CN102909726B - Behavior realizing method for service robot - Google Patents
Behavior realizing method for service robot Download PDFInfo
- Publication number
- CN102909726B CN102909726B CN201210384568.9A CN201210384568A CN102909726B CN 102909726 B CN102909726 B CN 102909726B CN 201210384568 A CN201210384568 A CN 201210384568A CN 102909726 B CN102909726 B CN 102909726B
- Authority
- CN
- China
- Prior art keywords
- action
- behavior
- service robot
- joint
- fragment
- 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.)
- Expired - Fee Related
Links
Abstract
The invention discloses a behavior realizing method for a service robot. A behavior comprises at least one action, and the action comprises at least one action segment. The method comprises the steps of environment and intention analysis and action realization and correction. In the environment and intention analysis step, an environmental structure of the service robot is analyzed, an intention of a commander is comprehended, a behavior object is determined, and the behavior actions are designed, that is, a behavior of the service robot is broken up into a series of actions arranged in sequence, and each action is embodied to a set of action segments arranged in sequence. In the action realization and correction step, each action segment is realized by using a sensor and mechanical drive components of the service robot, and each action and each behavior are further accomplished. According to the behavior realizing method, a certain quantity of action segments are defined, the action segments are combined together according to behavior requirements, continuous switching of action states is realized by adopting a connection and smoothing technique, and finally the environment and restriction adaptability is realized through a dynamic behavior management mechanism under environmental restriction.
Description
Technical field
The invention belongs to robotics, particularly a kind of service robot behavior implementation method.
Background technology
Service robot is the one of robot, has huge using value as a kind of autonomous robot.Mainly be engaged in the tasks such as maintaining, individual high-end business, security guard patrol, rescue monitoring.Therefore the capacity for service robot has high requirement.In addition, because the working environment of service robot and condition restriction have stronger uncertainty usually, so all there is higher requirement for the adaptive capacity to environment of service robot and constraint adaptive capacity.
In order to realize above-mentioned requirement, the usual more complicated of the action behavior that service robot must have, in order to realize its prerequisite everything, the movements design for service robot also becomes comparatively complicated.
Traditionally, the behavior of service robot is usually only according to the setting that designing requirement is carried out in advance, so the capacity of robot is subject to great constraint, the application outside design object is just difficult to the demand meeting user or application.And will be very complicated and difficult for arranging of new behavior, and robot lacks and effectively imitates mechanism in behavior, cannot realize the action imitation of whole limbs or part of limb.
Summary of the invention
The object of this invention is to provide a kind of service robot behavior implementation method, to solve the problem that service robot in prior art cannot meet complex behavior setting and realize.
The present invention is based upon on the combination foundation of action fragment, realizes behavior management and the control of all kinds of service robot.
All kinds of basic action that the present invention utilizes each limbs of robot to complete, forms the ensemble space of action.On the basis in set of actions space, select the simple action of some and type and utilize space vector and time vector to carry out sequence formation action fragment to action.For determining that the space vector that action is sorted and time vector have relativeness, space vector and time vector are mapped in actual real world, as the basis that behavior realizes by formulation reference frame in the application of reality.
The behavior of robot in the present invention refers to complete to have certain practical significance, meets the set that people determine the set of demand, arrives somewhere point, article etc. of taking as advanced.The behavior of robot can by the sub-line of some for forming in the present invention.Behavior is the set of actions with definite real implication.
So-called action fragment refers to the most basic element forming robot behavior in the present invention, cannot further be decomposed into significant behavior component.
Technical scheme of the present invention is, a kind of service robot behavior implementation method, and described behavior is made up of at least one action, and described action is made up of at least one action fragment, comprises the following steps:
Environment with intention analyze, the environmental structure namely described in Analysis Service robot, understands directorial intention, determines object of action, the behavior object be at least one;
Behavior act designs, and is a series of actions of arranged in sequence, and further each action is embodied as the set of the action fragment of multiple arranged in sequence by described service robot behavior decomposition;
Action realizes and corrects, and namely utilizes the sensor of service robot and each action fragment described in mechanical driving member realization, and completes each action and behavior further.
Further, seamlessly transit successive described service robot limbs joint attitude vectors sign each action fragment described with a series of, this joint attitude vectors is performed by the mechanical driving member identification of service robot.
Further, described joint attitude vectors comprises relative space position coordinate parameters and attitude angle parameter, attitude angle parameter is the angle rotated around privately owned coordinate system three axles, characterize pitching, driftage and roll respectively, the coordinate system of described joint attitude vectors initiatively determined in joint by its higher level, and namely the privately owned coordinate of driven joint is its main diarthrodial relative coordinate system.
Further, the joint attitude vectors described in mechanical driving member execution of described service robot is divided into the stage of entering, execution phase and exits transition stage, enters the stage and exits transition stage for being connected with front and back action fragment, wherein,
The stage that enters is by the done state of a upper action fragment, under the initial state of execution phase and the constraint of external condition, the transfer step by step of completion status, until be transferred to the beginning of execution phase, each exits transition state and determine next step shift direction under the constraint of input and external condition
Exit transition stage, the done state of execution phase is considered as input, under the state constraint of the beginning of next action fragment, completes transfer.
The present invention is a kind of behavior management technology of service robot vector action segment condense, the combination of some for the Behavioral availability of service robot action fragments is represented.These action fragments are described in the mode of vector quantization, and are connected by action fragment and smoothly interconnect.Service robot is condition environmentally, utilizes the combining form of behavior management mechanism dynamic conditioning action fragment to realize the adaptability of environment.The combination of the some action fragments of service robot Behavioral availability is represented, the combination of the behavior of a certain for robot specific purposes by some action fragments is expressed.Each action fragment is relatively independent, by by different fragment combination to produce dissimilar robot behavior.Action fragment adopts conversion and the switching of the vector form Describing Motion state of direction and relative amplitude.The direction of action is relevant with relative vector space with amplitude, can make robot under various different attitude condition, accurately complete the enforcement of all kinds of action.The description of action fragment be a small collections of action, comprising elemental motion mode, possible state transition path and error margin.Then utilize the difference between current state and next dbjective state to carry out state transfer when being connected between action, insert the smoothness properties that some level and smooth fragments realize transfer if desired.The environment of the present invention residing for service robot and constraints, the combination way of realization of the action fragment in a dynamic adjustment behavior, to adapt to the constraint of varying environment and condition.
The invention provides the theory and technology of a kind of service robot behavior management and control.By defining the action fragment of some, and according to the requirement of behavior, these action fragments are combined the brand-new behavior pattern of formation.In the implementation, action fragment, by a series of characterization vector, utilizes linking and smoothing technique to realize the continuous switching of operating state, under the constraint of environment restriction, finally utilize the dynamic management of behavior mechanism to realize the adaptability of environment and constraint.
Accompanying drawing explanation
Fig. 1 is that in one embodiment of the invention, service robot behavior realizes schematic diagram
Detailed description of the invention
As the behavior making service robot complete " pouring "; in the specific implementation of the technology protected at this patent; first be by factor determination operation objects such as the intentions of analysis environments structure and people; and then the behavior decomposition of pouring is become the combination of a series of actions fragment, finally utilize sensor and Mechanical Driven part to be realized.Therefore the technology protected of this patent can be divided into three levels to be also three steps of main process in the process realized simultaneously, and namely environment and intention are analyzed, behavior act design and action realizes and correction.
After the instruction receiving " pouring ", first system determines the object of action key element of this instruction, i.e. water source and container.For household or office environment, the object at water source and position are normally determined with known, as water dispenser, and water heater or thermos flask etc.And container can be pointed out by vision sensor or directly by effector, as teacup, dixie cup etc.This completes basic environment and intention analysis.
In the enterprising every trade in basis of a upper link be and movements design.It may be noted that in the technology, to think that behavior is that action is formed, action is made up of a series of actions fragment, as described below.
Behavior comprises: extracting container, march to water source, open water source water intaking, weigh vessel level height, close water source in good time and return destination or starting point;
Action comprises: holding container, direct of travel and distance, to open and liquid level monitors, return path;
Action fragment comprises: holding mode, dynamics feedback, strategy of advancing, open and close water source mode, monitoring picture process, return strategy.
The structure of stratification shown in above-mentioned is actual is that action is in the process being embodied as action fragment by behavior decomposition exactly.
Robot system according to instruction determination behavior, then is decomposed into a series of actions according to behavior, finally each action is embodied as the fragment of action.As shown in Figure 1.
After the instruction knowing pouring, first form behavior according to action effective object, namely need action to implement to each link and form, extracting container, advance to water source, open, monitor liquid level, close, return.In action, adopt different behaviors according to the difference of object, as teacup and dixie cup, the mode of gripping will there are differences, different water sources, and the method for operating as tap and water dispenser will be different.And the uncertainty of environment also can make advancing to water source and returning and is full of obstacle and uncertainty.
Because behavior all has clear and definite real implication, the relational implementation sub-line between semantic description behavior therefore can be adopted to be combined into senior behavior pattern.Joining relation is presented, if as precedence between the description of semantization refers to the form lines of description of word and is ... so .... etc.
Action fragment changes expression by the vector in joint.The current pose of a turning joint of robot can be made up of its position coordinates and attitude angle.Position coordinates adopts relative coordinate to represent that the position coordinates namely under the privately owned coordinate system in this joint is made up of xyz tri-parameters in space, and attitude angle is then the angle rotated around privately owned coordinate system three axles, characterizes pitching, driftage and roll respectively.Therefore the attitude in joint is determined by 6 parameters, can be expressed as a joint attitude vectors:
The coordinate system in joint initiatively determined in joint by its higher level.Namely the privately owned coordinate of driven joint is its main diarthrodial relative coordinate system.Under the privately owned coordinate system in joint, attitude vectors adopts relative coordinate system to represent, the maximum of joint in each dimension is decided to be 1, sets up relative coordinate system and attitude vectors with this.
The joint attitude vectors that action fragment is joining relation by a series of front and back is represented, the implementation of composition action.These successive joint attitude vectors are divided into three parts, are respectively to enter stage, execution phase, exit transition stage.
Wherein enter the stage and exit transition stage for being connected with other action fragments, also taking into account the mode realizing action under different constraints simultaneously.These two stages adopt the form of Markov chain to be represented.
The main effect of approach section is by the done state of a upper action fragment, and under the execution initial state of section and the constraint of external condition, the transfer step by step of completion status, until be transferred to the beginning performing section.Each transition state determines next step shift direction under the constraint of input and external condition.
Exit changeover portion more similar on 26S Proteasome Structure and Function with approach section, certain Chengdu can be thought the done state performing section is considered as input, under the state constraint of the beginning of next action fragment, complete transfer.
The state transition path that the adaptability of environment is just embodied under constraints is selected.
Claims (1)
1. a service robot behavior implementation method, described behavior is made up of at least one action, and described action is made up of at least one action fragment, it is characterized in that, comprises the following steps:
Environment with intention analyze, the environmental structure namely described in Analysis Service robot, understands directorial intention, determines object of action, the behavior object be at least one;
Behavior act designs, and is a series of actions of arranged in sequence, and further each action is embodied as the set of the action fragment of multiple arranged in sequence by described service robot behavior decomposition;
Action realizes and corrects, and namely utilizes the sensor of service robot and each action fragment described in mechanical driving member realization, and completes each action and behavior further;
Wherein, seamlessly transit successive described service robot limbs joint attitude vectors sign each action fragment described with a series of, this joint attitude vectors is performed by the mechanical driving member identification of service robot;
Described joint attitude vectors comprises relative space position coordinate parameters and attitude angle parameter, attitude angle parameter is the angle rotated around privately owned coordinate system three axles, characterize pitching, driftage and roll respectively, the coordinate system of described joint attitude vectors initiatively determined in joint by its higher level, and namely the privately owned coordinate of driven joint is its main diarthrodial relative coordinate system;
The attitude in joint is determined by 6 parameters, is expressed as a joint attitude vectors:
Under the privately owned coordinate system in joint, attitude vectors adopts relative coordinate system to represent, the maximum of joint in each dimension is decided to be 1, sets up relative coordinate system and attitude vectors with this;
The joint attitude vectors described in mechanical driving member execution of described service robot is divided into the stage of entering, execution phase and exits transition stage, enters the stage and exits transition stage for being connected with front and back action fragment, wherein,
The stage that enters is by the done state of a upper action fragment, under the initial state of execution phase and the constraint of external condition, the transfer step by step of completion status, until be transferred to the beginning of execution phase, each exits transition state and determine next step shift direction under the constraint of input and external condition
Exit transition stage, the done state of execution phase is considered as input, under the state constraint of the beginning of next action fragment, completes transfer.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210384568.9A CN102909726B (en) | 2012-10-11 | 2012-10-11 | Behavior realizing method for service robot |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210384568.9A CN102909726B (en) | 2012-10-11 | 2012-10-11 | Behavior realizing method for service robot |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102909726A CN102909726A (en) | 2013-02-06 |
CN102909726B true CN102909726B (en) | 2015-01-28 |
Family
ID=47608413
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210384568.9A Expired - Fee Related CN102909726B (en) | 2012-10-11 | 2012-10-11 | Behavior realizing method for service robot |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102909726B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPWO2017199565A1 (en) * | 2016-05-20 | 2019-01-10 | シャープ株式会社 | Robot, robot operation method, and program |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001300876A (en) * | 2000-04-20 | 2001-10-30 | Yamatake Corp | Service robot and service system using it |
JP4394602B2 (en) * | 2005-04-20 | 2010-01-06 | 富士通株式会社 | Service robot |
CN100348381C (en) * | 2006-01-06 | 2007-11-14 | 华南理工大学 | Housekeeping service robot |
KR20080090150A (en) * | 2007-04-04 | 2008-10-08 | 삼성전자주식회사 | Service robot, service system using service robot and controlling method of the service system using service robot |
CN101084817B (en) * | 2007-04-26 | 2012-08-22 | 复旦大学 | Opening intelligent calculation frame household multifunctional small-sized service robot |
CN101549498B (en) * | 2009-04-23 | 2010-12-29 | 上海交通大学 | Automatic tracking and navigation system of intelligent aid type walking robots |
-
2012
- 2012-10-11 CN CN201210384568.9A patent/CN102909726B/en not_active Expired - Fee Related
Also Published As
Publication number | Publication date |
---|---|
CN102909726A (en) | 2013-02-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Lu et al. | Adaptive control of time delay teleoperation system with uncertain dynamics | |
Zhang et al. | A data-and knowledge-driven framework for digital twin manufacturing cell | |
Dimarogonas et al. | A connection between formation infeasibility and velocity alignment in kinematic multi-agent systems | |
Kohrt et al. | An online robot trajectory planning and programming support system for industrial use | |
Barenji et al. | A digital twin-driven approach towards smart manufacturing: reduced energy consumption for a robotic cellular | |
Su et al. | Task-independent robotic uncalibrated hand-eye coordination based on the extended state observer | |
CN102708377B (en) | Method for planning combined tasks for virtual human | |
Dean-Leon et al. | Robotic technologies for fast deployment of industrial robot systems | |
CN104834308A (en) | Optimal itineration control method satisfying complex requirement | |
Merkt et al. | Continuous-time collision avoidance for trajectory optimization in dynamic environments | |
Chen et al. | Time delay prediction for space telerobot system with a modified sparse multivariate linear regression method | |
Ruchanurucks et al. | Humanoid robot motion generation with sequential physical constraints | |
Blanchini et al. | A convex programming approach to the inverse kinematics problem for manipulators under constraints | |
Raessa et al. | Teaching a robot to use electric tools with regrasp planning | |
Pavlichenko et al. | Autonomous dual-arm manipulation of familiar objects | |
CN102909726B (en) | Behavior realizing method for service robot | |
Tarbouriech et al. | An admittance based hierarchical control framework for dual-arm cobots | |
Zaborovsky et al. | Cyber-physical approach in a series of space experiments “kontur” | |
Banerjee et al. | A survey on physics informed reinforcement learning: Review and open problems | |
Maithripala et al. | A geometric virtual structure approach to decentralized formation control | |
Kohlbrecher et al. | A comprehensive software framework for complex locomotion and manipulation tasks applicable to different types of humanoid robots | |
Baxi et al. | Towards factory-scale edge robotic systems: Challenges and research directions | |
Alami et al. | Reasoning about humans and its use in a cognitive control architecture for a collaborative robot | |
Zaborovsky et al. | Cyber-physical approach to the network-centric robot control problems | |
Schierl | Object-oriented modeling and coordination of mobile robots |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150128 Termination date: 20211011 |
|
CF01 | Termination of patent right due to non-payment of annual fee |