+

WO2018171019A1 - Système et procédé de commande et d'entraînement d'optimisation multivariable pouvant s'adapter de manière intelligente à une surface - Google Patents

Système et procédé de commande et d'entraînement d'optimisation multivariable pouvant s'adapter de manière intelligente à une surface Download PDF

Info

Publication number
WO2018171019A1
WO2018171019A1 PCT/CN2017/084095 CN2017084095W WO2018171019A1 WO 2018171019 A1 WO2018171019 A1 WO 2018171019A1 CN 2017084095 W CN2017084095 W CN 2017084095W WO 2018171019 A1 WO2018171019 A1 WO 2018171019A1
Authority
WO
WIPO (PCT)
Prior art keywords
module
driving
user
optimization
value group
Prior art date
Application number
PCT/CN2017/084095
Other languages
English (en)
Chinese (zh)
Inventor
辛志宇
闵苏
叶鹏
Original Assignee
魔玛智能科技（上海）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 魔玛智能科技（上海）有限公司 filed Critical 魔玛智能科技（上海）有限公司
Publication of WO2018171019A1 publication Critical patent/WO2018171019A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers

Definitions

  • the present invention relates to the field of computer and intelligent system control technologies, and in particular, to a multi-variable optimized drive control system and method for intelligently accommodating surfaces.
  • a self-adjusting surface support system it is often necessary to drive the system to simultaneously make the same or different changes and adjustments to multiple points and multiple areas of the surface.
  • these multi-variable driver implementations often need to be coordinated or completed simultaneously or simultaneously.
  • the driver system also needs a relatively flexible optimization strategy to implement the resource allocation scheme to meet the implementation requirements of different tasks for the user's individual needs and the real-time changing environment state.
  • the above task requirements could not be achieved.
  • the invention realizes the above multi-variable cooperation and achieves the requirement of optimal driving execution effect through a distributed modular optimization driving system and method combining software and hardware.
  • the multi-variable optimized driving control system for intelligent adaptive surface comprises: a data calling module, a data access module, a user behavior and state pattern recognition module, a user environment adaptation decision module, a multivariable optimization solving module, and a distributed Drive subsystem control module and distributed multiple drive subsystem modules, wherein:
  • the data invoking module is configured to transmit the contact object duration and real-time behavior and status tag data retrieved from the data access module to the user behavior and state pattern recognition module and the user environment adaptation decision module;
  • a data access module for accessing the diachronic and real-time behavior and status tag data of the contact object
  • the user behavior and state pattern recognition module is configured to compare the acquired real-time and chronological contact object behavior and state tag data with the pattern category features in the database, and perform pattern recognition and classification marking on the current user mode category, and the current user mode
  • the category value is written into the data access module
  • the user environment adaptation decision module is configured to obtain a current user mode category from the data access module, and retrieve a user environment adaptation target value group corresponding to the current user mode category from the data access module, and output the target value group to at most
  • the variable optimization solving module obtains the returned driving target value group and outputs to the distributed driving subsystem control module;
  • the multi-variable optimization solving module is configured to generate a driving strategy for the current user mode category, parse, optimize and correct the user environment adaptive target value group according to the driving strategy, and output the driving target value group to the user environment adaptive decision module;
  • the distributed driving subsystem control module is configured to receive a driving target value group that is adapted by the user environment to the decision module output, generate a corresponding driving value group and/or a task command, and output to the corresponding driving subsystem module for cooperative driving execution to complete the target. Task, and returning an operation value, the operation value is saved as a device driver record in the data access module;
  • a plurality of distributed drive subsystem modules for performing drive tasks for supporting surface adjustment.
  • the data access module comprises: a data temporary storage module and a database, wherein: the data temporary storage module stores the current user mode category value, and the database stores the duration, real-time behavior and status tag data of the contact object.
  • the contact object comprises: a partial or all body area where the user lies, sits, and is in contact with the support surface.
  • the multivariate optimization solution module comprises: a user customized optimization strategy module, a user environment adaptation optimization strategy module, a driving resource optimization strategy module, and a target task solving decision module, specifically:
  • the user customization optimization policy module is configured to set a user environment adaptation optimization strategy and a driving resource optimization strategy scheme corresponding to the current user mode category under different user personalized requirement conditions;
  • different user personalization strategies include:
  • the light sleep stage uses a comfort priority strategy and/or a non-perceive adaptation priority strategy
  • attitude-priority strategy and/or the non-perceive adaptation priority strategy are adopted in the deep sleep phase;
  • an uncomfortable strategy and/or an immediate adaptation priority strategy is employed during the sleep awakening period;
  • a gesture priority strategy is adopted for all sleep stages
  • the user environment adaptation optimization policy module is configured to set different adjustment strategies, and a priority relationship of target value adjustment between driving variables under the policy.
  • the attitude priority strategy the strategy of adjusting the attitude (height deformation) variable to the target value priority
  • Comfort priority strategy the strategy of adjusting the pressure variable to the target value priority
  • the global pressure balance strategy a strategy of balancing the pressure variables of various parts of the body to the target value
  • the driving resource optimization policy module is configured to set a calling scheme of different driving resources under different driving resource optimization strategies; specifically, for example:
  • the target task solving decision module is configured to accept a user environment adaptation target value group output by the user environment adaptation decision module, and generate the condition constraint constraint of the user customization optimization strategy module, the user environment adaptation optimization strategy module, and the driving resource optimization strategy module.
  • the driving target value group adapted to the driving operation is output to the driving subsystem control module.
  • the distributed drive subsystem control module comprises: an instruction accepting and communication module, a driver execution module,
  • the instruction accepting and communicating module is configured to accept a driving value group and a driving task command output by the driving execution module, and output a driving value group and/or a task command to the corresponding driving subsystem module;
  • the driver execution module is configured to accept the driving target value group output by the multivariate optimization solving module, generate a driving value group and The task command is driven to the instruction accept and communication module, and the return operation value is saved in the data access module.
  • the multi-variable optimization driving control method for intelligent adaptive surface comprises the following steps:
  • Data invocation step transferring the contact object duration and real-time behavior and status tag data retrieved from the data access module to the user behavior and state pattern recognition module and the user environment adaptation decision module;
  • Data access step accessing the duration and real-time behavior and status tag data of the contact object
  • User behavior and status pattern recognition step compare the acquired real-time and duration contact object behavior and status tag data with the pattern category features in the database, and perform pattern recognition and classification marking on the current user mode category, and the current user mode category value Write to the data access module;
  • the user environment adapts to the decision step: obtaining the current user mode category from the data access module, and calling the user environment adaptation target value group corresponding to the current user mode category from the data access module, and outputting the target value group to multivariate optimization
  • the solving module obtains the returned driving target value group and outputs to the distributed driving subsystem control module;
  • the multi-variable optimization solving step generating a driving strategy for the current user mode category, parsing, optimizing and correcting the user environment adaptation target value group according to the driving strategy, and outputting the driving target value group to the user environment adaptation decision module;
  • the distributed driving subsystem control step receiving the driving target value group output by the user environment adaptation decision module, generating a corresponding driving value group and/or task command, outputting to the corresponding driving subsystem module for cooperative driving execution, and completing the target task, And returning an operation value, the operation value is saved as a device driver record in the data access module;
  • Distributed multiple drive subsystem steps Perform drive tasks to support surface adjustments.
  • the present invention has the following beneficial effects:
  • the intelligent adaptive surface multi-variable optimization driving control system and method provided by the invention are particularly suitable for optimal control of multi-target tasks of multiple driving systems, and can cooperatively perform complex surface motion changes under the control of the global optimization strategy. Good user experience. And it can meet the user's personalized customized priority strategy and calculation method.
  • the invention changes from a simple target value drive to a multi-variable optimization drive that considers optimization strategies such as user customization, multiple environment adaptation and drive resources, and improves the user experience by data-driven and artificial intelligence.
  • the distributed modular multiple drive subsystems of the present invention are capable of cooperatively performing complex surface change actions under a global optimization strategy control system.
  • FIG. 1 is a schematic diagram of a multi-variable optimized drive control system for an intelligent adaptive surface provided in the present invention.
  • the multi-variable optimized driving control system for intelligent adaptive surface comprises: a data calling module, a data access module, a user behavior and state pattern recognition module, a user environment adaptation decision module, a multivariable optimization solving module, and a distributed Drive subsystem control module and distributed multiple drive subsystem modules, wherein:
  • the data invoking module is configured to transmit the contact object duration and real-time behavior and status tag data retrieved from the data access module to the user behavior and state pattern recognition module and the user environment adaptation decision module;
  • a data access module for accessing the diachronic and real-time behavior and status tag data of the contact object
  • the user behavior and state pattern recognition module is configured to compare the acquired real-time and chronological contact object behavior and state tag data with the pattern category features in the database, and perform pattern recognition and classification marking on the current user mode category, and the current user mode
  • the category value is written into the data access module
  • the user environment adaptation decision module is configured to obtain a current user mode category from the data access module, and retrieve a user environment adaptation target value group corresponding to the current user mode category from the data access module, and output the target value group to at most
  • the variable optimization solving module obtains the returned driving target value group and outputs to the distributed driving subsystem control module;
  • the multi-variable optimization solving module is configured to generate a driving strategy for the current user mode category, parse, optimize and correct the user environment adaptive target value group according to the driving strategy, and output the driving target value group to the user environment adaptive decision module;
  • the distributed driving subsystem control module is configured to receive a driving target value group that is adapted by the user environment to the decision module output, generate a corresponding driving value group and/or a task command, and output to the corresponding driving subsystem module for cooperative driving execution to complete the target. Task, and returning an operation value, the operation value is saved as a device driver record in the data access module;
  • a plurality of distributed drive subsystem modules for performing drive tasks for supporting surface adjustment.
  • the data access module includes: a data temporary storage module and a database, wherein: the data temporary storage module stores a current user mode category value, and the database stores the duration, real-time behavior, and status tag data of the contact object.
  • the contact object includes a partial or total body area where the user lies, sits, and is in contact with the support surface.
  • the multivariate optimization solution module includes: a user customized optimization strategy module, and a user environment adaptation optimization strategy module Block, drive resource optimization strategy module, target task solution decision module, specifically:
  • the user customization optimization policy module is configured to set a user environment adaptation optimization strategy and a driving resource optimization strategy scheme corresponding to the current user mode category under different user personalized requirement conditions;
  • different user personalization strategies include:
  • the light sleep stage uses a comfort priority strategy and/or a non-perceive adaptation priority strategy
  • attitude-priority strategy and/or the non-perceive adaptation priority strategy are adopted in the deep sleep phase;
  • an uncomfortable strategy and/or an immediate adaptation priority strategy is employed during the sleep awakening period;
  • a gesture priority strategy is adopted for all sleep stages
  • the user environment adaptation optimization policy module is configured to set different adjustment strategies, and a priority relationship of target value adjustment between driving variables under the policy;
  • Attitude priority strategy a strategy of adjusting the attitude (height deformation) variable to the target value priority
  • Comfort priority strategy a strategy that prioritizes pressure variables to target values
  • a global stress equalization strategy that balances the pressure variables of various parts of the body to the target value
  • the driving resource optimization policy module is configured to set a calling scheme of different driving resources under different driving resource optimization strategies; specifically, for example:
  • the target task solving decision module is configured to accept a user environment adaptation target value group output by the user environment adaptation decision module, and generate the condition constraint constraint of the user customization optimization strategy module, the user environment adaptation optimization strategy module, and the driving resource optimization strategy module.
  • the driving target value group adapted to the driving operation is output to the driving subsystem control module.
  • the distributed drive subsystem control module includes: an instruction accepting and communication module, and a driver execution module,
  • the instruction accepting and communicating module is configured to accept a driving value group and a driving task command output by the driving execution module, and output a driving value group and/or a task command to the corresponding driving subsystem module;
  • the driver execution module is configured to accept the driving target value group output by the multivariate optimization solving module, generate a driving value group and drive the task command to the instruction accepting and communication module, and return the operation value to be saved in the data access module.
  • the multi-variable optimization driving control method for intelligent adaptive surface comprises the following steps:
  • Data invocation step transferring the contact object duration and real-time behavior and status tag data retrieved from the data access module to the user behavior and state pattern recognition module and the user environment adaptation decision module;
  • Data access step accessing the duration and real-time behavior and status tag data of the contact object
  • User behavior and status pattern recognition step compare the acquired real-time and duration contact object behavior and status tag data with the pattern category features in the database, and perform pattern recognition and classification marking on the current user mode category, and the current user mode category value Write to the data access module;
  • the user environment adapts to the decision step: obtaining the current user mode category from the data access module, and calling the user environment adaptation target value group corresponding to the current user mode category from the data access module, and outputting the target value group to multivariate optimization
  • the solving module obtains the returned driving target value group and outputs to the distributed driving subsystem control module;
  • the multi-variable optimization solving step generating a driving strategy for the current user mode category, parsing, optimizing and correcting the user environment adaptation target value group according to the driving strategy, and outputting the driving target value group to the user environment adaptation decision module;
  • the distributed driving subsystem control step receiving the driving target value group output by the user environment adaptation decision module, generating a corresponding driving value group and/or task command, outputting to the corresponding driving subsystem module for cooperative driving execution, and completing the target task, And returning an operation value, the operation value is saved as a device driver record in the data access module;
  • Distributed multiple drive subsystem steps Perform drive tasks to support surface adjustments.
  • the steps in the multi-variable optimization driving control method of the intelligent adaptive surface provided by the present invention may be implemented by using corresponding modules, devices, units, etc. in the multi-variable optimization driving control system of the intelligent adaptive surface.
  • the steps of the method can be implemented by a person skilled in the art with reference to the technical solution of the system. That is, the embodiment in the system can be understood as a preferred example of implementing the method, and details are not described herein.
  • the method of the present invention is applied to a support surface control in which the distributed drive adapts to the user's sleep behavior and state.
  • the process of completing the deformation drive of the support surface is also very important for the user experience.
  • a set of optimization strategies are needed to control and manage the driving process. For example, for different weight body parts, how to naturally complete the adaptation action at the same time is very important for the user experience.
  • the database module is invoked to obtain the duration and immediate behavior and status tag data of the contact object, and the user behavior and state mode evaluation module and the user environment adaptation mode evaluation module are input.
  • the user behavior and state pattern recognition module compares the pre-entered pattern category features according to the acquired instant and diachronic contact object behavior and state tag data, and determines the current user mode category according to the sleep cycle and phase characteristics, the body motion behavior characteristics, and the posture.
  • the state feature and the like perform pattern recognition and classification marking, and write the current user's sleep cycle/stage and posture state mode category values into the data access module-data temporary storage module.
  • the user environment adaptation decision module obtains the current user's sleep cycle/stage, posture state mode category value from the data access module-data temporary storage module, and calls the user environment adaptation target value group corresponding to the mode category from the database module. Then, the user customization optimization strategy module, the user environment adaptation optimization strategy module, and the driving resource optimization strategy module in the multivariate optimization solution module are invoked to evaluate the current user mode and the environmental state, for example, user comfort priority is adopted during the sleep phase.
  • the strategy adopts the user attitude priority strategy in the deep sleep stage; according to the driving resource optimization strategy, the global driver is optimized to obtain a natural coordinated driving implementation process suitable for the user experience.
  • the target task solving decision module is called to optimize and correct the target value group.
  • the drive target value is output to the drive subsystem control module according to the corrected drive target value. .
  • the driving subsystem control module generates a driving value group and a driving task command according to the received driving target value group, performs cooperative driving execution on the distributed multiple driving subsystem modules, completes the target task, and returns the operation value to the driving subsystem control.
  • the module is written by the module as a device driver record to the data storage module - the database module.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)
  • Traffic Control Systems (AREA)

Abstract

L'invention concerne un système et un procédé de commande et d'entraînement d'optimisation multivariable pouvant s'adapter de manière intelligente à une surface. Le système comprend : un module d'appel de données; un module d'accès aux données; un module de reconnaissance de mode d'état et de comportement d'utilisateur; un module de décision d'adaptation d'environnement d'utilisateur; un module de calcul d'optimisation multivariable; un module de commande de sous-système d'entraînement de type distribué; et de multiples modules de sous-système d'entraînement de type distribué. Le système et le procédé sont applicables à la commande d'optimisation de tâches multi-cibles d'une pluralité de systèmes d'entraînement et permettent de modifier de manière coopérative des actions sur des surfaces complexes sous la commande d'une stratégie d'optimisation globale, disposent d'une bonne expérience utilisateur, et peuvent satisfaire un procédé de calcul et une stratégie de préférence sur mesure et personnalisée d'un utilisateur. L'entraînement suivant une valeur cible simple est remplacé par l'entraînement d'optimisation multivariable en tenant compte de stratégies d'optimisation en vue d'une personnalisation sur mesure d'un utilisateur, d'une adaptation multi-environnement et de recours d'entraînement, de telle sorte que l'expérience utilisateur est améliorée au moyen d'une commande de données et d'une intelligence artificielle, et des actions de changement sur des surfaces complexes peuvent être exécutées de manière coopérative à l'aide d'un système de commande de la stratégie d'optimisation globale.
PCT/CN2017/084095 2017-03-20 2017-05-12 Système et procédé de commande et d'entraînement d'optimisation multivariable pouvant s'adapter de manière intelligente à une surface WO2018171019A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710166523.7A CN107168111B (zh) 2017-03-20 2017-03-20 智能适应表面的多变量优化驱动控制系统及方法
CN201710166523.7 2017-03-20

Publications (1)

Publication Number Publication Date
WO2018171019A1 true WO2018171019A1 (fr) 2018-09-27

Family

ID=59849306

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/084095 WO2018171019A1 (fr) 2017-03-20 2017-05-12 Système et procédé de commande et d'entraînement d'optimisation multivariable pouvant s'adapter de manière intelligente à une surface

Country Status (2)

Country Link
CN (1) CN107168111B (fr)
WO (1) WO2018171019A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102298346A (zh) * 2011-05-26 2011-12-28 江苏科技大学 一种智能轮椅语音驱动控制器及识别与控制方法
US20120255121A1 (en) * 2011-04-11 2012-10-11 Receveur Timothy J Low noise linear diaphragm compressor by variable amplitude driver
CN103780691A (zh) * 2014-01-20 2014-05-07 辛志宇 智慧睡眠系统及其用户端系统和云端系统
CN203838535U (zh) * 2014-01-20 2014-09-17 辛志宇 灵活布置更换的矩阵模块化睡眠数据采集和环境动作系统
CN106959636A (zh) * 2017-03-20 2017-07-18 魔玛智能科技（上海）有限公司 智能适应支撑表面的模块化分布式驱动控制系统及方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN201097174Y (zh) * 2007-10-29 2008-08-06 吴远宁 医疗保健床垫电气控制器
CN102436246A (zh) * 2011-12-19 2012-05-02 厦门万安智能股份有限公司 具有环境适应情景模式的智能家居集控装置
US9186479B1 (en) * 2014-06-05 2015-11-17 Morphy Inc. Methods and systems for gathering human biological signals and controlling a bed device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120255121A1 (en) * 2011-04-11 2012-10-11 Receveur Timothy J Low noise linear diaphragm compressor by variable amplitude driver
CN102298346A (zh) * 2011-05-26 2011-12-28 江苏科技大学 一种智能轮椅语音驱动控制器及识别与控制方法
CN103780691A (zh) * 2014-01-20 2014-05-07 辛志宇 智慧睡眠系统及其用户端系统和云端系统
CN203838535U (zh) * 2014-01-20 2014-09-17 辛志宇 灵活布置更换的矩阵模块化睡眠数据采集和环境动作系统
CN106959636A (zh) * 2017-03-20 2017-07-18 魔玛智能科技（上海）有限公司 智能适应支撑表面的模块化分布式驱动控制系统及方法

Also Published As

Publication number Publication date
CN107168111B (zh) 2019-03-08
CN107168111A (zh) 2017-09-15

Similar Documents

Publication Publication Date Title
US11783227B2 (en) Method, apparatus, device and readable medium for transfer learning in machine learning
US20210142143A1 (en) Artificial intelligence systems and methods
US20200143291A1 (en) Hierarchical Clustered Reinforcement Machine Learning
WO2018171020A1 (fr) Système et procédé de commande de pilotage distribuée modulaire de surface de support adaptative intelligente
CN107003929A (zh) 用于异构核设备上的机器学习的应用特征化
GB2624839A (en) Creating multiple prioritized clinical summaries using artificial intelligence
JP7520123B2 (ja) 人間－ロボット混在製造プロセスにおける自動異常検出のためのシステムおよび方法
Hoffmann et al. Seec: A framework for self-aware computing
CN108984257A (zh) 一种支持自定义算法组件的机器学习平台
KR20230108293A (ko) 입력 값 처리 방법 및 시스템
CN104778254B (zh) 一种非参数的话题自动标注的分布式系统和标注方法
WO2018171019A1 (fr) Système et procédé de commande et d'entraînement d'optimisation multivariable pouvant s'adapter de manière intelligente à une surface
CN119396497B (zh) 基于云端大模型增强的边缘ai协同方法及系统
US11369290B2 (en) Assist device control system
CN118136222B (zh) 一种面向医疗环境的数字化智能制氧系统
CN109471520A (zh) 智能升降桌的调整方法、装置及存储介质和智能升降桌
CN109301820A (zh) 一种企业电力控制方法及系统
CN117114137A (zh) 一种自行闭环优化的嵌入式算法模型系统
US20220129071A1 (en) Systems and Methods for Nonlinear Latent Spatiotemporal Representation Alignment Decoding for Brain-Computer Interfaces
WO2023015500A1 (fr) Calcul hétérogène à modèles multiples
US11557898B2 (en) Method and devices for processing sensor data by applying one or more processing pipelines to the sensor data
CN113796335B (zh) 小鼠行为训练系统的控制方法、装置、设备及存储介质
KR20180065476A (ko) 자세 모니터링 방법 및 장치
Rad et al. Control Algorithms for Medical Recovery Robots Used in Physiotherapy
WO2025106937A1 (fr) Décomposition en modes dynamiques pour évaluation d'arrêt précoce d'entraînement de modèles d'apprentissage automatique

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17902095

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17902095

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 03/01/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17902095

Country of ref document: EP

Kind code of ref document: A1

点击 这是indexloc提供的php浏览器服务，不要输入任何密码和下载