CN113268352A - Multi-instruction response type task collaborative management method facing general service robot - Google Patents
Multi-instruction response type task collaborative management method facing general service robot Download PDFInfo
- Publication number
- CN113268352A CN113268352A CN202110653407.4A CN202110653407A CN113268352A CN 113268352 A CN113268352 A CN 113268352A CN 202110653407 A CN202110653407 A CN 202110653407A CN 113268352 A CN113268352 A CN 113268352A
- Authority
- CN
- China
- Prior art keywords
- task
- instruction
- execution
- tut
- action
- 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
- 230000004044 response Effects 0.000 title claims abstract description 58
- 238000007726 management method Methods 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 claims abstract description 116
- 230000008569 process Effects 0.000 claims abstract description 111
- 230000007246 mechanism Effects 0.000 claims abstract description 23
- 230000009471 action Effects 0.000 claims description 148
- 238000012545 processing Methods 0.000 claims description 60
- 238000011161 development Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 7
- XWECJLRBIDJOMF-UHFFFAOYSA-N 1-[3-(2-tert-butylphenoxy)propoxy]-6,6-dimethyl-1,3,5-triazine-2,4-diamine Chemical compound CC(C)(C)C1=CC=CC=C1OCCCON1C(C)(C)NC(N)=NC1=N XWECJLRBIDJOMF-UHFFFAOYSA-N 0.000 claims description 6
- 230000002457 bidirectional effect Effects 0.000 claims description 6
- 238000004806 packaging method and process Methods 0.000 claims description 6
- 230000002093 peripheral effect Effects 0.000 claims description 6
- 230000006378 damage Effects 0.000 claims description 5
- 238000007499 fusion processing Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000011084 recovery Methods 0.000 claims description 4
- 230000008707 rearrangement Effects 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 238000005728 strengthening Methods 0.000 abstract 1
- 230000006399 behavior Effects 0.000 description 7
- 230000008859 change Effects 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 5
- 230000033001 locomotion Effects 0.000 description 5
- 238000012937 correction Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- FPQVGDGSRVMNMR-JCTPKUEWSA-N [[(z)-(1-cyano-2-ethoxy-2-oxoethylidene)amino]oxy-(dimethylamino)methylidene]-dimethylazanium;tetrafluoroborate Chemical compound F[B-](F)(F)F.CCOC(=O)C(\C#N)=N/OC(N(C)C)=[N+](C)C FPQVGDGSRVMNMR-JCTPKUEWSA-N 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000010391 action planning Effects 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004886 process control Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/54—Interprogram communication
- G06F9/542—Event management; Broadcasting; Multicasting; Notifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/54—Interprogram communication
- G06F9/546—Message passing systems or structures, e.g. queues
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Numerical Control (AREA)
- Manipulator (AREA)
Abstract
The invention provides a multi-instruction response type task collaborative management method facing a universal service robot. These three processes are independent of each other and cooperate with each other. The data exchange between the processes is realized through a message bus of the TCP. In the running process of the application system, the basic software carries out ordered scheduling on each application process task according to the characteristics of the kernel running mechanism and the application scene mode, thereby ensuring the correctness and rationality of resource use, strengthening the running management mechanism and ensuring the normal work of the service robot.
Description
Technical Field
The invention relates to the field of robot software, in particular to a multi-instruction response type task collaborative management method and device for a universal service robot.
Background
With the continuous maturity of software and hardware technologies and products, the service robot gradually enters the terminal consumer market. The service robot takes a robot body as a carrier, and provides help of entertainment, information service, behavior assistance and the like for daily life of terminal consumers in a man-machine interaction mode through software. The service robot has wide application scenes and has strong product requirements in the fields of a large number of families, public spaces, commercial facilities, entertainment facilities, cultural education, sanitation and the like. In the actual application process, the scenes have certain difference on the form and the function of the product. But has certain commonality to the technology and product requirements of the hardware carrier and the basic software level of the bottom layer robot. The aforementioned differences are more reflected in the application software level. Therefore, the development of general software and a matched platform system aiming at the service robot has positive significance for promoting the development of future intelligent industry in China. Meanwhile, the method has wide market prospect.
The service robot is a highly complex integrated body integrating hardware, software and a network. Meanwhile, the robot takes a large amount of voice, images and actions in a natural form as an interactive means. It is the product with the highest level of intellectualization and integration in the current terminal market. In the software construction process, the current development level and development trend of the main technology need to be fully considered. As general software, it is necessary to fully consider factors such as product form, application mode, user habit characteristics, and the like in the design stage. The existing voice \ video \ image recognition, high-precision positioning navigation, autonomous behavior based on machine recognition, leading-edge human-computer interaction technology and the like are organically combined together. The construction of a complete set of functions is achieved. On the basis of the function set, manufacturers realize the development and integration of application systems.
In the task execution process of the robot, the body of the robot is influenced by various factors, and due to the characteristics of the body motion, various external environment changes need to be adjusted correctly in time from the perspective of safety. In a conventional process control method, an application is composed of one process, and the application process is generally run in one CPU. The process instantiates the task according to the instructions. In this task instance, the task is broken up into a set of actions that are executed sequentially in a stack or queue fashion. During the movement of the body, the CPU in the process needs to process various external information, including voice \ image, laser radar scanning, etc. On the other hand, the execution of various actions is also scheduled. The CPU may be in a highly loaded operating state, which may lead to a lag in information processing and out of synchronization in motion scheduling. Under special conditions, the accidents of abnormal body behaviors and even harm to the safety of peripheral objects can be caused.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a multi-instruction response type task collaborative management method facing a universal service robot.
In order to achieve the technical purpose, the technical scheme adopted by the invention is as follows:
the multi-instruction response type task collaborative management method facing the universal service robot comprises the following steps:
the first step is as follows: the system is composed of three independent threads which are respectively a situation processing unit SU, a task processing unit TU and an action processing unit AU, the three threads independently run after the system is started, the whole system is deployed on an operating system of the robot, data exchange is realized among the three processes through a TCP message channel,
the second step is that: in the application development process, the definition of service logic is realized according to actual requirements, meanwhile, corresponding response action operations are defined according to different types of events, the response action definitions are packaged by a uniform interface, in the operation process, the system calls corresponding examples through configuration information,
the third step: when the robot runs and is in an on-duty state in the initial stage, waiting for an external instruction, the SU performs data processing on the basis of the received external information, identifies and filters out a definite task instruction,
the fourth step: the SU sends a task instruction to the TU, after receiving the instruction, the TU loads a task corresponding to the instruction according to the service logic definition, the task is an independent thread object TUT, the TUT completes an operation set specified by the instruction, the TUT is established and then placed in a local thread pool of the TU to perform management and scheduling in a life cycle,
the fifth step: the TU plans the tasks according to the strategy, decomposes the tasks into a group of independent instruction execution actions, arranges the tasks according to the peripheral conditions and the self-state of the robot, sets the instruction execution action arrangement scheme to the TUT after finishing the operation,
and a sixth step: after the TUT task arrangement scheme is finished, the AU is informed through the message bus, after receiving the information, the AU locally creates a corresponding task thread object AUT according to the information, the AUT and the TUT realize mapping binding, meanwhile, the AUT is placed in a local thread pool by the AU to carry out bidirectional synchronous life cycle management with the TUT,
the seventh step: after the operation is finished, the TUT starts to execute the single-step instruction action according to the task arranging scheme, in the instruction executing process, an independent sub-thread object TUTO is established for each single-step instruction execution, meanwhile, the AUTO also establishes a corresponding instruction execution sub-thread object AUTO, the AUTO and the TUTO realize the data bidirectional exchange and the state information synchronization through a message bus,
eighth step: after the AUTO and the TUTO are created and bound, the AUTO starts to act according to the instruction of the task thread object, the AUTO calls a driver of the bottom layer equipment through a uniformly packaged interface to realize the operation of the bottom layer equipment, the robot realizes the action execution in the process,
the ninth step: after the AUTO action is completed, packaging the state into a corresponding event message and feeding the event message back to the TUTO, reporting the event message to the TUT by the TUTO, finishing the current life cycle of the TUTO \ AUTO by the TUT according to a task scheduling scheme, reclaiming the occupied resources, starting and executing subsequent actions,
the tenth step: after finishing all or actions, the TUT constructs an event message of task completion and reports the event message to the TU, the TU realizes the recovery and destruction of TUT \ AUT thread object resources according to the message and puts the robot in the watch standby state again,
the eleventh step: in the process of TUTO \ AUTO action execution, the robot is influenced by the external environment, the AUTO acquires the information of action execution failure through the throwing exception called by the bottom layer, constructs corresponding event information according to information classification and sends the information to the corresponding TUTO through the information bus, the TUTO retrieves the corresponding strategy to perform corresponding processing operation after receiving the information,
the twelfth step: when the TUTO can not retrieve the current event message to correspond to the response processing, the message is reported to the TUT, the TUT carries out the rearrangement of the action task according to the configured response strategy, constructs a new response action, realizes external feedback, after the response is finished, the task continues to execute in the eighth step until the completion or the termination,
the thirteenth step: in the process of task execution, the robot receives a new instruction of a user, in the process, SU identifies the instruction and sends the instruction to TU \ AU, TU \ AU suspends the action currently executed according to a response strategy, TU judges the priority of the current instruction according to a scheduling strategy, if the current instruction needs to be executed preferentially, the current task thread TUT \ AUT is suspended and the resources occupied by the task are released, a new task object is created for the current instruction, the new task object is placed in a task pool to be executed preferentially, after the task execution is finished, whether the suspended task is executed continuously or not is judged according to the strategy, if the current instruction does not need to be executed preferentially, the suspended task is placed in a message stack, after the current task is executed, the subsequent execution scheduling is realized according to the management mechanism of the message stack,
the fourteenth step is that: in the process of executing the task of the robot, the system can generate some events, the AUTO \ TUTO receives the event messages and reports the event messages to the TUT \ AUT, the TUT \ AUT executes corresponding response strategy actions according to the event messages, and meanwhile, state information is fed back to the TU \ AU, so that the TU \ AU can schedule the task in time.
In order to optimize the technical scheme, the specific measures adopted further comprise:
the SU is responsible for receiving, processing and fusing data of external input equipment and identifying, processing and issuing instruction messages; the TU is responsible for task planning and action arrangement, and in the running process, action response and task scheduling operation during various state changes are realized; the AU is responsible for the execution of single-step actions, which implements underlying output or action mechanism pair calls through a uniformly encapsulated interface.
In daily operation, the SU establishes interception for the instruction according to system definition, when the instruction definition is hit by the data acquired by the specific equipment, the instruction is identified and extracted in a multi-channel fusion processing mode, and after the instruction extraction is completed, the instruction jobMsg is issued to the TU \ AU.
In the instruction extraction process, data hit is obtained through a single device according to definition, then a fixed time window is established, the SU obtains corresponding acquired data through different devices, whether the acquired data meet the message hit threshold setting of the application definition is judged, after the setting is met, the data are aggregated into instruction information according to the definition, if all the data meeting the requirements are not collected in the time window, the current instruction is judged to be invalid, the time window is cancelled, and the time window enters an on-duty state to wait for the issuance of a subsequent instruction.
After receiving a jobMsg instruction sent by an SU, a TU constructs a task object THT according to instruction classification and data setting, wherein the object is an independent thread object, the running state of the object does not influence the running process of the TU, the TU manages the TUT through a thread pool, the TU manages the received instruction sent by the SU through a message stack mode, after the task is executed, the next instruction is obtained from the message stack to construct a task and the task is executed, and in the scheduling process, if one task TUT is executed, the TUT constructs a task1Needs to be interrupted and wait for other tasks TUT2After the execution is finished, the system executes the TUT1The corresponding jobMsg is reset at the appropriate position of the message stack, and the TUT is terminated from the thread pool1\AUT1And after the execution of other task objects is finished, acquiring the jobMsg from the message stack again, updating the task layout and finishing the task execution.
AUT corresponding to THT is established in AU, data exchange and state information synchronization are realized between AU and THT through message bus, various action instructions of THT are issued through AUT, and AUT collects execution state information in the execution process and feeds back the execution state information to THT.
The TUTO builds a corresponding AUTO in the AUTO, the AUTO realizes the calling execution of bottom layer output/action mechanisms, and in the execution process, the AUTO controls the resource lock of the bottom layer output/action mechanisms according to the definition so as to ensure the correctness and the safety of the action execution.
And after the AUTO finishes the single action, packaging the state information into an event message and feeding the event message back to the corresponding TUTO, after the TUTO finishes the response operation, reporting the state event message to the TUT, recovering and destroying resources occupied by the TUTO \ the AUTO by the TUT, finishing the whole life cycle management operation of the single action, and then sequentially executing the next action until the whole task is finished.
And after the TUT finishes all the actions, the TUT informs the AUT and reports the AUT to the TU, the TU recovers and destroys resources occupied by the TUT \ AUT, finishes the whole life cycle management operation of the task, and acquires subsequent task instructions from the message stack according to a scheduling strategy for execution until the task is completely executed.
When the robot cannot continuously complete a task due to the influence of external conditions in the execution process, the AUTO acquires state information by capturing the exception thrown by the underlying equipment driver in the process, identifies the reason of action execution failure according to exception classification, retrieves a response strategy defined by application, and realizes the creation, execution and calling of a strategy instance in a reflection calling mode.
In the process of executing the robot task, receiving an external new command jobMsg, sending the command jobMsg to the TU by the SU, and sending the command jobMsg to the TU by the TU according to the current task THT1Comparing with the priority of the new instruction to decide scheduling execution, when the new task instruction jobMsg is higher than the current task THT1Then TU interrupts the current task THT1/AHT1Releasing THT1/AHT2Occupied resource, and constructing a new task object THT for the current command jobMsg2/AHT2Completion of execution of a task, THT2/AHT1After the execution is finished, the TU judges whether to continue to execute the THT according to the scheduling strategy1/AHT1If the execution is continued, the task programming execution scheme is re-planned, otherwise the THT is cancelled1/AHT1An object.
When the system generates some events in the process of executing the robot task, after the TU \ AU receives the messages, the TUT \ AUT which is running is subjected to scheduling response of suspending or terminating or continuing execution according to the strategy.
The operation process of the service robot has certain particularity. Typically, a task is initiated by a structured instruction. The system breaks the instructions into a specific set of actions according to the predefinition and completes all the contents one by one. On the other hand, the application scenario of the service robot has certain openness and uncertainty. In the complete execution process of a task, the state adjustment such as suspension, termination, waiting, restoration and the like needs to be carried out by combining condition identification and judgment strategies, which may be influenced by factors such as external environment interference, instruction change and the like. The service robot is composed of a hardware device having a highly complex structure and an action executing mechanism. In the above adjustment process, it is necessary to avoid situations such as destruction of the surrounding environment due to self-operation, in consideration of the safety of the body behavior. Thus, the service robot cannot implement execution of application actions in a fully task-driven manner. In the process of task execution, the self-execution action is adjusted according to the closed-loop feedback of the external environment change, and the correctness and the safety of the robot behavior are ensured.
The invention is composed of three independent processes, namely a situation processing unit (SU), a task processing unit (TU) and an action processing unit (AU). These three processes are independent of each other and cooperate with each other. The data exchange between the processes is realized through a message bus of the TCP. In the mechanism, a perception-decision-action three-unit independent cooperative organization framework is adopted according to the behavior characteristics of the robot. Each unit is made up of an independent process. The process can be run for three independent CPU nodes or for three independent kernels on one node. Each process unit independently processes respective task content. The invention can combine condition recognition and judgment strategy to adjust the state of pause, termination, waiting, recovery and the like under the influence of factors such as external environment interference, instruction change and the like, realizes the ordered scheduling of each application process task, ensures the correctness and rationality of resource use, strengthens the operation management mechanism and ensures the normal work of the service robot.
Drawings
FIG. 1 is a flow chart of the execution of a corresponding task according to the content of a particular voice command;
FIG. 2 is a flow chart of the actions performed by the present invention;
FIG. 3 is a flow chart of task scheduling of the present invention;
FIGS. 4-6 are flow diagrams of the response and processing of the multiple instruction event of the present invention;
FIG. 7 is a flow chart of the response and processing of environmental feedback events according to the present invention;
FIG. 8 is a flow chart of the response and processing of status events according to the present invention.
Detailed Description
Examples of the present invention are described in further detail below.
The multi-instruction response type task collaborative management method facing the universal service robot comprises the following steps:
the first step is as follows: the system is composed of three independent threads which are respectively a situation processing unit SU, a task processing unit TU and an action processing unit AU, the three threads independently run after the system is started, the whole system is deployed on an operating system of the robot, data exchange is realized among the three processes through a TCP message channel,
the second step is that: in the application development process, the definition of service logic is realized according to actual requirements, meanwhile, corresponding response action operations are defined according to different types of events, the response action definitions are packaged by a uniform interface, in the operation process, the system calls corresponding examples through configuration information,
the third step: when the robot runs and is in an on-duty state in the initial stage, waiting for an external instruction, the SU performs data processing on the basis of the received external information, identifies and filters out a definite task instruction,
the fourth step: the SU sends a task instruction to the TU, after receiving the instruction, the TU loads a task corresponding to the instruction according to the service logic definition, the task is an independent thread object TUT, the TUT completes an operation set specified by the instruction, the TUT is established and then placed in a local thread pool of the TU to perform management and scheduling in a life cycle,
the fifth step: the TU plans the tasks according to the strategy, decomposes the tasks into a group of independent instruction execution actions, arranges the tasks according to the peripheral conditions and the self-state of the robot, sets the instruction execution action arrangement scheme to the TUT after finishing the operation,
and a sixth step: after the TUT task arrangement scheme is finished, the AU is informed through the message bus, after receiving the information, the AU locally creates a corresponding task thread object AUT according to the information, the AUT and the TUT realize mapping binding, meanwhile, the AUT is placed in a local thread pool by the AU to carry out bidirectional synchronous life cycle management with the TUT,
the seventh step: after the operation is finished, the TUT starts to execute the single-step instruction action according to the task arranging scheme, in the instruction executing process, an independent sub-thread object TUTO is established for each single-step instruction execution, meanwhile, the AUTO also establishes a corresponding instruction execution sub-thread object AUTO, the AUTO and the TUTO realize the data bidirectional exchange and the state information synchronization through a message bus,
eighth step: after the AUTO and the TUTO are created and bound, the AUTO starts to act according to the instruction of the task thread object, the AUTO calls a driver of the bottom layer equipment through a uniformly packaged interface to realize the operation of the bottom layer equipment, the robot realizes the action execution in the process,
the ninth step: after the AUTO action is completed, packaging the state into a corresponding event message and feeding the event message back to the TUTO, reporting the event message to the TUT by the TUTO, finishing the current life cycle of the TUTO \ AUTO by the TUT according to a task scheduling scheme, reclaiming the occupied resources, starting and executing subsequent actions,
the tenth step: after finishing all or actions, the TUT constructs an event message of task completion and reports the event message to the TU, the TU realizes the recovery and destruction of TUT \ AUT thread object resources according to the message and puts the robot in the watch standby state again,
the eleventh step: during the execution of the actions of TUTO \ AUTO, the robot is affected by the external environment, for example: when the barrier can not cross and can not execute the action, the AUTO acquires the information of action execution failure through the throwing exception called by the bottom layer, constructs corresponding event information according to information classification, sends the information to the corresponding TUTO through the information bus, retrieves the corresponding strategy to perform corresponding processing operation after the TUTO receives the information,
the twelfth step: when the TUTO can not retrieve the current event message to correspond to the response processing, the message is reported to the TUT, the TUT carries out the rearrangement of the action task according to the configured response strategy, constructs a new response action, realizes external feedback, after the response is finished, the task continues to execute in the eighth step until the completion or the termination,
the thirteenth step: in the process of task execution, the robot receives a new instruction of a user, in the process, SU identifies the instruction and sends the instruction to TU \ AU, TU \ AU suspends the action currently executed according to a response strategy, TU judges the priority of the current instruction according to a scheduling strategy, if the current instruction needs to be executed preferentially, the current task thread TUT \ AUT is suspended and the resources occupied by the task are released, a new task object is created for the current instruction, the new task object is placed in a task pool to be executed preferentially, after the task execution is finished, whether the suspended task is executed continuously or not is judged according to the strategy, if the current instruction does not need to be executed preferentially, the suspended task is placed in a message stack, after the current task is executed, the subsequent execution scheduling is realized according to the management mechanism of the message stack,
the fourteenth step is that: in the process of executing the task of the robot, the system can generate some events, the AUTO \ TUTO receives the event messages and reports the event messages to the TUT \ AUT, the TUT \ AUT executes corresponding response strategy actions according to the event messages, and meanwhile, state information is fed back to the TU \ AU, so that the TU \ AU can schedule the task in time.
In the embodiment, the SU is responsible for receiving, processing and fusing data of external input equipment and identifying, processing and issuing instruction messages; the TU is responsible for task planning and action arrangement, and in the running process, action response and task scheduling operation during various state changes are realized; the AU is responsible for the execution of single-step actions, which implements underlying output or action mechanism pair calls through a uniformly encapsulated interface.
In daily operation, the SU establishes interception for the instruction according to system definition, when the instruction definition is hit by the data acquired by the specific equipment, the instruction is identified and extracted in a multi-channel fusion processing mode, and after the instruction extraction is completed, the instruction jobMsg is issued to the TU \ AU.
In the instruction extraction process, data hit is obtained through a single device according to definition, then a fixed time window is established, the SU obtains corresponding acquired data through different devices, whether the acquired data meet the message hit threshold setting of the application definition is judged, after the setting is met, the data are aggregated into instruction information according to the definition, if all the data meeting the requirements are not collected in the time window, the current instruction is judged to be invalid, the time window is cancelled, and the time window enters an on-duty state to wait for the issuance of a subsequent instruction.
After receiving a jobMsg instruction sent by an SU, a TU constructs a task object THT according to instruction classification and data setting, wherein the object is an independent thread object, the running state of the object does not influence the running process of the TU, the TU manages the TUT through a thread pool, the TU manages the received instruction sent by the SU through a message stack mode, after the task is executed, the next instruction is obtained from the message stack to construct a task and the task is executed, and in the scheduling process, if one task TUT is executed, the TUT constructs a task1Needs to be interrupted and wait for other tasks TUT2After the execution is finished, the system executes the TUT1The corresponding jobMsg is reset at the appropriate position of the message stack, and the TUT is terminated from the thread pool1\AUT1After the execution of other task objects is finished, the jobMsg is obtained again from the message stack, and the task arrangement is finished after the task arrangement is updatedAnd executing the task.
AUT corresponding to THT is established in AU, data exchange and state information synchronization are realized between AU and THT through message bus, various action instructions of THT are issued through AUT, and AUT collects execution state information in the execution process and feeds back the execution state information to THT.
The TUTO builds a corresponding AUTO in the AUTO, the AUTO realizes the calling execution of bottom layer output/action mechanisms, and in the execution process, the AUTO controls the resource lock of the bottom layer output/action mechanisms according to the definition so as to ensure the correctness and the safety of the action execution.
And after the AUTO finishes the single action, packaging the state information into an event message and feeding the event message back to the corresponding TUTO, after the TUTO finishes the response operation, reporting the state event message to the TUT, recovering and destroying resources occupied by the TUTO \ the AUTO by the TUT, finishing the whole life cycle management operation of the single action, and then sequentially executing the next action until the whole task is finished.
And after the TUT finishes all the actions, the TUT informs the AUT and reports the AUT to the TU, the TU recovers and destroys resources occupied by the TUT \ AUT, finishes the whole life cycle management operation of the task, and acquires subsequent task instructions from the message stack according to a scheduling strategy for execution until the task is completely executed.
When the robot cannot continuously complete a task due to the influence of external conditions in the execution process, the AUTO acquires state information by capturing the exception thrown by the underlying equipment driver in the process, identifies the reason of action execution failure according to exception classification, retrieves a response strategy defined by application, and realizes the creation, execution and calling of a strategy instance in a reflection calling mode.
In the process of executing the robot task, receiving an external new command jobMsg, sending the command jobMsg to the TU by the SU, and sending the command jobMsg to the TU by the TU according to the current task THT1Comparing with the priority of the new instruction to decide scheduling execution, when the new task instruction jobMsg is higher than the current task THT1Then TU interrupts the current task THT1/AHT1Releasing THT1/AHT2Occupied resource, and constructing a new task object THT for the current command jobMsg2/AHT2Completion of execution of a task, THT2/AHT1After the execution is finished, the TU judges whether to continue to execute the THT according to the scheduling strategy1/AHT1If the execution is continued, the task programming execution scheme is re-planned, otherwise the THT is cancelled1/AHT1An object.
When the system generates some events in the process of executing the robot task, after the TU \ AU receives the messages, the TUT \ AUT which is running is subjected to scheduling response of suspending or terminating or continuing execution according to the strategy.
Situation processing unit (SU) of the present invention: the situation processing unit is connected with various external input devices including audio and video acquisition devices, a laser radar, accurate positioning and the like through the bottom layer. In the operation process, the situation processing unit establishes a corresponding data receiving channel and a data processing module for the equipment. In the increment process, the acquisition of external data and the extraction of information are realized. Meanwhile, according to task definition, an on-duty mechanism is established, various information collected in real time is subjected to fusion processing, various task related messages required by application are identified and captured, and the generated messages are issued to other process units, so that bases are provided for decision and action execution;
task processing unit (TU): the task processing unit is responsible for executing the task process according to the application definition. It breaks up the task into a set of execution actions and implements the execution order organization through the execution stack. And sending the message schema defined by the task to the situation processing unit. During the task processing, the execution action instructions are sent to the action processing unit according to the execution sequence. In the action processing process, a corresponding binding relation is established, and various feedback information of the machine body is collected in time when the action processing unit executes the current action. On the other hand, it also acquires external environment information through the situation processing unit. On the basis of the fusion processing of the two kinds of information, responses such as pause, termination, change and the like are made in time to the current action according to the application predefined strategy, and the correctness and the safety of the feedback response of the machine body are ensured while the task execution is met;
motion processing unit (AU): the action processing unit is responsible for performing specific single-step tasks. Which controls the output device and the action executing mechanism. During operation, the task processing unit receives the action instruction sent by the task processing unit. And controlling the operation of the related equipment/mechanism according to the action instruction. And collecting the state information, the execution process information and the like of the machine body in the action execution process and feeding back the state information, the execution process information and the like to the task processing unit in time. Meanwhile, the situation processing unit receives the relevant messages sent by the situation processing unit. And performing corresponding control adjustment on the current body action according to the response definition.
The following describes the task and action execution scheduling method of the present invention specifically:
1.1 initiation and execution of tasks
In the design stage of the service robot, developers can realize the customized development of different tasks according to application requirements, and specific task flow logics are arranged. Meanwhile, according to the basic principle of the service robot, response definition is carried out on the external event message received in the task execution process. Ensuring the normal operation of the robot. In daily operation, the service robot is in a standby state. Which typically performs the corresponding task execution based on the specific voice command content. In the process, the robot receives voice data through the situation processing unit (SU) in the on-duty state, and after the voice data is identified as effective instruction content, the information is sent to the task processing unit (TU). And after receiving the task instruction, the TU schedules the task according to the current machine body state and executes the task content required by the current voice instruction. This process is illustrated in figure 1.
And (3) the whole task execution process:
1. a task instruction receiving stage: and the situation processing unit (SU) receives the instruction information understanding instruction through the audio acquisition equipment, maps the instruction information understanding instruction into an instruction code, and retrieves the corresponding message rule according to the code. And acquiring relevant data from other peripheral equipment within the time window of the constraint according to the message rule definition. Information is extracted from different peripherals, a message is constructed and sent to the task processing unit to start subsequent execution.
2. And (3) task decision and planning stage: the task processing unit (TU) receives the message and retrieves a corresponding response operation definition based on the message type.
And if the task is an emergency task, directly calling the behavior processing unit, and if the current task is a common task, planning and arranging the task. The system instantiates and loads the task object according to the message content, plans according to the application definition and the current machine body state, decomposes the task into a group of execution actions and plans and arranges. The associated response seed policy is configured at each execution action. Meanwhile, the priority of the execution action is configured according to the priority of the current task. And after all processing is finished, the task information is sent to the action processing unit.
3. And an action execution stage: and an action processing unit (AU) receives the action execution information of the task and schedules the task action according to the priority. And the AU starts to execute the current task, acquires the current state and records the current state. According to the task plan, the method is executed according to the steps. And in the execution process, calling each external device according to the strategy. And in the execution process of each planning step, maintaining the current state, completing the execution follow-up of all tasks, recovering the state according to the definition and informing the system.
1.2 execution of actions
During the execution of the whole task, the TU establishes an independent thread object (TUTO) for the single-step execution action based on the task main thread TUT. Meanwhile, the AUT bound in the AU establishes a corresponding executive thread object (AUTO), and the two thread objects realize mapping binding. In the operation process, the synchronization of the state information is realized through the message bus.
In the action execution starting phase, the TUT first instantiates the TUTO of the instruction action according to the current state, and simultaneously completes the loading and the setting of the related data. After the processing is completed, the object is placed in an object pool for subsequent management. The TUT then sends relevant information to the AUT over the bus based on the TUTO serialization. And after receiving the information, the AUT automatically establishes a corresponding AUTO object locally and manages the AUTO object in a local object pool. And the TUT binds the TUTO and the AUTO after receiving the feedback information, and then a task main thread of the TUT starts the thread execution of the TUTO. And after the thread is started, the TUTO performs corresponding starting execution operation on the corresponding AUTO thread through the task main thread of the AUT. The thread internal of the AUTO continues to execute according to the action defined by the application until the current action is executed or terminated by an external condition. This process is illustrated in fig. 2.
Usually one process is required for one action execution. In the process, the AUTO continuously controls the operation of the output device/action mechanism in an independent thread mode, and feeds back the state information to the TOTU through a message bus. Since AUTO and TOTU are both independent of TUT and TUT task main threads, the thread is an action child thread. Therefore, the task main thread can adjust the action sub-thread according to the external environment change and the task execution state so as to achieve the purposes of timely responding and correcting.
1.3 task adjustment and action correction during execution
In the action execution process, when influenced by external environment (such as new voice instruction receiving, the existence of an unavoidable obstacle in the current action route, the limitation of the motion of the current action mechanism and the like), the robot needs to adjust the current action in time to ensure the safety. This correction according to the body state mainly comprises two links:
1. suspending current action execution: and stopping the current robot in time according to the current state of the robot. And generates a related event message to transmit to the TU.
2. And correcting the current execution action: and the TU performs new instruction action planning on the current execution action according to the event message and the preset strategy of the application. On this basis, the suspended TUTO \ AUTO is replaced with the newly programmed TUTO \ AUTO, and execution is started.
1.4 task scheduling
The robot is interrupted by new instructions in certain situations while the task is being executed. In the process, the TU manages the received instruction sent by the SU in a message stack manner. After the task is executed, the slave messageAnd acquiring a next instruction construction task in the stack and executing. During scheduling, if a task TUT1Needs to be interrupted and wait for other tasks TUT2And executing the operation after the execution is finished. The system will TUT1The corresponding jobMsg is reset at the appropriate position of the message stack, and the TUT is terminated from the thread pool1\AUT1And after the execution of other task objects is finished, acquiring the jobMsg from the message stack again, updating the task layout and finishing the task execution. As shown in fig. 3.
2. Event response and handling
During the task execution of the robot, it is necessary to execute a correction task due to factors such as its own operation restriction, the surrounding environment, and external commands. In this process, the system needs to perform various intervention operations on the execution threads of the application tasks according to the characteristics of various events. In a system, events are basically composed of several categories:
1. a multi-instruction event: and sending the instructions to the robot in a voice mode, a network connection transmission mode and the like. The robot needs to execute corresponding tasks according to the instruction content. If the robot is currently in the execution state of a certain task. Scheduling is carried out according to the priority of the instruction, and the normal work of the robot is ensured.
2. Environmental feedback events: during operation, events are formed due to obstacles, ambient temperature, and the like. And after receiving the relevant information through the SU, the robot constructs a corresponding event message and sends the event message to the TU. And the TU replans the tasks according to the strategy, corrects the currently executed action and ensures that the robot smoothly finishes the current operation.
3. Executing the state event: in the task execution process of TU \ AU, the system executes some state events of each stage in the instruction action. The event system can master the starting, intermediate process, ending, pause and other states of each action execution, so that the corresponding configuration strategy can be called to ensure the correct execution of the action.
4. System state events: some events such as internal network communication, bottom layer data reading and writing and the like in the system operation process. Under specific conditions, corresponding strategies are applied and configured so as to respond in time when the system has internal faults and ensure the overall safety and reliability of the system.
2.1 response and handling of multiple instruction events
And in the running process of the robot, the SU acquires external audio data through the audio acquisition equipment. The instruction information is identified after being processed. And after the instruction information is identified, fusing other data according to the message format customized by the application to form an instruction event message, and then sending the instruction event message to the task processing (TU). This process is illustrated in fig. 4.
And after receiving the instruction event message, the TU controls the task main thread which is currently executed by a scheduling mechanism and judges whether the current instruction is consistent with the content of the task which is currently executed. If the instruction is consistent with the instruction, the event message is sent to the task main thread, and the task processes the instruction by itself. And if not, establishing a new task object according to the instruction. And scheduling the current task object and the newly-built task object to be executed according to the strategy. The basic scheduling architecture is shown in fig. 5. The specific execution flow is shown in fig. 6.
2.2 response and handling of environmental feedback events
During the task execution process, the robot is influenced by external environmental factors, so that the task execution cannot be continued. Under the condition, the system forms an environment feedback event message, the TU adjusts the current action execution according to the event message, replans the execution action according to the configured strategy and finishes the execution process correction of the task. The robot achieves a correct feedback response.
As mentioned before, the ambient feedback event is initiated by the AU. And acquiring the state information of the action execution failure by the action execution of the AU in the running process in an abnormal capture mode. And forming a corresponding environment feedback event message by combining the state definition of the application configuration and sending the environment feedback event message to the TU. And the TU replans according to the content and feeds back the result to the AU. The AU re-executes the action or executes a new action according to the content, completing the responsive execution of the environmental feedback event. This process is illustrated in fig. 7.
2.3 response and handling of State events
During the running process of the task, each thread object can experience the states of starting, executing, suspending, continuing, withdrawing, interrupting and the like. Because a certain cooperative relationship exists among a plurality of tasks, the objects need to respond in time according to the state changes of other objects so as to ensure the normal operation of the machine body.
In the above process, when the state of the object changes, corresponding event messages are constructed, and by using these messages, state synchronization of the mapping thread object between the TU and the AU and scheduling execution of the upper thread object are realized. The process is shown in fig. 8.
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-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.
Claims (12)
1. The multi-instruction response type task collaborative management method facing the universal service robot is characterized in that: the method comprises the following steps:
the first step is as follows: the system is composed of three independent threads which are respectively a situation processing unit SU, a task processing unit TU and an action processing unit AU, the three threads independently run after the system is started, the whole system is deployed on an operating system of the robot, data exchange is realized among the three processes through a TCP message channel,
the second step is that: in the application development process, the definition of service logic is realized according to actual requirements, meanwhile, corresponding response action operations are defined according to different types of events, the response action definitions are packaged by a uniform interface, in the operation process, the system calls corresponding examples through configuration information,
the third step: when the robot runs and is in an on-duty state in the initial stage, waiting for an external instruction, the SU performs data processing on the basis of the received external information, identifies and filters out a definite task instruction,
the fourth step: the SU sends a task instruction to the TU, after receiving the instruction, the TU loads a task corresponding to the instruction according to the service logic definition, the task is an independent thread object TUT, the TUT completes an operation set specified by the instruction, the TUT is established and then placed in a local thread pool of the TU to perform management and scheduling in a life cycle,
the fifth step: the TU plans the tasks according to the strategy, decomposes the tasks into a group of independent instruction execution actions, arranges the tasks according to the peripheral conditions and the self-state of the robot, sets the instruction execution action arrangement scheme to the TUT after finishing the operation,
and a sixth step: after the TUT task arrangement scheme is finished, the AU is informed through the message bus, after receiving the information, the AU locally creates a corresponding task thread object AUT according to the information, the AUT and the TUT realize mapping binding, meanwhile, the AUT is placed in a local thread pool by the AU to carry out bidirectional synchronous life cycle management with the TUT,
the seventh step: after the operation is finished, the TUT starts to execute the single-step instruction action according to the task arranging scheme, in the instruction executing process, an independent sub-thread object TUTO is established for each single-step instruction execution, meanwhile, the AUTO also establishes a corresponding instruction execution sub-thread object AUTO, the AUTO and the TUTO realize the data bidirectional exchange and the state information synchronization through a message bus,
eighth step: after the AUTO and the TUTO are created and bound, the AUTO starts to act according to the instruction of the task thread object, the AUTO calls a driver of the bottom layer equipment through a uniformly packaged interface to realize the operation of the bottom layer equipment, the robot realizes the action execution in the process,
the ninth step: after the AUTO action is completed, packaging the state into a corresponding event message and feeding the event message back to the TUTO, reporting the event message to the TUT by the TUTO, finishing the current life cycle of the TUTO \ AUTO by the TUT according to a task scheduling scheme, reclaiming the occupied resources, starting and executing subsequent actions,
the tenth step: after finishing all or actions, the TUT constructs an event message of task completion and reports the event message to the TU, the TU realizes the recovery and destruction of TUT \ AUT thread object resources according to the message and puts the robot in the watch standby state again,
the eleventh step: in the process of TUTO \ AUTO action execution, the robot is influenced by the external environment, the AUTO acquires the information of action execution failure through the throwing exception called by the bottom layer, constructs corresponding event information according to information classification and sends the information to the corresponding TUTO through the information bus, the TUTO retrieves the corresponding strategy to perform corresponding processing operation after receiving the information,
the twelfth step: when the TUTO can not retrieve the current event message to correspond to the response processing, the message is reported to the TUT, the TUT carries out the rearrangement of the action task according to the configured response strategy, constructs a new response action, realizes external feedback, after the response is finished, the task continues to execute in the eighth step until the completion or the termination,
the thirteenth step: in the process of task execution, the robot receives a new instruction of a user, in the process, SU identifies the instruction and sends the instruction to TU \ AU, TU \ AU suspends the action currently executed according to a response strategy, TU judges the priority of the current instruction according to a scheduling strategy, if the current instruction needs to be executed preferentially, the current task thread TUT \ AUT is suspended and the resources occupied by the task are released, a new task object is created for the current instruction, the new task object is placed in a task pool to be executed preferentially, after the task execution is finished, whether the suspended task is executed continuously or not is judged according to the strategy, if the current instruction does not need to be executed preferentially, the suspended task is placed in a message stack, after the current task is executed, the subsequent execution scheduling is realized according to the management mechanism of the message stack,
the fourteenth step is that: in the process of executing the task of the robot, the system can generate some events, the AUTO \ TUTO receives the event messages and reports the event messages to the TUT \ AUT, the TUT \ AUT executes corresponding response strategy actions according to the event messages, and meanwhile, state information is fed back to the TU \ AU, so that the TU \ AU can schedule the task in time.
2. The universal service robot-oriented multi-instruction responsive task collaborative management method according to claim 1, characterized in that: the SU is responsible for receiving, processing and fusing data of external input equipment and identifying, processing and issuing instruction messages; the TU is responsible for task planning and action arrangement, and in the running process, action response and task scheduling operation during various state changes are realized; the AU is responsible for the execution of single-step actions, which implements underlying output or action mechanism pair calls through a uniformly encapsulated interface.
3. The universal service robot-oriented multi-instruction responsive task collaborative management method according to claim 1, characterized in that: in daily operation, the SU establishes interception for the instruction according to system definition, when the instruction definition is hit by the data acquired by the specific equipment, the instruction is identified and extracted in a multi-channel fusion processing mode, and after the instruction extraction is completed, the instruction jobMsg is issued to the TU \ AU.
4. The universal service robot-oriented multi-instruction responsive task collaborative management method according to claim 1, characterized in that: in the instruction extraction process, data hit is obtained through a single device according to definition, then a fixed time window is established, the SU obtains corresponding acquired data through different devices, whether the acquired data meet the message hit threshold setting of the application definition is judged, after the setting is met, the data are aggregated into instruction information according to the definition, if all the data meeting the requirements are not collected in the time window, the current instruction is judged to be invalid, the time window is cancelled, and the time window enters an on-duty state to wait for the issuance of a subsequent instruction.
5. The universal service robot-oriented multi-instruction responsive task collaborative management method according to claim 1, characterized in that: after receiving a jobMsg instruction sent by an SU, a TU constructs a task object THT according to instruction classification and data setting, wherein the object is an independent thread object, the running state of the object does not influence the running process of the TU, the TU manages the TUT through a thread pool, the TU manages the received instruction sent by the SU through a message stack mode, after the task is executed, a next instruction construction task is obtained from the message stack and executed, and in the scheduling process, the next instruction construction task is executedIf a task TUT1Needs to be interrupted and wait for other tasks TUT2After the execution is finished, the system executes the TUT1The corresponding jobMsg is reset at the appropriate position of the message stack, and the TUT is terminated from the thread pool1\AUT1And after the execution of other task objects is finished, acquiring the jobMsg from the message stack again, updating the task layout and finishing the task execution.
6. The universal service robot-oriented multi-instruction responsive task collaborative management method according to claim 1, characterized in that: AUT corresponding to THT is established in AU, data exchange and state information synchronization are realized between AU and THT through message bus, various action instructions of THT are issued through AUT, and AUT collects execution state information in the execution process and feeds back the execution state information to THT.
7. The universal service robot-oriented multi-instruction responsive task collaborative management method according to claim 1, characterized in that: the TUTO builds a corresponding AUTO in the AUTO, the AUTO realizes the calling execution of bottom layer output/action mechanisms, and in the execution process, the AUTO controls the resource lock of the bottom layer output/action mechanisms according to the definition so as to ensure the correctness and the safety of the action execution.
8. The cooperative management method for multi-instruction responsive tasks of the universal service robot as claimed in claim 5, 6 or 7, wherein: and after the AUTO finishes the single action, packaging the state information into an event message and feeding the event message back to the corresponding TUTO, after the TUTO finishes the response operation, reporting the state event message to the TUT, recovering and destroying resources occupied by the TUTO \ the AUTO by the TUT, finishing the whole life cycle management operation of the single action, and then sequentially executing the next action until the whole task is finished.
9. The universal service robot-oriented multi-instruction responsive task collaborative management method according to claim 8, wherein: and after the TUT finishes all the actions, the TUT informs the AUT and reports the AUT to the TU, the TU recovers and destroys resources occupied by the TUT \ AUT, finishes the whole life cycle management operation of the task, and acquires subsequent task instructions from the message stack according to a scheduling strategy for execution until the task is completely executed.
10. The cooperative management method of multi-instruction responsive tasks for a universal service robot as claimed in any one of claims 1 to 6, wherein: when the robot cannot continuously complete a task due to the influence of external conditions in the execution process, the AUTO acquires state information by capturing the exception thrown by the underlying equipment driver in the process, identifies the reason of action execution failure according to exception classification, retrieves a response strategy defined by application, and realizes the creation, execution and calling of a strategy instance in a reflection calling mode.
11. The universal service robot-oriented multi-instruction responsive task collaborative management method according to claim 3, wherein: in the process of executing the robot task, receiving an external new command jobMsg, sending the command jobMsg to the TU by the SU, and sending the command jobMsg to the TU by the TU according to the current task THT1Comparing with the priority of the new instruction to decide scheduling execution, when the new task instruction jobMsg is higher than the current task THT1Then TU interrupts the current task THT1/AHT1Releasing THT1/AHT2Occupied resource, and constructing a new task object THT for the current command jobMsg2/AHT2Completion of execution of a task, THT2/AHT1After the execution is finished, the TU judges whether to continue to execute the THT according to the scheduling strategy1/AHT1If the execution is continued, the task programming execution scheme is re-planned, otherwise the THT is cancelled1/AHT1An object.
12. The universal service robot-oriented multi-instruction responsive task collaborative management method according to claim 1, characterized in that: when the system generates some events in the process of executing the robot task, after the TU \ AU receives the messages, the TUT \ AUT which is running is subjected to scheduling response of suspending or terminating or continuing execution according to the strategy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110653407.4A CN113268352B (en) | 2021-06-11 | 2021-06-11 | Multi-instruction responsive task collaborative management method for universal service robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110653407.4A CN113268352B (en) | 2021-06-11 | 2021-06-11 | Multi-instruction responsive task collaborative management method for universal service robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113268352A true CN113268352A (en) | 2021-08-17 |
| CN113268352B CN113268352B (en) | 2024-03-08 |
Family
ID=77234871
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110653407.4A Active CN113268352B (en) | 2021-06-11 | 2021-06-11 | Multi-instruction responsive task collaborative management method for universal service robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113268352B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114416049A (en) * | 2021-12-23 | 2022-04-29 | 北京来也网络科技有限公司 | Configuration method and device of service interface combining RPA and AI |
| CN116823203A (en) * | 2023-07-17 | 2023-09-29 | 先看看闪聘(江苏)数字科技有限公司 | Recruitment system and recruitment method based on AI large language model |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101401002A (en) * | 2005-12-20 | 2009-04-01 | 俄亥俄州立大学研究基金会 | Nanoporous substrates for analytical methods |
| CN102914967A (en) * | 2012-09-21 | 2013-02-06 | 浙江工业大学 | Autonomous navigation and man-machine coordination picking operating system of picking robot |
| CN103077140A (en) * | 2013-02-05 | 2013-05-01 | 北京配天大富精密机械有限公司 | Communication method and communication device for motion control card of robot and host |
| CN104965426A (en) * | 2015-06-24 | 2015-10-07 | 百度在线网络技术（北京）有限公司 | Intelligent robot control system, method and device based on artificial intelligence |
| CN204971277U (en) * | 2015-08-24 | 2016-01-20 | 华南理工大学 | Can realize health service robot of brain electric -examination survey |
| CN106030427A (en) * | 2014-02-20 | 2016-10-12 | M·奥利尼克 | Methods and systems for food preparation in a robotic cooking kitchen |
| CN107671887A (en) * | 2017-08-22 | 2018-02-09 | 广东美的智能机器人有限公司 | Robot self-test control method, robot and dispatch server |
| CN107756395A (en) * | 2016-08-19 | 2018-03-06 | 腾讯科技（深圳）有限公司 | Control system, the method and apparatus of intelligent robot |
| CN109165055A (en) * | 2018-08-30 | 2019-01-08 | 百度在线网络技术（北京）有限公司 | A kind of component loading method, device, computer equipment and storage medium |
| CN109605388A (en) * | 2018-12-27 | 2019-04-12 | 南京熊猫电子股份有限公司 | A kind of long-range control method based on service robot stage task |
| CN111309880A (en) * | 2020-01-21 | 2020-06-19 | 清华大学 | Multi-agent action strategy learning method, device, medium and computing equipment |
| CN111427751A (en) * | 2020-04-15 | 2020-07-17 | 赞同科技股份有限公司 | Method and system for processing service based on asynchronous processing mechanism |
| CN111797048A (en) * | 2020-05-27 | 2020-10-20 | 深圳壹账通智能科技有限公司 | Data processing method, device, equipment and computer readable storage medium |
| CN112486171A (en) * | 2020-11-30 | 2021-03-12 | 中科院软件研究所南京软件技术研究院 | Robot obstacle avoidance method based on vision |
| CN112518778A (en) * | 2020-12-22 | 2021-03-19 | 上海原圈网络科技有限公司 | Control method of intelligent man-machine fusion scene based on service robot |
-
2021
- 2021-06-11 CN CN202110653407.4A patent/CN113268352B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101401002A (en) * | 2005-12-20 | 2009-04-01 | 俄亥俄州立大学研究基金会 | Nanoporous substrates for analytical methods |
| CN102914967A (en) * | 2012-09-21 | 2013-02-06 | 浙江工业大学 | Autonomous navigation and man-machine coordination picking operating system of picking robot |
| CN103077140A (en) * | 2013-02-05 | 2013-05-01 | 北京配天大富精密机械有限公司 | Communication method and communication device for motion control card of robot and host |
| CN106030427A (en) * | 2014-02-20 | 2016-10-12 | M·奥利尼克 | Methods and systems for food preparation in a robotic cooking kitchen |
| CN104965426A (en) * | 2015-06-24 | 2015-10-07 | 百度在线网络技术（北京）有限公司 | Intelligent robot control system, method and device based on artificial intelligence |
| CN204971277U (en) * | 2015-08-24 | 2016-01-20 | 华南理工大学 | Can realize health service robot of brain electric -examination survey |
| CN107756395A (en) * | 2016-08-19 | 2018-03-06 | 腾讯科技（深圳）有限公司 | Control system, the method and apparatus of intelligent robot |
| CN107671887A (en) * | 2017-08-22 | 2018-02-09 | 广东美的智能机器人有限公司 | Robot self-test control method, robot and dispatch server |
| CN109165055A (en) * | 2018-08-30 | 2019-01-08 | 百度在线网络技术（北京）有限公司 | A kind of component loading method, device, computer equipment and storage medium |
| CN109605388A (en) * | 2018-12-27 | 2019-04-12 | 南京熊猫电子股份有限公司 | A kind of long-range control method based on service robot stage task |
| CN111309880A (en) * | 2020-01-21 | 2020-06-19 | 清华大学 | Multi-agent action strategy learning method, device, medium and computing equipment |
| CN111427751A (en) * | 2020-04-15 | 2020-07-17 | 赞同科技股份有限公司 | Method and system for processing service based on asynchronous processing mechanism |
| CN111797048A (en) * | 2020-05-27 | 2020-10-20 | 深圳壹账通智能科技有限公司 | Data processing method, device, equipment and computer readable storage medium |
| CN112486171A (en) * | 2020-11-30 | 2021-03-12 | 中科院软件研究所南京软件技术研究院 | Robot obstacle avoidance method based on vision |
| CN112518778A (en) * | 2020-12-22 | 2021-03-19 | 上海原圈网络科技有限公司 | Control method of intelligent man-machine fusion scene based on service robot |
Non-Patent Citations (5)
| Title |
|---|
| AMANDA HOLLOMAN: "Leveraging Neurophysiological Information to Augment Interpretation of Responses to Vulnerable Robot Behaviors", 《2019 14TH ACM/IEEE INTERNATIONAL CONFERENCE ON HUMAN-ROBOT INTERACTION (HRI)》 * |
| HAOMING GUO: "Autonomous and dynamic Web service composition in wireless grids", 《INTERNATIONAL CONFERENCE ON INFORMATION TECHNOLOGY: CODING AND COMPUTING (ITCC\'05) - VOLUME II》 * |
| 刘兵: "基于行为编程的移动机器人室内导航系统研究", 《中国优秀硕士论文全文数据库》, 15 February 2021 (2021-02-15) * |
| 工业机器人培训: "【超级干货】机器人控制系统相关知识大汇集", 《HTTPS://WWW.SOHU.COM/A/215042705_202023》, 6 January 2018 (2018-01-06) * |
| 郭黎敏等: "基于停留时间的语义行为模式挖掘", 《计算机研究与发展》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114416049A (en) * | 2021-12-23 | 2022-04-29 | 北京来也网络科技有限公司 | Configuration method and device of service interface combining RPA and AI |
| CN116823203A (en) * | 2023-07-17 | 2023-09-29 | 先看看闪聘(江苏)数字科技有限公司 | Recruitment system and recruitment method based on AI large language model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113268352B (en) | 2024-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113268352B (en) | Multi-instruction responsive task collaborative management method for universal service robot | |
| CN112549029B (en) | Robot behavior control method and device based on behavior tree | |
| CN103645678B (en) | A kind of real-time response network control system | |
| CN105260248B (en) | A kind of task scheduling system and dispatching method being dynamically adapted | |
| CN106293919A (en) | The built-in tasks dispatching device of a kind of Time Triggered and method | |
| Ramamritham | Where do time constraints come from? Where do they go? | |
| CN110580020A (en) | AGV (automatic guided vehicle) scheduling method and device, computer equipment and storage medium | |
| Maimon | Real-time operational control of flexible manufacturing systems | |
| CN109684057A (en) | Task processing method, device and storage medium | |
| CN107566460B (en) | Method and system for distributed deployment of planning tasks | |
| CN111105155B (en) | Role-based group unmanned system collaborative task management method | |
| CN102176122B (en) | Automatic control system based on finite automaton and control method thereof | |
| CN112579267A (en) | Decentralized big data job flow scheduling method and device | |
| CN110270998A (en) | Robot dispatching method, electronic equipment and computer readable storage medium | |
| CN111240935A (en) | Automatic intelligent operation and maintenance system and operation and maintenance method | |
| CN104635506B (en) | A kind of semiconductor production line handling system device simulating method | |
| CN113687927A (en) | Method, device, equipment and storage medium for scheduling and configuring flash tasks | |
| CN117149378A (en) | Task scheduling method, device, equipment and medium of intelligent automobile operating system | |
| Lin et al. | An agent-based flexible routing manufacturing control simulation system | |
| CN115793496A (en) | Dispatching simulation method and system for AGV logistics system | |
| CN105227648A (en) | A kind of subway comprehensive monitoring system time sequence triggering method based on time bucket | |
| CN105677515B (en) | A kind of Online Database Backup method and system | |
| CN115056234A (en) | RPA controller scheduling method and system based on event driving and infinite state machine | |
| Ramamritham | The Origin of TCs. | |
| Pimentel et al. | Implicit priorities in adaptation requirements |
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 |