Integrated Motion And Force Acquisition System For Tracking (imfast) - Research Paper - Danushka Marasinghe

Profiles and Projects: Graduate Yearbook 2008 School of Engineering Monash University

INTEGRATED MOTION AND FORCE ACQUISITION SYSTEM FOR TRACKING (IMFAST) M.M.Danushka Ranjana Marasinghe Supervised by Dr. S.M.N.A.Senanayake ABSTRACT An integrated system has been developed to measure the forces generated on the starting blocks and time taken by the sprinter to reach a specific displacement. The focus of the system is to provide a user friendly environment for coaches and sprinters to interpret the data conveniently. Time and displacement is later converted to velocity and acceleration for analysis purposes. The forces are measured using a MLP500 Load Cell and the time is measured by a SignalTEC™ timing light system. A user friendly Graphical User Interface (GUI) has been created using LabVIEW™ for real-time acquisition, analyzing, saving and comparison of the data. The software is also able to generate a sound of a ‘Gun-shot’ to indicate the start of the race. The load cell is calibrated using a static force technique to convert voltage readings to corresponding force values and then the results are compared with the actual force readings. The time measurement system has been validated by comparing with the SMARTiming software (recommended by the manufacturer). The integrated system shows high reliability and stability relative to the existing softwares of the individual systems.

1. INTRODUCTION “On you Marks.... Get Set”.... is the standard pre starting sequence of a sprint race. Commencing of a sprint race is from the firing of the gun to the departure of the athlete from the starting blocks(Lynch, 2003). During sprinting several factors decide the winning or losing of the race. Among them, speed of the athlete is fundamental. On the other hand proficiency of starting has a significant influence on the performance. (Ozolin) . It is observed that the faster starts apply more force than the slower ones while pushing off the starting blocks(Lynch, 2003). The athlete should apply the maximum force at the optimal time in order to increase the efficiency of a start (Lynch, 2003).

The force asserted at the start point has a significant impact on the performance of the sprinter. It is known that high leg stiffness is needed for the high running speeds(Chelly and Denis, 2001). The acceleration in the start takes place over the first 30 meters of the distance, during which the athlete reaches around 90 – 95% of the maximal speed(Ozolin) The best athlete can be defined as the one who has an uniform performance during the entire competition maintaining his high level(Aldana, Gomez et al.). Therefore it is important for the athletes and coaches to perform evaluations on both sprint starting forces and acquired initial speeds. The impact of technology on sports has produced several products that assist the coaches and athletes to analyze each technique. For example, RFID (Radio Frequency Identification) Sports Timing System (Electro-Com, 2008) is a system capable of collecting timing data using TiRFID™ technology, eTIMER40 Professional Sports Timing System(ATHNETIX, 2001) uses wireless timing gates to provide precise timing data. FinishLynx (Lynx, 2007) is a system that uses RFID system to measure the reaction time and with an add-on it can be modified to record time. Starting blocks with a force transducer is the system mostly used to analyze starting force. These systems provide individual data and do not offer the relationship between starting forces and speed. The Sports Biomechanics centre at National Sports Institute (NSI) in Malaysia uses separate systems to analyze these parameters. Another shortcoming is the high investment for these equipments. The prospective solution is to develop a low cost integrated system. This initiative influenced on the creating of ‘IMFAST’ – ‘Integrated Motion and Force Acquisition System for Tracking’ which is presented as the end product of this research. Page 1

The Integrated system uses an Instrumented Starting block with a Load cell to analyze the starting forces. Since the use of optical timing gates may provide a convenient and quick accessible source of information on running speed(Harrison, 2005), the SignalTEC™ timing light system is used to measure the time taken to travel a specified distance. This time measurement is later converted to velocity and acceleration by taking the derivatives. LabVIEW™ is the main software that is used to program and operate the system. Features such as displaying useful data, integration of the system and the user friendly GUI with the capability of settling performance parameters are controlled by the LabVIEW software.

2. OBJECTIVES The research is conducted in collaboration with Monash University Sunway Campus and the National Sports Institute of Malaysia (NSI). It is required to meet the common objectives set by both parties, which are stated below,  Integrate both instrumented starting block and wireless timing lights to a single system.  Develop the user friendly GUI using LabVIEW™ to acquire data in real time.  Provide a single system capable of determining the sprinters forces on the starting blocks and subsequent displacement down the sprint lane.  Provide such a system with a computer interface which allows a user to set up performance parameters for individual training or for use of new training programs.  Provide useful information that can be analyzed by the trainers to give feedback to the sprinter on the performance and make suggestions for improvement.  Provide the ability to compare two different sessions.  Provide a starting sound for the sprinter to start the race.

3. HARDWARE & SOFTWARE The subsequent hardware and software is utilized, 3.1 Hardware 1. SignalTEC wireless timing gates system. It is used to measure the time. It consists of six timing light sets that connect to the PC via Bluetooth. Each set equipped with a gate and a reflector. 2. MLP-500 Load Cell manufactured by Transducer Techniques. It is used to measure the force exerted by the foot of the athlete. 3. TMO-1 Signal Conditioner. Used to amplify the signal generated by the load cell.

4. DI-148 Data Acquisition (DAQ) device. 5. Instrumented Starting block including two foot blocks. The load cell is attached to a foot block. 3.2 Software 1. LabVIEW™ version 8.5. This software is used to program and run the integrated system. 2. Bluesoleil version 1.6.1.4. Used to create virtual serial ports to communicate with the timing lights

4. OVERVIEW OF THE SYSTEM The integrated system executes two parallel processes to obtain the force on the starting block and the time taken by the sprinter to run a specified distance. Process 1 Process 3

Process 2

Figure 1 – Overview of the Integrated System

The system is divided into three sections for the ease of explanation. 4.1 Process 1  The force applied on the starting block is measured using the Load Cell.  The signal is amplified using a signal conditioner.  Amplified signal is converted to computer readable data and fed into the computer using the Data acquisition device. 4.2 Process 2  Time taken by the sprinter to pass a specified distance is measured using the timing lights. The distance depends on the placement and the number of timing lights used for the session.  The data is sent to the computer through Bluetooth™. 4.3 Process 3  Data from process 1 and process 2 are acquired using the developed GUI in LabVIEW.  All the attributes of the data can be analyzed after acquisition.  The data can be saved to a file.  Saved sessions can be compared.

Page 2

4.4 System Operation System Operation is shown in Figure 2.

1

4

2 3

4.5 Features of the integrated system The features available in the IMFAST are as follows,  The system is capable of acquiring data in real-time from the force acquisition system and the time acquisition system.  The user has the freedom to select the number of timing lights that can be used for each session.  Both systems initiate data collection by pressing a single START button.  The operator has the ability to generate a Gun-Shot sound by press of a button.  Time, distance, acceleration, velocity, force distribution and maximum force exerted are displayed.  The parameters are displayed on the GUI in graphs and on a figurative representation for user friendly analysis.  The collected data can be saved in a “.xls1” file. The file is stored inside a Folder created under the Athlete name.  The comparison feature enables to analyze all the attributes of two different saved sessions.

5. RESULTS The developed GUI is separated into five tabs. Each tab has given a different functionality to increase the userfriendliness. “Settings” tab comes first.

Figure 2 - System Operation

Start button at (1) correspond to number (1) at Figure 5. The operation runs in three parallel processes when the system starts. Force acquisition (2) and Motion acquisition (3) systems run concurrently with the press of the Start button (1). Combination of above mentioned two processes is called the Data acquisition system. The Comparison process (4) runs parallel with the data acquisition system to enable the user to compare two saved sessions without interfering with the data acquisition system

1

4 3

5

2

Figure 3 - GUI - Settings tab

As shown in Figure 3, the selection knob at (1) can be used to choose the number of timing lights that are active during a particular session. COM (Communication) port 1

.xls - A commonly used spreadsheet file format invented by Microsoft Excel™ Page 3

selection menus at (2) are used to select the virtual serial port numbers to all active timing lights. The user has to enter the distances between timing gates to the inputs at (3). The COM port menu at (4) is to select the port for the Data Acquisition (DAQ) device to communicate with the load cell. The LEDs at (5) are used to verify the connectivity of the timing lights.

With reference to the Figure 5, the athlete information is displayed at (1). The file path indicator (2) displays the destination of the saved file that contains information of the session and (3) shows the maximum force in Newton that incurred during the session. The force variation graph (4) displays the variation of the applied forces on the foot block. The respective y-axis value of the cursor position in graph (4) is displayed at (5).

The second tab is focused on the data acquisition. It is called the “Real-time” tab.

1 3

2

5

1

2

6 3

4

3

Figure 6 – GUI- Results tab – Motion Figure 4 - GUI - Real-time tab As illustrated in Figure 4, the integrated system begins to acquire data with the press of the START button at (1). “Gun-shot” sound can be generated by pressing the switch at (2). The graphs at (3) and (4) display the force, velocity and acceleration details in real-time. The times between timing lights are displayed at (5). LED at (6) notifies the user that the integrated system has finished capturing data. The results acquired from the real-time process are displayed in the “offline” tab on the GUI. This tab is further divided into two sections for viewing convenience. First section (Figure 5) displays the force related data while the second (Figure 6) displays the parameters related to the motion tracking. 2 1 3

4

5

As illustrated in the Figure 6, the graph at (1) represents the velocities obtained during the run. Each point on the graph symbolizes the average velocity between two adjacent timing lights. The graph at (2) indicates acceleration data in a similar manner. The figurative representation (3) shows the average velocity/acceleration and the distance between adjacent timing lights. These values correspond to the cursor positions of graphs (1) and (2). The obtained results have to be validated to ensure reliability of the system, the validation process employed is elaborated in the following section. 5.1 Force Validation The performance of the load cell is assessed by applying static loads to the transducer. The load cell is dismantled from the foot block and reoriented at 900. The load is applied on the load cell from 0N – 170N in 20N increments. The corresponding force values displayed on the software are noted. The same procedure is implemented for deceasing load. The average of the two displayed load values are plotted against the actual load values as shown in Figure 7.

Figure 5 – GUI – Results tab – Force

Page 4

The error for each session is calculated using following formula.

Error  (Measured Value )  (True Value ) (1.1)  IMFAST  SMART

Figure 7 - Force Validation graph The Actual Load vs. the Load Cell Reading graph above shows a linear relationship with a nonlinearity of 1.4%. 5.2 Time Validation The time readings acquired from the timing lights using the developed software and the SMARTiming software are compared.

The Average error between Gate 1 and 2 is 6.59% and between Gate 2 and 3 is 3.40%. The time difference between the values obtained from each software is less than 0.06 seconds. The comparison above has sufficient evidence to conclude that the time values obtained from the IMFAST system correspond to the practical values.

6. DISCUSSION The Load Cell restricts the angle adjustment of the foot block. This results in an inconvenience which can affect the performance of the sprinter.

The timing lights are placed with 2m distance between each other. The same test subject is used to run ten times in the same manner. The sessions are distributed equally among the two softwares.

The load cell converts the force applied on the starting blocks to corresponding voltage values. Thus, it is essential to covert these voltage values back to corresponding force values for ease of analyzing.

It is assumed that the commercially available software provides accurate readings and that the test subject performs equally in all ten sessions.

The procedure explained in section 5.1 is implied to obtain voltage values for the corresponding forces. The Voltage vs. Force graph has been plotted from the average values.

Comparison of a testing conducted using three gates is shown below. Session

Dist.

1 2 3 4 5 Ave.

3 3 3 3 3 3

Session

Dist.

1 2 3 4 5 Ave.

3 3 3 3 3 3

SMART 1.03 1.10 1.01 1.26 1.11 1.10

Gate 1 & 2 IMFAST Diff. 1.02 -0.01 1.05 -0.06 1.14 0.13 1.17 -0.09 1.04 -0.07 1.05 -0.06

Error % 0.87 5.34 13.29 7.14 6.32 6.59

SMART 0.78 0.83 0.83 0.95 0.87 0.85

Gate 2 & 3 IMFAST Diff. 0.95 0.17 0.89 0.06 0.82 -0.01 0.90 -0.04 0.83 -0.05 0.88 0.03

Error % 21.58 7.26 1.69 4.44 5.28 3.49

Figure 8 - Load Cell Calibration (Voltage vs. Force)

Force  0.0382(Voltage )  24.085

(1.2)

The above equation (1.1) is hardcoded in LabVIEW™ to convert the values in real time.

Table 1 - Comparison of time between IMFAST & SMARTiming Page 5

8. CONCLUSION

1

Figure 9 - Comparison Tab – Acceleration Section Figure 9 illustrates the acceleration comparison tab in the GUI that can be used to compare two saved sessions. This tab is divided into three sections to compare Force, Velocity and Acceleration. The cursor of the second graph follows the cursor movement of the first graph. Functionality of the figurative representation (1) is similar to that stated in section 5. The only difference is that the higher acceleration/velocity value is indicated in “Green” color while the lower in “Red”.

7. FUTURE RECOMMENDATIONS There are several improvements that can be implemented to the system which would increase the user friendliness as well as the accuracy of the data.  The program can be enhanced to show the time of the gun shot sound on the force acquisition graph. This can be useful to evaluate the reaction time of the player.  The number of timing lights can be increased for better resolution of data.  Another Load cell can be attached to the remaining foot block so force exerted by both feet can be analyzed.  Presently the force transducer and the data acquisition system have a wired connection to the PC. It can be programmed to acquire data wirelessly through the Wi-Fi data acquisition.  Investigate the possibility of using other wireless mediums to communicate with the timing lights instead of Bluetooth™ that has more range.

In conclusion, the integrated system that has been developed which the forces generated by a sprinter against the starting blocks and time taken by the sprinter to run from starting position to a given displacement. The integrated system is developed with a user friendly GUI where coaches and athletes can easily interpret data. Athlete information, Time, distance, acceleration, velocity, force distribution and maximum force are displayed on the GUI. Obtained results have been validated by comparison means and shown that the developed system shows high reliability and stability relative to the existing software of the individual systems.

9. ACKNOWLEDGEMENT The author would like to gratefully acknowledge Dr. S.M.N.A.Senanayake for the advices and the support given to complete this research, Mr. Khoo Boon How for the knowledge and support given, Mr. Edin Suwarganda for the continuous feedback given and also to the lab technicians, colleagues for the enormous support given to complete this research successfully.

10. REFERENCES 1. 2. 3. 4.

5.

6. 7. 8.

Aldana, N. S., Gomez, J., et al. "SpeedMed: Device for measuring velocity in Track Sports." Revista Engenieria Biomedica Mayo 2007(1): 33-37. ATHNETIX. (2001). "eTIMER40 - Professional Sports Timing System." from http://www.athnetix.com/Forms/etimer40.htm. Chelly, S. and Denis, C. (2001). "Leg Power and Hoping stiffness: relationships with Sprint running Performance." Med Sci Sports Exerc 33: 326-333. Electro-Com. (2008). "Electro-Com RFID solutions of SPORTS TIMING." Retrieved 9th April 2008, 2008, from http://www.electrocom.com.au/rfid_sportstiming1.htm Harrison, A., Jensen, Randall (2005). "A Comparison of Laser and Video Techniques for Determining Displacement and Velocity During Running " Measurement in Physical Education and Exercise Science 9(4): 12. Lynch, M. (2003). "Sprint Starting ", from http://www.lollylegs.com/training/Starting.aspx Lynx. (2007). "Lynx System Developers FinishLynx." Retrieved 25th May, 2008, from http://www.finishlynx.com/. Ozolin, E. "The Technique of the Sprint Start."

Page 6