FootStep-Tracker:An Anchor-Free Indoor Localization System via Sensing Foot Steps Chang Liu Jie Wu Abstract State Key Laboratory for Novel Department of Computer Recently,indoor localization has been a key supporting Software Technology Information and Sciences technology for most ubiquitous applications.However,most Nanjing University,P.R.China Temple University,USA localization schemes require the deployment of the anchor liuchang@dislab.nju.edu.cn jiewu@temple.edu nodes in advance to assist indoor localization.In this demo. we develop FootStep-Tracker,an indoor localization system Lei Xie Sanglu Lu via sensing the user's footsteps,without any support infras- State Key Laboratory for Novel State Key Laboratory for Novel tructures.By embedding tiny Sensor-Tag into the user's Software Technology Software Technology shoes,FootStep-Tracker is able to accurately perceive the Nanjing University,P.R.China Nanjing University,P.R.China user's moving trace,including the moving direction and dis- Ixie@nju.edu.cn sanglu@nju.edu.cn tance.Moreover,by detecting the user's activities such as going upstairs/downstairs and taking an elevator,FootStep- Chuyu Wang Tracker can correlate with the specified positions such as State Key Laboratory for Novel stairs and elevators,and determine the exacted moving Software Technology traces in the indoor map. Nanjing University,P.R.China wangcyu217@126.com Author Keywords Indoor Localization;User Activity Sensing;Sensing Foot Steps. Permission to make digital or hard coples of part or all of this work for personal or classroom use is granted without tee provided that copies are not made or distributed ACM Classification Keywords for profit or commercial advantage and that coples bear this notice and the full citation on the tirst page.Copyrights for components of this work owned by others than ACM H.5.m [Information interfaces and presentation (e.g.,HCI)]: must be honored.Abstracting with credit is permitted.To copy otherwise,to republish Miscellaneous to post on servers,or to redistribute to lists,requires prior specific permission and/or a fee.Request permissions from permissionsacm.org or Publications Dept,ACM, lnc,fax+1(212)869-0481. Introduction UbiComp'15 Adjunct,September 710,2015,Osaka.Japan. ACM978-1-4503-3575-1/15/D9. Nowadays,indoor localization schemes have been widely hmtp1k.dol.0rg10.1145/2800835.2B00844 used to support various applications.The state art indoor
FootStep-Tracker: An Anchor-Free Indoor Localization System via Sensing Foot Steps Chang Liu State Key Laboratory for Novel Software Technology Nanjing University, P.R. China liuchang@dislab.nju.edu.cn Jie Wu Department of Computer Information and Sciences Temple University, USA jiewu@temple.edu Lei Xie State Key Laboratory for Novel Software Technology Nanjing University, P.R. China lxie@nju.edu.cn Sanglu Lu State Key Laboratory for Novel Software Technology Nanjing University, P.R. China sanglu@nju.edu.cn Chuyu Wang State Key Laboratory for Novel Software Technology Nanjing University, P.R. China wangcyu217@126.com Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org or Publications Dept., ACM, Inc., fax +1 (212) 869-0481. UbiComp ’15 Adjunct, September 710, 2015, Osaka, Japan. ACM 978-1-4503-3575-1/15/09. http://dx.doi.org/10.1145/2800835.2800844 Abstract Recently, indoor localization has been a key supporting technology for most ubiquitous applications. However, most localization schemes require the deployment of the anchor nodes in advance to assist indoor localization. In this demo, we develop FootStep-Tracker, an indoor localization system via sensing the user’s footsteps, without any support infrastructures. By embedding tiny Sensor-Tag into the user’s shoes, FootStep-Tracker is able to accurately perceive the user’s moving trace, including the moving direction and distance. Moreover, by detecting the user’s activities such as going upstairs/downstairs and taking an elevator, FootStepTracker can correlate with the specified positions such as stairs and elevators, and determine the exacted moving traces in the indoor map. Author Keywords Indoor Localization;User Activity Sensing;Sensing Foot Steps. ACM Classification Keywords H.5.m [Information interfaces and presentation (e.g., HCI)]: Miscellaneous Introduction Nowadays, indoor localization schemes have been widely used to support various applications. The state art indoor
localization schemes mainly leverage WiFi[1][3]or Blue- SensorTag tooth to locate the users in the indoor environment.In most Android Smartphone cases,the deployment of anchor nodes such as WiFi AP- Accelerometer Walking s and Bluetooth beacons is required in these schemes. Activity Distance However,for a number of indoor environments,deploying Classifier Walking Gyroscope Direction the localization anchor nodes is either impossible or rather Taking Elevato可 expensive.Therefore,an anchor-free system for indoor lo- Up/Downestairs】 calization is essential.In this demo,we develop FootStep- Indoor Relereno Tracker,an anchor-free indoor localization system purely Map Position based on sensing the user's footsteps. Location Figure 2:Framework of FootStep-Tracker. references the position of stairs and elevators by detect- ing the user's activities such as going upstairs/downstairs and taking an elevator.By combining the moving trace and reference position,FootStep-Tracker can locate the user accurately on the given indoor map. Activity Classifier.It extracts features from stream raw sensors'data.According to the extracted features,Activity Figure 1:The SensorTag we used for FootStep-Tracker.We Classifier classifies the user's current activities into different embed SensorTag into the insole in the shoes and it sends data to classes,by leveraging several classify techniques,such as an Android smart phone via bluetooth. Decision Tree and Support Vector Machine.Table 1 showes the label and description of eight classes. System Design As is shown in Figure 1,we embed the tiny sensor like the Step Counter.It segments the walking data stream step SensorTag [2]into the user's shoes.SensorTag collects by step according to the pattern of the accelerometer in and sends the sensors'data to an Android smart phone via walking activities.The segment data here is not only used bluetooth.The smart phone analyzes the data and illus- for step counting,but also used for step length estimation. trates the user's exact location in the indoor map.Figure Step Length Estimator.It measures the step length of 2 shows the framework of our system.FootStep-Tracker each step.By intensively analyzing the foot movement dur- depicts the user's moving trace by estimating the walking ing walking,we build a geometric model to depict the move- distance and walking direction.Besides,FootStep-Tracker ment and rotation of the insteps along one step.Based on
localization schemes mainly leverage WiFi[1][3] or Bluetooth to locate the users in the indoor environment. In most cases, the deployment of anchor nodes such as WiFi APs and Bluetooth beacons is required in these schemes. However, for a number of indoor environments, deploying the localization anchor nodes is either impossible or rather expensive. Therefore, an anchor-free system for indoor localization is essential. In this demo, we develop FootStepTracker, an anchor-free indoor localization system purely based on sensing the user’s footsteps. Figure 1: The SensorTag we used for FootStep-Tracker. We embed SensorTag into the insole in the shoes and it sends data to an Android smart phone via bluetooth. System Design As is shown in Figure 1, we embed the tiny sensor like the SensorTag [2] into the user’s shoes. SensorTag collects and sends the sensors’ data to an Android smart phone via bluetooth. The smart phone analyzes the data and illustrates the user’s exact location in the indoor map. Figure 2 shows the framework of our system. FootStep-Tracker depicts the user’s moving trace by estimating the walking distance and walking direction. Besides, FootStep-Tracker Accelerometer Gyroscope Activity Classifier Walking Distance Walking Direction Reference Position Walking Walking Trace Indoor Map Location Figure 2: Framework of FootStep-Tracker. references the position of stairs and elevators by detecting the user’s activities such as going upstairs/downstairs and taking an elevator. By combining the moving trace and reference position, FootStep-Tracker can locate the user accurately on the given indoor map. Activity Classifier. It extracts features from stream raw sensors’ data. According to the extracted features, Activity Classifier classifies the user’s current activities into different classes, by leveraging several classify techniques, such as Decision Tree and Support Vector Machine. Table 1 showes the label and description of eight classes. Step Counter. It segments the walking data stream step by step according to the pattern of the accelerometer in walking activities. The segment data here is not only used for step counting, but also used for step length estimation. Step Length Estimator. It measures the step length of each step. By intensively analyzing the foot movement during walking, we build a geometric model to depict the movement and rotation of the insteps along one step. Based on
Table 1:Label Description eter and gyroscope data among 5 different users.Then we use them to train the classify model.We perform the clas- UST going upstairs sifier on the previous 5 users and 3 new users.They go DST going downstairs up/downstairs,walk or take elevator.Figure 3(a)shows the WALK walking accuracy of Activity Classifier on the collected data.Particu- EHG hypergravity in elevator. larly,we don't evaluate the EUP/EDOWN because that EUP EWL weightlessness in elevator. is the combination of EHG and EWL,and EDOWN is the E UP elevator going up combination of EWL and EHG.On average,we achieve a E_DOWN elevator going down classification accuracy of 96.2%. SS stand still Location Accuracy.To evaluate the location performance of FootStep-Tracker,we tested it among 5 users.They wear the proposed model,we calibrate the step length estimation FootStep-Tracker,walking along an approximately 100m- by efficiently employing the accelerometer combined with long path which is the full red line in Figure 3(b)in our de- gyroscopes. partment building.Users taked the elevator (the left bottom one)down to this floor,and moved along the path.After Step Direction Estimator.It measures the moving direc- he/she turned six times,he/she stopped at another elevator tion once we detect the turning steps from the heading in the map,and we stopped the tracking.Figure 3(c)shows forward steps.By merging consecutive turning steps in a the average location accuracy for each user.On average, whole turning process,and extracting the integral,max and the location accuracy is within 1m. variance as features,we classify the current turning direc- tions into three classes:left turn,right turn. Demo Application FootStep-Tracker interacts with user via an Android applica- Reference Position Estimator.It estimates the reference tion.Figure 4 shows the snapshot of the application.When positions of the moving trace based on the current activity user connecting the SensorTag and moving in the indoor and an indoor map.As the location of elevators and stairs environment,the application timely reports the users'step are fixed in the indoor environment,Reference Position Es- number,moving trace,current activity,sensors'data and timator treats the position of elevators or stairs as the refer- the user's location in the indoor map. ence position of the moving trace. Evaluation Acknowledgements This work is supported in part by National Natural Sci- Classification accuracy of Activity Classifier.To eval- ence Foundation of China under Grant Nos.61472185. uate the accuracy of the Activity Classifier,we embed the 61373129,61321491.91218302:Key Project of Jiangsu FootStep-Tracker into user's shoes and collect sensors'da- Research Program under Grant No.BE2013116;EU FP7 ta.The Sensor-Tag's sample frequency is set as 10 Hz,and IRSES MobileCloud Project under Grant No.612212.This it is embedded in the insole as depicted in Figure 1.For work is partially supported by Collaborative Innovation Cen- each activity,we collect about 500 windows of accelerom-
Table 1: Label Description U_ST going upstairs D_ST going downstairs WALK walking E_HG hypergravity in elevator. E_WL weightlessness in elevator. E_UP elevator going up E_DOWN elevator going down SS stand still the proposed model, we calibrate the step length estimation by efficiently employing the accelerometer combined with gyroscopes. Step Direction Estimator. It measures the moving direction once we detect the turning steps from the heading forward steps. By merging consecutive turning steps in a whole turning process, and extracting the integral, max and variance as features, we classify the current turning directions into three classes: left turn, right turn. Reference Position Estimator. It estimates the reference positions of the moving trace based on the current activity and an indoor map. As the location of elevators and stairs are fixed in the indoor environment, Reference Position Estimator treats the position of elevators or stairs as the reference position of the moving trace. Evaluation Classification accuracy of Activity Classifier. To evaluate the accuracy of the Activity Classifier, we embed the FootStep-Tracker into user’s shoes and collect sensors’ data. The Sensor-Tag’s sample frequency is set as 10 Hz, and it is embedded in the insole as depicted in Figure 1. For each activity, we collect about 500 windows of accelerometer and gyroscope data among 5 different users. Then we use them to train the classify model. We perform the classifier on the previous 5 users and 3 new users. They go up/downstairs, walk or take elevator. Figure 3 (a) shows the accuracy of Activity Classifier on the collected data. Particularly, we don’t evaluate the EUP/EDOWN because that EUP is the combination of EHG and EWL, and EDOWN is the combination of EWL and EHG. On average, we achieve a classification accuracy of 96.2%. Location Accuracy. To evaluate the location performance of FootStep-Tracker, we tested it among 5 users. They wear FootStep-Tracker, walking along an approximately 100mlong path which is the full red line in Figure 3 (b) in our department building. Users taked the elevator (the left bottom one) down to this floor, and moved along the path. After he/she turned six times, he/she stopped at another elevator in the map, and we stopped the tracking. Figure 3 (c) shows the average location accuracy for each user. On average, the location accuracy is within 1m. Demo Application FootStep-Tracker interacts with user via an Android application. Figure 4 shows the snapshot of the application. When user connecting the SensorTag and moving in the indoor environment, the application timely reports the users’ step number, moving trace, current activity, sensors’ data and the user’s location in the indoor map. Acknowledgements This work is supported in part by National Natural Science Foundation of China under Grant Nos. 61472185, 61373129, 61321491, 91218302; Key Project of Jiangsu Research Program under Grant No. BE2013116; EU FP7 IRSES MobileCloud Project under Grant No. 612212. This work is partially supported by Collaborative Innovation Cen-
g 095 0.00 .05 000 00 0.00 E_HG 0.00 t.00 Q.00 D0 0.00 0.00 E0.B EWL 0.00 0.00 1.00 0.00 0.00 0.00 UST 000 0.00 000 099 0000.01 D.ST 000 0.00 0.00000 0.01 WALX 000 0.00 0.00a000.11 D.的 SS E HG EWL UST DST WALK 56▣ User1 User2 User3 User4 User5 (a)Activity Classifier performance. (a)Walking path. (b)Location accuracy for each user. Figure 3:System Evaluation. ter of Novel Software Technology and Industrialization.Jie Wu is supported in part by US National Science Foundation SETTING grants ECCS 1231461,ECCS 1128209,CNS 1138963,C- Moving Trace NS 1065444,and CCF 1028167.Lei Xie is the correspond- Step 15 ing author. REFERENCES 1.Sebastian Hilsenbeck,Dmytro Bobkov,Georg Schroth, Robert Huitl,and Eckehard Steinbach.2014. Sensors'data Graph-based data fusion of pedometer and WiFi measurements for mobile indoor positioning.In Proc.of Ubicomp.IEEE. 2.TEXAS INSTRUMENTS.2014.SensorTag.(2014). http://www.ti.com.cn/tool/cn/cc2541dk-sensor. 3.He Wang,Souvik Sen,Ahmed Elgohary,Moustafa Farid,Moustafa Youssef,and Romit Roy Choudhury. 2012.No need to war-drive:unsupervised indoor localization.In Proc.of MobiSys.ACM. Figure 4:FootStep-Tracker App
(a) Activity Classifier performance. 56m 63m (a) Walking path. (b) Location accuracy for each user. Figure 3: System Evaluation. Figure 4: FootStep-Tracker App. ter of Novel Software Technology and Industrialization. Jie Wu is supported in part by US National Science Foundation grants ECCS 1231461, ECCS 1128209, CNS 1138963, CNS 1065444, and CCF 1028167. Lei Xie is the corresponding author. REFERENCES 1. Sebastian Hilsenbeck, Dmytro Bobkov, Georg Schroth, Robert Huitl, and Eckehard Steinbach. 2014. Graph-based data fusion of pedometer and WiFi measurements for mobile indoor positioning. In Proc. of Ubicomp. IEEE. 2. TEXAS INSTRUMENTS. 2014. SensorTag. (2014). http://www.ti.com.cn/tool/cn/cc2541dk-sensor. 3. He Wang, Souvik Sen, Ahmed Elgohary, Moustafa Farid, Moustafa Youssef, and Romit Roy Choudhury. 2012. No need to war-drive: unsupervised indoor localization. In Proc. of MobiSys. ACM
