外文翻译电气1001-201009006-黄辉英文文献Sensor technology for smart homes_第1页
外文翻译电气1001-201009006-黄辉英文文献Sensor technology for smart homes_第2页
外文翻译电气1001-201009006-黄辉英文文献Sensor technology for smart homes_第3页
外文翻译电气1001-201009006-黄辉英文文献Sensor technology for smart homes_第4页
外文翻译电气1001-201009006-黄辉英文文献Sensor technology for smart homes_第5页
免费预览已结束,剩余1页可下载查看

下载本文档

版权说明:本文档由用户提供并上传,收益归属内容提供方,若内容存在侵权,请进行举报或认领

文档简介

Maturitas 69 2011 131 136 Contents lists available at ScienceDirect Maturitas journal homepage Review Sensor technology for smart homes Dan Dinga b Rory A Coopera b Paul F Pasquinac Lavinia Fici Pasquinad aDepartment of Rehabilitation Science and Technology 5044 Forbes Tower University of Pittsburgh Pittsburgh 15260 PA USA bHuman Engineering Research Laboratories VA Pittsburgh Healthcare System 7180 Highland Drive 151R1 H Pittsburgh 15206 PA USA cDepartment of Orthopedics and Rehabilitation Walter Reed Army Medical Center Washington 20307 DC USA dDepartment of Architecture Catholic University of America Washington 20064 DC USA a r t i c l ei n f o Article history Received 7 March 2011 Accepted 11 March 2011 Keywords Smart homes Sensor technology Independent living Aging Disability a b s t r a c t Asmarthomeisaresidenceequippedwithtechnologythatobservestheresidentsandprovidesproactive services Most recently it has been introduced as a potential solution to support independent living of people with disabilities and older adults as well as to relieve the workload from family caregivers and healthproviders Oneofthekeysupportingfeaturesofasmarthomeisitsabilitytomonitortheactivities ofdailylivingandsafetyofresidents andindetectingchangesintheirdailyroutines Withtheavailability of inexpensive low power sensors radios and embedded processors current smart homes are typically equipped with a large amount of networked sensors which collaboratively process and make deductions from the acquired data on the state of the home as well as the activities and behaviors of its residents This article reviews sensor technology used in smart homes with a focus on direct environment sensing and infrastructure mediated sensing The article also points out the strengths and limitations of different sensortechnologies aswellasdiscusseschallengesandopportunitiesfromclinical technical andethical perspectives It is recommended that sensor technologies for smart homes address actual needs of all stake holders including end users their family members and caregivers and their doctors and therapists More evidence on the appropriateness usefulness and cost benefi ts analysis of sensor technologies for smart homes is necessary before these sensors should be widely deployed into real world residential settings and successfully integrated into everyday life and health care services 2011 Elsevier Ireland Ltd All rights reserved Contents Introduction 132 Materials and methods 132 Sensor technology used by smart homes 132 Direct environmental sensing 132 Simple binary sensors 132 Video cameras 133 Radio frequency identifi cation RFID 133 Other sensors 133 Infrastructure mediated systems 133 Discussion 134 Clinical considerations 134 Technical considerations 134 Ethical considerations 134 Conclusions 135 Contributors 135 Confl ict of interest 135 Corresponding author at Human Engineering Laboratories VA Pittsburgh Healthcare System 7180 Highland Drive 151R1 H Pittsburgh 15260 PA USA Tel 1 412 954 5278 fax 1 412 954 5340 E mail address dad5 pitt edu D Ding 0378 5122 see front matter 2011 Elsevier Ireland Ltd All rights reserved doi 10 1016 j maturitas 2011 03 016 132D Ding et al Maturitas 69 2011 131 136 Provenance and peer review 135 Acknowledgements 135 References 135 1 Introduction A smart home refers to a residence augmented with sensors to observe the environment and devices actuators to provide proac tive services with the goal to improve the occupant s experience Thesmarthomeconceptwasoriginallydevelopedwiththeprimary focus on providing convenience improving security and saving energy Over the past decade smart home technology has increas ingly targeted use by people with reduced capabilities due to aging ordisability 1 3 Thisincreaseininterestisrelatedtothepotential for smart homes to support the ability of older adults and peo ple with disabilities to remain living independently in their own homes and to reduce the sometimes considerable efforts of family or professional caregivers 2 Today there are a number of smart homesaroundtheworld 1 12 Mostofthesehomesarecorporate showrooms or research laboratories that are not occupied by reg ular residents and have been primarily focused on demonstrations of technological capabilities 4 7 8 12 A small number of these homesareresidentialhomesincludinggrouphomesequippedwith a sensors to monitor resident behaviors and detect anomalous events 5 13 One of the key supporting features of a smart home is its ability to monitor the activities of daily living and safety of its residents With the availability of inexpensive low power sensors radio fre quency chips and embedded processors existing smart homes are typicallyequippedwithalargeamountofnetworkedsensorswhich collaboratively process and make deductions from the acquired data on the state of the home as well as the activities and behav iors of its residents This article reviews sensor technologies used insmarthomes discussesthestrengthsandlimitationsofdifferent types of sensors and offers suggestions for future directions 2 Materials and methods Three types of sensor technologies have demonstrated ability to address the challenges of sensing human activity in a smart home including wearable devices where sensors are worn by the residents direct environment components where sensors are dis tributed in the environment and infrastructure mediated systems where sensors are installed on an existing home infrastructure Sensors worn by the residents can be embedded into clothes eye glasses shoes and wristwatches or positioned directly on the body 14 15 They monitor features that are descriptive of the person s physiological state or movement For example wear able accelerometers are commonly used for activity recognition including fall detection 16 and wearable RFID radio frequency identifi cation readerscanbeusedtodetectinteractionwithtagged objects 17 Direct environment components typically consists of a set of sensors and an associated sensor network wired or wire less to transfer data to a centralized monitoring system where sensor fusion and activity inference take place 12 Infrastructure mediated systems leverage existing home infrastructure such as the plumbing or electrical system to mediate the transduction of events 18 Thisarticlefocusedondirectenvironmentcomponents and infrastructure mediated systems technologies This literature review includes published work that has under gone peer review The search was restricted to articles in journals chapters of periodicals and proceedings of conferences written in English and published between 2000 and 2010 Keyword searches were conducted in PubMed IEEE Xplore or using the Google search engine The keywords used were smart home smart environ ment intelligent environment sensor technology ubiquitous sensing wireless sensor network infrastructure mediated sens ing RFID activityrecognition contextaware smartappliance motion detectors binary sensors contact switches video sen sors privacy ethical issues These keywords were used alone or in combination 3 Sensor technology used by smart homes Many smart homes today adopt the concept of ubiquitous sens ing where a network of sensors integrated with a network of processing devices yield a rich multi modal stream of data These sensory data are analyzed to recognize and monitor basic and instrumental activities of daily living performed by the residents such as bathing dressing preparing a meal and taking medication This approach has the potential to allow smart homes to capture patterns possibly refl ecting physical and cognitive health condi tions and then recognize when the patterns begin to deviate from individualizednormsandwhenatypicalbehaviorthatmayindicate problems or require intervention 19 3 1 Direct environmental sensing Direct environmental sensing is commonly based upon many simple binary sensors throughout the home video cameras and RFID technology This type of sensing may provide valuable insight into actual activities and contexts but comes with practical and complexity costs 18 Installation and maintenance can be cost prohibitive especially when dealing with installation in existing homes It can also create privacy concerns especially with the use of video cameras 20 Finally an instrumented environment alone has not proven suffi cient for detecting activities when multiple residents are present 3 1 1 Simple binary sensors One particular type of sensor that is commonly used in smart homes is binary sensor which simply detects the state of an object or movement with a single digit 1 or 0 Various types of binary sensors have been used in smart homes including motion detec tors pressure sensors and contact switches Motion detectors and pressure sensors are usually used to detect occupant presence and locationsthroughoutthehouse 21 Forexample theSmartCondo in Alberta used thirteen passive infrared motion sensors spaced throughout the condominium to locate its occupant at key areas 12 TigerPlace a retirement community helping residents age in place used a motion sensor on the ceiling above the shower to detect showering activity 22 The Ubiquitous Home in Japan was instrumented with 18cm 18cm binary pressure detection units throughout the apartment fl oor to track residents and detects fur niture positions 6 Contact switches are usually installed on the doors in a smart home such as the front door and doors of cabinets and appliances to provide information on the specifi c interaction that the occupant performs with objects and appliances For exam ple the Ubiquitous Home installed 131 contact switches to detect the open close states of doors and drawers 6 Majeed and Brown classifi eddataloggedfromcontactswitchesandmotionsensorsvia fuzzyrulesintooneofsixgeneralactivitiessuchassleeping prepar D Ding et al Maturitas 69 2011 131 136133 ing and eating food and receiving visitors 23 The advantages of simple binary sensors include low cost easy installation and low privacy concerns However these sensors only give information at anabstractlevelandthusarelimitedwheninferringactivities 21 For example it is impossible to infer what item is removed from a cabinet by simply knowing the state of the cabinet 3 1 2 Video cameras Video cameras are considered high content sensors which pro vide rich sources of information both for human observation and forcomputerinterpretation However theyusuallyintroducemore technical challenges with respect to storage requirements and information extraction as well as social challenges around privacy when compared with simple binary sensors 20 24 Tabar et al employed a network of three cameras in a room to complement a wearable accelerometer for detecting falls These cameras were triggered when a fall was detected based upon the accelerome ter signals and real time image processing was used to analyze the data and estimate the occupant s posture This setup not only conserved the privacy of the occupant when there was no need to transmit visual data but helped reduce the number of false alarms causedbysensitiveaccelerometersignals 25 TheTigerPlaceteam used an event driven video sensor network to track relevant data on gait and range of motion and to detect falls 5 To preserve the privacy of the residents they developed algorithms to iden tify a moving person in the image and extract a silhouette The PlaceLab alive inlaboratoryforthestudyofubiquitouscomputing technologies in the home used nine infrared cameras nine color cameras and eighteen microphones throughout their apartment in cabinets and above working surfaces such as the offi ce desk and kitchen counters Image processing algorithms were used to select from the four video streams and the one audio stream that best captured an occupant s behavior based on motion and the camera layout in the environment These data streams were synchronized with other sensor data and saved to a portable disk drive 8 3 1 3 Radio frequency identifi cation RFID RFIDisatechnologythatusescommunicationviaradiowavesto exchange data between a reader and an electronic tag When inter rogated by a RFID reader the tags respond with a unique identifi er as well as information stored in their memory There are usually two types of RFID tags i e passive tags that have no power source and are often attached to an object for detecting user object inter action and active RFID tags that contain a battery and provide greater range and could be carried by a person for personal iden tifi cation One of the major limitations of RFID technology is their reliability and stability especially when reading through liquid or metals Nonetheless the ability of RFID tags to be placed out of sight and to track multiple people make them well suited for smart homes This section provides two examples on how RFID tags and readersandtagshavebeenembeddedintoobjectsandtheenviron menttodetectuser objectinteraction TheGatorTechSmartHouse developed a SmartPlug system and a SmartWave system based on RFID technology In the SmartPlug system every power outlet in the house was equipped with a RFID reader and the plugs of elec trical devices like lamps or radios were attached with a RFID tag Whenever a device was plugged into an outlet the RFID reader in the outlet read the tag and sent data to the main computer The computer could directly identify each device and its location as well as control the device 26 Monitoring the use of electrical devices at home could provide valuable information on the occu pant s lifestyle and help detect deviations from daily routines that may be associated with functional decline The SmartWave sys tem is a pervasive microwave based cooking system where an RFID reader placed under the countertop was used to determine which food packet was being scanned to obtain the cooking instructions from a database and automatically program the microwave based on the instructions 27 3 1 4 Other sensors A variety of other sensors can be deployed in smart homes to help infer activities as well as trigger automatic services 8 22 28 30 These sensors usually provide more specifi c informa tion than simple binary sensors but give fewer details than video cameras The PlaceLab used 34 temperature sensors 10 humidity sensors 5lightsensors andonebarometricpressuresensortocap tureandadjustinteriorconditionsoftheirapartment 8 TigerPlace used a bed sensor a pneumatic strip installed under the bed linens todetectpresence respiration normal abnormal pulse low nor mal or high and movement in the bed and reported four levels of bed restlessness to remote caregivers They reported a case study on a 96 year old woman living alone in her apartment who had a signifi cant cardiac event When reviewing her bed sensor data collected before the event occurred a signifi cant increase in bardy cardia slowpulserateof1 30beatsperminute wasdetectedanda decrease in bed restlessness at all levels was also noted during the same time period 22 Chen et al used a single miniature omni directional microphone near the sink to detect and classify major activitiesoccurringwithinabathroomincludingshowering brush ing of teeth washing hands fl ushing and urination Preliminary resultswereencouragingwiththeaccuracyrateformostsoundcat egoriesbeingabove84 Theirgoalwastoautomaticallygenerating customized personal hygiene behavioral reports for caregivers and geriatric clinicians Clearly in spaces such as bathrooms where the useofvideosensorsisnotsociallyacceptable soundscouldprovide an alternative source of information 28 3 2 Infrastructure mediated systems Infrastructure mediated systems have recently generated con siderableinterestasanunobtrusive low cost andpracticalmethod for activity classifi cation in smart homes 31 Compared to installingsimplebinarysensorsthroughoutanentirehouse infras tructure mediate systems require the installation of one or a few sensors along the existing infrastructure which signifi cantly reduces the cost and complexity of deployment and maintenance 18 However it may not provide suffi cient detail on the activities and their contexts 32 Patel et al leveraged the existing ductwork infrastructure of central heating ventilation and air conditioning HVAC systems for detecting whole house gross movement and room transition They instrumented an HVAC s air fi lter with fi ve pressure sensors each sensing in both directions The magnitude of the pressure change across all the sensors was used to identify unique changes in airfl ow in the physical space caused by human inter room movements such as a person walking through a par ticular doorway or the opening and closing of a particular door Preliminary results showed they could classify unique transition events with up to 75 80 accuracy 18 Froehlich et al developed HydoSense a customized pressure sensor that could be installed at any accessible location within a home s existing water infras tructure such as an exterior hose bib and utility sink spigot for activity sensing The continuous analysis of pressure allowed them to identify individual fi xtures at which water was being used as well as an estimate of the amount of water being used Prelimi nary results showed they could identify fi xture events with 97 9 aggregateaccuracyandestimatewaterusagewitherrorratescom parable to utility supplied water meters 33 Patel et al developed a powerline noise analyzer that can be plugged into an ordinary wall outlet to detect a variety of electrical events throughout the home The analyzer can detect the electrical noise on residential power lines created by the abrupt switching of electrical devices and the noise created by certain devices while in operation The 134D Ding et al Maturitas 69 2011 131 136 system was tested in one home for several weeks and in fi ve homes for one week each and results showed they could learn and clas sify various electrical events with accuracies ranging from 85 to 90 34 4 Discussion Inspiteofthegrowingnumberofinitiativesinsmarthomearea cha

温馨提示

  • 1. 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
  • 2. 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
  • 3. 本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
  • 4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
  • 5. 人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
  • 6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
  • 7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。

最新文档

评论

0/150

提交评论