Perttu Laurinen

Activity recognition using a wrist-worn inertial measurement unit: A case study for industrial assembly lines

As wearable sensors are becoming more common, their utilization in real-world applications is also becoming more attractive. In this study, a single wrist-worn inertial measurement unit was attached to the active wrist of a worker and acceleration and angular speed information was used to decide what activity the worker was performing at certain time intervals. This activity information can then be used for proactive instruction systems or to ensure that all the needed work phases are performed. In this study, the selected activities were basic tasks of hammering, screwing, spanner use and using a power drill for screwing. In addition, a null activity class consisting of other activities (moving around the post, staying still, changing tools) was defined. The performed activity could then be recognized online by using a sliding window method to divide the data into two-second intervals and overlapping two adjacent windows by 1.5 seconds. Thus, the activity was recognized every half second. The method used for the actual recognition was the k nearest neighbor method with a specific distance boundary for classifying completely new events as null data. In addition, the final class was decided by using a majority vote to classifications of three adjacent windows. The results showed that almost 90 percent accuracy can be achieved with this kind of setting; the activity-specific accuracies for hammering, screwing, spanner use, power drilling and null data were 96.4%, 89.7%, 89.5%, 77.6% and 89.0%, respectively. In addition, in a case with completely new null events, use of the specific distance measure improved accuracy from 68.6% to 82.3%.

Efficient accelerometer-based swimming exercise tracking

The study concentrates on tracking swimming exercises based on the data of 3D accelerometer and shows that human activities can be tracked accurately using low sampling rates. The tracking of swimming exercise is done in three phases: first the swimming style and turns are recognized, secondly the number of strokes are counted and thirdly the intensity of swimming is estimated. Tracking is done using efficient methods because the methods presented in the study are designed for light applications which do not allow heavy computing. To keep tracking as light as possible it is studied what is the lowest sampling frequency that can be used and still obtain accurate results. Moreover, two different sensor placements (wrist and upper back) are compared. The results of the study show that tracking can be done with high accuracy using simple methods that are fast to calculate and with a really low sampling frequency. It is shown that an upper back-worn sensor is more accurate than a wrist-worn one when the swimming style is recognized, but when the number of strokes is counted and intensity estimated, the sensors give approximately equally accurate results.

Clustering-based activity classification with a wrist-worn accelerometer using basic features

Automatic recognition of activities using time series data collected from exercise can facilitate development of applications that motivate people to exercise more frequently and actively. This article presents a method for recognizing nine different everyday sport activities, such as running, walking, aerobics and Nordic walking, using only two-dimensional wrist-worn accelerometer. The suggested method is based on clustering the data by first using an EM -algorithm to form homogeneous groups and then applying C4.5-based decision trees inside these groups. The features extracted for classification process are simple features, such as variance and mean, which are calculated from compressed signals that contain only such points of the original time series where the derivative is equal to zero. The data were collected by ten subjects and they contained nine different sports. Using the presented method, the data were classified with an accuracy of 85%, whereas the accuracy using an automatically generated decision tree was 80%. The purpose of this method is to recognize activities in order to form an activity diary.

Smart archive: a component-based data mining application framework

Implementation of data mining applications is a challenging and complicated task, and the applications are often built from scratch. In this paper, a component-based application framework, called smart archive (SA) designed for implementing data mining applications, is presented. SA provides functionality common to most data mining applications and components for utilizing history information. Using SA, it is possible to build high-quality applications with shorter development times by configuring the framework to process application-specific data. The architecture, the components, the implementation and the design principles of the framework are presented. The advantages of a framework-based implementation are demonstrated by presenting a case study which compares the framework approach to implementing a real-world application with the option of building an equivalent application from scratch. In conclusion, the paper presents a lucid framework for creating data mining applications and illustrates the importance and advantages of using the presented approach.

Methods for classifying spot welding processes: a comparative study of performance

Resistance spot welding is an important and widely used method for joining metal objects. In this paper, various classification methods for identifying welding processes are evaluated. Using process identification, a similar process for a new welding experiment can be found among the previously run processes, and the process parameters leading to high-quality welding joints can be applied. With this approach, good welding results can be obtained right from the beginning, and the time needed for the set-up of a new process can be substantially reduced. In addition, previous quality control methods can also be used for the new process. Different classifiers are tested with several data sets consisting of statistical and geometrical features extracted from current and voltage signals recorded during welding. The best feature set - classifier combination for the data used in this study is selected. Finally, it is concluded that welding processes can be identified almost perfectly by certain features.

Application of the Extended

Resistance spot welding is used to join two or more metal objects, and the technique is widely used in, for example, the automotive and electrical industries. This paper introduces the use of the k-nearest-neighbor (knn) method to identify similar welding processes. The two main benefits achieved from knowing the most similar process are the following: 1) The time needed for the setup of a new process can be substantially reduced by restoring the process parameters leading to high-quality joints, and 2) the quality of new welding spots can be predicted and improved using the stored information of a similar process. In this paper, the basic knn method was found to be inadequate, and an extension of the knn method, which is called similarity measure, was developed. The similarity measure provides information of how similar the new process is by using the distance to the knns. Based on the results, processes can be classified, and the similarity measure proved to be a valuable addition to the existing methodology. Furthermore, process information can provide a major benefit to welding industry.

k

Process identification in the field of resistance spot welding can be used to improve welding quality and to speed up the set-up of a new welding process. Previously, good classification results of welding processes have been obtained using a feature set consisting of

nn Method to Resistance Spot Welding Process Identification and the Benefits of Process Information

54

Feature Selection for Identification of Spot Welding Processes

features extracted from current and voltage signals recorded during welding. In this study, the usability of the individual features is evaluated and various feature selection methods are tested to find an optimal feature subset to be used in classification. Ways are sought to further improve classification accuracy by discarding features containing less classification-relevant information. The use of a small feature set is profitable in that it facilitates both feature extraction and classification. It is discovered that the classification of welding processes can be performed using a substantially reduced feature set. In addition, careful selection of the features used also improves classification accuracy. In conclusion, selection of the feature subset to be used in classification notably improves the performance of the spot welding process identification system.

Resistance Spot Welding Process Identification Using an Extended knn Method

Resistance spot welding is used to join two or more metal objects together, and the technique is in widespread use in, for example, the automotive and electrical industries. This paper introduces the use of the k- nearest neighbours (knn) method to identify different welding processes. Process information can be used to find suitable initialisation parameters for welding machines or to predict the quality of welding spots using previously gathered data. In this study, the basic knn method was found to be inadequate, and an extension to the knn method was developed. The distance to the k-nearest neighbours was considered important information, and a similarity measure was formulated to provide this information to the user. According to the results, processes can be classified using the method and specific features. The similarity measure proved to be a valuable addition, which helps the user to decide whether the closest process is close enough to be classified as the same process.