CN109108985A - The method for diagnosing faults and system of mobile robot cluster node - Google Patents
The method for diagnosing faults and system of mobile robot cluster node Download PDFInfo
- Publication number
- CN109108985A CN109108985A CN201811263751.7A CN201811263751A CN109108985A CN 109108985 A CN109108985 A CN 109108985A CN 201811263751 A CN201811263751 A CN 201811263751A CN 109108985 A CN109108985 A CN 109108985A
- Authority
- CN
- China
- Prior art keywords
- node
- path
- local
- quasi
- malfunctioning
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 12
- 238000012545 processing Methods 0.000 claims description 17
- 238000005259 measurement Methods 0.000 claims description 10
- 238000003745 diagnosis Methods 0.000 claims description 9
- 238000005265 energy consumption Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims 1
- 230000010365 information processing Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1674—Programme controls characterised by safety, monitoring, diagnostic
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses the method for diagnosing faults of mobile robot cluster node and systems, it include: that each node of cluster connect with cloud database and uploads path data in real time, set cycle period and refresh interval, cloud database periodically carries out assessment marking and ranking to each node entirety path, set first m and latter n of respectively super node and quasi- malfunctioning node, their each local paths are mapped and assessed, set local fault threshold value and overall failure threshold value, make a reference value with local path score mean value each in super node, local fault is defined as when the ratio between certain local path score and corresponding a reference value reach local fault threshold value in certain quasi- malfunctioning node.When the ratio between the sum of local fault path distance of quasi- malfunctioning node and whole path distance are higher than overall failure threshold value, which is determined as malfunctioning node, otherwise reverts to normal node.
Description
Technical field
The present invention relates to fault diagnosis technology scheme and systems, examine more particularly to the failure of mobile robot cluster node
Disconnected method and system.
Background technique
Existing robot cluster is generally made of several ordinary nodes and a central node, each ordinary node
Data endlessly can be transmitted to central node, influence whether entire distribution as long as wherein some node breaks down
The normal operation of cluster.Therefore, how real-time fault diagnosis and positioning right and wrong effectively to be carried out to entire robot cluster system
Often it is necessary to.
It generallys use a large amount of sensor in the prior art to be acquired the operation conditions of each robot node, such as work
Movement speed, inclination angle, vibration during work etc., then be compared with the data of central node, ordinary node is judged with this
Whether robot breaks down.But due to the mechano-electronic components for having a large amount of precisions in robot, the course of work must
It can be lost, with the increasing of loss, the collected data of sensor can be increasing with the data difference of central node, because
This is easy to happen erroneous judgement after a period of work, needs periodically manually to be adjusted Centroid data.
During moveable robot movement, especially in the presence of the outdoor of many disturbing factors, road conditions have diversity and
Unpredictability, existing partial movement robot failure diagnosis scheme is then by robot movement velocity and expected inconsistent definition
Subject to failure, have ignored influence of the local road conditions to moveable robot movement, lack to identification result validation verification.
Summary of the invention
In view of the above-mentioned problems, the present invention provides the method for diagnosing faults and system of mobile robot cluster node.
In a first aspect, the present invention provides the method for diagnosing faults of mobile robot cluster node, comprising:
Each node connect with cloud database and uploads in real time path data in the robot cluster, setting cycle period T and
Refresh interval △ t, cloud database periodically carry out assessment marking and ranking to each node entirety path, set super node
Number m and quasi- malfunctioning node number n, first m and latter n of ranking is respectively super node and quasi- malfunctioning node.
Mapping matching and assessment marking are carried out to each local path of super node and quasi- malfunctioning node, set local fault
Threshold value Ylocal-meanWith overall failure threshold value Ywhole-mean, with the score mean value E [S of local path each in super nodelocal-mean]
As a reference value of the local path, as certain local path score S of quasi- malfunctioning node Plocal-pIt is reached with the ratio between corresponding a reference value
To local fault threshold value Ylocal-meanWhen be defined as local fault, whether true fault also needs to observe.When quasi- malfunctioning node
The ratio between the sum of local fault path distance and whole path distance are higher than overall failure threshold value Ywhole-mean-maxWhen, by the quasi- failure
Node is determined as malfunctioning node, and the ratio between the sum of such as quasi- malfunctioning node local fault path distance and whole path distance are lower than whole
Fault threshold Ywhole-mean-minWhen, then the node is reverted into normal node.
The mobile robot cluster should include at least the mobile robot of 2 and the above quantity.
The path data includes whole path data and local path data, specifically has path distance, by the path
Spent time, the energy consumption three parts by the path.
Each node entirety path is carried out to further include establishing respective algorithms, determine road in assessment marking and ranking described
Diameter distance, by the path the spent time, by the energy consumption in the path to the influence degree and weight of score.
In the setting super node number m and quasi- malfunctioning node number n, first m and latter n of ranking respectively super
Node and quasi- malfunctioning node.Wherein super node and quasi- malfunctioning node are to carry out periodical dynamic evaluation, it may occur that periodically
Change, evaluation criteria may be implemented, the adaptive dynamic of entire group system is adjusted.
Marking is matched and assessed in each local path to super node and quasi- malfunctioning node, further includes single
The whole path of node can be subdivided into several local paths, and the local path of super node should be with the local path of quasi- malfunctioning node
Carry out mapping matching, it is ensured that the corresponding consistency for calculating data.
In the setting local fault threshold value Ylocal-meanWith overall failure threshold value Ywhole-mean, with office each in super node
The score mean value E [S in portion pathlocal-mean] a reference value as the local path, when certain local path point of quasi- malfunctioning node P
Number Slocal-pReach local fault threshold value Y with the ratio between corresponding a reference valuelocal-meanWhen be defined as local fault, whether it is true therefore
Barrier also needs to observe.Wherein, the setting of local fault threshold value is related with artificial allowable fluctuation range, and a reference value can be with super node
Cyclically-varying and change, therefore local fault threshold value Ylocal-meanOnly to a certain particular moment.
Described when the ratio between the sum of local fault path distance of quasi- malfunctioning node and whole path distance are higher than whole event
Hinder threshold value Ywhole-mean-maxWhen, which is determined as malfunctioning node, such as quasi- malfunctioning node local fault path distance
The sum of be lower than overall failure threshold value Y with the ratio between whole path distancewhole-mean-minWhen, then the node is reverted into normal node.
Wherein, the sum of described local fault path distance be individual node local fault path distance all in whole path it
With malfunctioning node is directed at by its accounting in whole path and is further determined that, the difference for excluding local road conditions is played a game
The interference of portion's path data.
Second aspect, the present invention provide the fault diagnosis system of mobile robot cluster node, comprising: memory, wireless
Transceiver module, sensor module, message processing module, identification module, six part of label model.
The memory is for storing path data, and each robot node has memory, in addition, cloud
Database is also connected and periodically updates with memory.
Local storage of the radio receiving transmitting module for robot node is communicated with cloud database, robot
Own path data both can be transferred to cloud database plus timestamp by radio receiving transmitting module by node, can also be passed through
The push of radio receiving transmitting module reception cloud path data.
The sensor module specifically includes path distance, passes through for corresponding index data to be measured and acquired
The path spent time, the energy consumption three parts by the path.
The message processing module is connected with the sensor module, memory, the data that sensor module measurement obtains
Needing to be filtered by message processing module can just be saved in memory, mainly be filtered using anti-frequency in measurement, after measurement
Using noise reduction filtering.Finally by the respective algorithms of foundation, message processing module will be calculated and be evaluated to path data.
The identification module is for belonging to super node, ordinary node, standard to after message processing module, then to node
Malfunctioning node is recognized, and to by message processing module determine subject to malfunctioning node whether belong to malfunctioning node carry out into
The identification of one step.
The label model is used to stick the node after judgment module differentiates electronic tag classification, and label model has
Four kinds of super node, ordinary node, quasi- malfunctioning node, malfunctioning node labels.
Detailed description of the invention
Fig. 1 is the embodiment of the present invention Technology Roadmap;
Fig. 2 is the path profile of present example;
Fig. 3 is the local path statistical chart of super node;
The whole path of malfunctioning node and local path statistical chart subject to Fig. 4;
Fig. 5 is the flow diagram of the fault diagnosis system of mobile robot cluster node provided by the invention.
Specific embodiment
In order to make the purpose of the embodiment of the present invention, technical solution and compared with prior art the advantages of be more clear it is clear,
Below in conjunction with the attached drawing in embodiment, technical scheme in the embodiment of the invention is clearly and completely described.It is aobvious and easy
Insight, described embodiment are only a part of the embodiment of the present invention, instead of all the embodiments.Based on implementation of the invention
Example, relevant technical staff in the field's every other embodiment obtained without creative efforts, should all belong to
In the scope of protection of the invention.
The embodiment of the present invention is suitable for the fault diagnosis of mobile robot cluster node, the mobile robot cluster interior joint
Quantity is not less than two.Optionally, the robot cluster system of the present embodiment and existing robot cluster system difference exist
In: existing group system is made of several ordinary nodes and a central node, and ordinary node transfers data to central node,
Fault diagnosis is carried out to ordinary node by central node.And all nodes are most starting all to be ordinary node in the present invention, do not deposit
It is decentralization management in central node.
Fig. 1 is the embodiment of the present invention Technology Roadmap, and Fig. 2 is the path profile of present example, and Fig. 3 is super node
Local path statistical chart, the whole path of malfunctioning node and local path statistical chart subject to Fig. 4.The present embodiment includes:
Shown in step 101, each node connect with cloud database and uploads path data in real time in robot cluster, by computer
Database is handled and is calculated to data beyond the clouds.
It shown in step 102, needs to set cycle period T, although too long cycle period possesses higher accurate
Property, but will affect the timeliness of discovery malfunctioning node simultaneously；Although too short cycle period is easy to find malfunctioning node earlier,
Erroneous judgement can be more also easy to produce because data volume is on the low side simultaneously, and increases computation burden, therefore, it is necessary to go selection to be suitble to according to the actual situation
Cycle period.
Shown in step 103, cloud database can periodically be assessed each node entirety path according to cycle period, and give
Specific score out, with lowest fractional for 0, highest score 1.Set super node number m and quasi- malfunctioning node number n, ranking
First m is super node, and selected super node then means that its whole path data in robot cluster is preferable.After ranking
Malfunctioning node subject to n, quasi- malfunctioning node then illustrates it, and whole path data is poor in robot cluster, and score is low.
Quasi- malfunctioning node mean the node still and may be it is normal, may due in most of local path road conditions compared with
Difference causes whole path scoring low, it is therefore desirable to be directed at malfunctioning node and be further determined that.
Fig. 2 is the path profile of present example, that is to say, that the arbitrary node in robot cluster may be by Fig. 2
Free routing.In conjunction with step 104, number of the super node selected in each local path is counted, such as Fig. 3 institute
Show, so super node, in whole path, if local path AG has been walked 5 times, local path GF has been walked 8 times, local path FE
It has walked 2 times, local path HI has been walked 0 time.Therefore each local path score of super node has been counted and and then divided by corresponding office
The number that portion is walked in path, using desired value as a reference value of this section of local path.It is especially noted that super when occurring
For node when certain section of local path walking number is 0, this section of local path will be marked as invalid data, be not involved in the later period
It is directed at the assessment of malfunctioning node.
The whole path of malfunctioning node and local path statistical chart subject to as shown in Figure 4, in conjunction with step 104,105, alignment
Each local path of malfunctioning node carries out walking number statistics, and if local path AG has been walked 1 time, local path GF has walked 1 time, office
Portion path FE has been walked 2 times.Assessment marking is carried out to each section of local path again, calculates quasi- malfunctioning node in each office actually walked
Portion path average mark.The basis point that quasi- malfunctioning node local path average mark and corresponding super node obtain is subjected to mapping
Match.
In conjunction with shown in step 105,106,107, local fault threshold value Y is setlocal-meanWith overall failure threshold value
Ywhole-mean, with the score mean value E [S of local path each in super nodelocal-mean] a reference value as the local path, when
Certain local path score S of quasi- malfunctioning node Plocal-pReach local fault threshold value Y with the ratio between corresponding a reference valuelocal-meanShi Ding
Justice is local fault, and whether true fault also needs to observe.When the sum of local fault path distance of quasi- malfunctioning node and entirety
The ratio between path distance is higher than overall failure threshold value Ywhole-mean-maxWhen, which is determined as malfunctioning node, such as quasi- event
Hinder the ratio between the sum of node local fault path distance and whole path distance and is lower than overall failure threshold value Ywhole-mean-minWhen, then will
The node reverts to normal node.
Fig. 5 is the flow diagram of the fault diagnosis system of mobile robot cluster interior joint provided by the invention, comprising:
Memory, radio receiving transmitting module, sensor module, message processing module, identification module, six part of label model.
Wherein, memory is for storing path data, and each robot node has memory, in addition, cloud
Client database is also connected and periodically updates with memory.
Wherein, radio receiving transmitting module is communicated for the local storage of robot node with cloud database, machine
Own path data both can be transferred to cloud database plus timestamp by radio receiving transmitting module by people's node, can also be led to
Cross the push that radio receiving transmitting module receives cloud path data.
Wherein, sensor module specifically includes path distance, passes through for corresponding index data to be measured and acquired
The path spent time, the energy consumption three parts by the path.
Wherein, message processing module is connected with the sensor module, memory, the data that sensor module measurement obtains
Needing to be filtered by message processing module can just be saved in memory, mainly be filtered using anti-frequency in measurement, after measurement
Using noise reduction filtering.Finally by the respective algorithms of foundation, message processing module will be calculated and be evaluated to path data.
Wherein, identification module is used for after message processing module, then to node belong to super node, ordinary node,
Quasi- malfunctioning node is recognized, and is carried out to whether malfunctioning node subject to determining by message processing module belongs to malfunctioning node
Further identification.
Wherein, label model is used to stick the node after judgment module differentiates electronic tag classification, label model
There are four kinds of super node, ordinary node, quasi- malfunctioning node, malfunctioning node labels.
Finally, it should be noted that the above various embodiments is only used to illustrate the technical scheme of the present invention, rather than it is carried out
Limitation；Although the present invention is described in detail referring to each embodiment above-mentioned, the related technical personnel of this field should
Understand: it is still possible to modify the technical solutions described in the foregoing embodiments, or to part of or whole skills
Art feature carries out equivalent or close replacement, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution
Scope.
Claims (10)
1. the method for diagnosing faults of mobile robot cluster node characterized by comprising
Each node connect with cloud database and uploads in real time path data in the robot cluster, setting cycle period T and
Refresh interval △ t, cloud database periodically carry out assessment marking and ranking to each node entirety path, set super node
Number m and quasi- malfunctioning node number n, first m and latter n of ranking is respectively super node and quasi- malfunctioning node；To super section
Each local path of point and quasi- malfunctioning node carries out mapping matching and assessment marking, sets local fault threshold value Ylocal-meanWith it is whole
Body fault threshold Ywhole-mean, with the score mean value E [S of local path each in super nodelocal-mean] as the local path
A reference value, as certain local path score S of quasi- malfunctioning node Plocal-pReach local fault threshold value with the ratio between corresponding a reference value
Ylocal-meanWhen be defined as local fault, whether true fault also needs to observe；When the local fault path of quasi- malfunctioning node away from
From the sum of be higher than overall failure threshold value Y with the ratio between whole path distancewhole-mean-maxWhen, which is determined as failure
Node, the ratio between the sum of such as quasi- malfunctioning node local fault path distance and whole path distance are lower than overall failure threshold value
Ywhole-mean-minWhen, then the node is reverted into normal node.
2. the method according to claim 1, wherein the path data includes whole path data and local road
Diameter data, specifically have path distance, by the path the spent time, by the energy consumption three parts in the path；To described each
Node entirety path carries out in assessment marking and ranking, further includes establishing respective algorithms, determines path distance, by the path institute
Spend time, influence degree and weight by the energy consumption in the path to score.
3. according to claim 1 with method as claimed in claim 2, which is characterized in that the super node number m and quasi- failure
Interstitial content n is by being manually set, and number size will affect calculation amount and accuracy rate；Meanwhile each super node and quasi- failure
Node will do it periodical dynamic evaluation and change, it can be achieved that evaluation criteria adjusts the adaptive dynamic of entire group system.
4. the method according to claim 1, wherein should be with quasi- failure section by the local path to super node
The local path of point carries out mapping matching, it can be ensured that the corresponding consistency for calculating data；Meanwhile setting local fault threshold value
Ylocal-meanThe identification to local fault can be achieved, set overall failure threshold value Ywhole-meanIt is achievable to malfunctioning node and normal
The identification of node.
5. the fault diagnosis system of mobile robot cluster node characterized by comprising memory, radio receiving transmitting module, biography
Sensor module, message processing module, identification module, six part of label model.
6. system according to claim 9, which is characterized in that the memory is for storing path data, often
A robot node has memory, and is connected with cloud database and is periodically updated.
7. system according to claim 9, which is characterized in that the sensor module is used to carry out corresponding index data
Measurement and acquisition, specifically include path distance, by the path the spent time, by the energy consumption three parts in the path.
8. system according to claim 9, which is characterized in that the message processing module and the sensor module are deposited
Reservoir is connected, and the data that sensor module measurement obtains, which need to be filtered by message processing module, can just be saved in memory
In, it is mainly filtered using anti-frequency in measurement, noise reduction filtering is used after measurement；Finally by the respective algorithms of foundation, information processing
Module will be calculated and be evaluated to path data.
9. according to system described in claim 9 and claim 12, which is characterized in that the identification module is used for by believing
After ceasing processing module, then super node, ordinary node, quasi- malfunctioning node are belonged to node and recognized, and to by information
Whether malfunctioning node belongs to malfunctioning node and is further recognized subject to processing module judgement.
10. according to system described in claim 9 and claim 13, which is characterized in that the label model is used for process
Judgment module differentiate after node stick electronic tag classification, label model have super node, ordinary node, quasi- malfunctioning node,
Four kinds of labels of malfunctioning node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811263751.7A CN109108985A (en) | 2018-10-28 | 2018-10-28 | The method for diagnosing faults and system of mobile robot cluster node |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811263751.7A CN109108985A (en) | 2018-10-28 | 2018-10-28 | The method for diagnosing faults and system of mobile robot cluster node |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109108985A true CN109108985A (en) | 2019-01-01 |
Family
ID=64855738
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811263751.7A Pending CN109108985A (en) | 2018-10-28 | 2018-10-28 | The method for diagnosing faults and system of mobile robot cluster node |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109108985A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111124765A (en) * | 2019-12-06 | 2020-05-08 | 中盈优创资讯科技有限公司 | Big data cluster task scheduling method and system based on node labels |
| CN112587239A (en) * | 2020-12-30 | 2021-04-02 | 上海微创医疗机器人（集团）股份有限公司 | Medical robot, fault detection method and storage medium |
| CN113542027A (en) * | 2021-07-16 | 2021-10-22 | 中国工商银行股份有限公司 | Flow isolation method, device and system based on distributed service architecture |
-
2018
- 2018-10-28 CN CN201811263751.7A patent/CN109108985A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111124765A (en) * | 2019-12-06 | 2020-05-08 | 中盈优创资讯科技有限公司 | Big data cluster task scheduling method and system based on node labels |
| CN112587239A (en) * | 2020-12-30 | 2021-04-02 | 上海微创医疗机器人（集团）股份有限公司 | Medical robot, fault detection method and storage medium |
| CN112587239B (en) * | 2020-12-30 | 2022-04-26 | 上海微创医疗机器人（集团）股份有限公司 | Medical robot, fault detection method and storage medium |
| CN113542027A (en) * | 2021-07-16 | 2021-10-22 | 中国工商银行股份有限公司 | Flow isolation method, device and system based on distributed service architecture |
| CN113542027B (en) * | 2021-07-16 | 2022-10-11 | 中国工商银行股份有限公司 | Flow isolation method, device and system based on distributed service architecture |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109108985A (en) | The method for diagnosing faults and system of mobile robot cluster node | |
| CN105590409B (en) | A kind of tumble detection method for human body and system based on big data | |
| CN106338406B (en) | The on-line monitoring of train traction electric drive system and fault early warning system and method | |
| CN107957357A (en) | A kind of automatic method of sampling of drainage pipeline water quality and device | |
| CA2939225C (en) | Method for locating animals | |
| CN103778470B (en) | There is the distributed power generation island detection method of automatic measure on line ability | |
| CN108376286A (en) | A kind of equipment fault diagnosis apparatus and system | |
| CN106777909A (en) | Gestational period health risk assessment system | |
| CN108009485A (en) | Wireless fingerprint storehouse update method based on crowdsourcing data | |
| CN107038460A (en) | A kind of ship monitor shortage of data value complementing method based on improvement KNN | |
| CN108038884A (en) | calibration method, device, storage medium and processor | |
| CN109525956A (en) | The energy-efficient method of data capture of sub-clustering in wireless sense network based on data-driven | |
| CN106131770A (en) | A kind of data fusion method of the wireless sensor network for greenhouse | |
| CN112423225A (en) | UWB technology-based method for positioning target object in railway passenger station | |
| CN107633883A (en) | A kind of health degree evaluation method based on big data analysis | |
| CN110084336A (en) | A kind of prison institute's article management system and method based on wireless location | |
| CN116342063A (en) | Data processing method applied to intelligent agriculture | |
| CN106202113A (en) | It is applied to information providing method and the device forming a team to move | |
| CN108966169A (en) | For the Analysis on heavy metal pollution early warning system of agricultural land soil | |
| CN109830001A (en) | A kind of Data Quality Assessment Methodology and device | |
| CN107195170A (en) | A kind of intelligent electric meter wireless kilowatt meter reading-out system | |
| CN105911966A (en) | Surface meteorological observation comprehensive integration processing system | |
| CN106295238A (en) | A kind of hypertensive nephropathy Forecasting Methodology based on increment type neural network model and prognoses system | |
| CN109413193A (en) | Soil quality monitoring device for precision agriculture | |
| CN109347969A (en) | Agricultural planting ambient intelligence monitoring system based on big data |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190101 |