system for division of labor and collaboration in intelligent robots | |
| CN120258275B (en) | Intelligent scheduling method and system based on artificial intelligent large model | |
| CN120630922B (en) | AGV automatic transport vehicle scheduling simulation method and system | |
| CN119270853B (en) | A networked multi-machine linkage wire-laying robot control system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |