Position Tracking Of Athletes Using IMU, Magneto Meter And UWB

Motivation for the Project

In order to provide motivation for this project it is important to identify the reason for undertaking this project which explain why this project is important to execute. Considering the current market for the high end consumer oriented tech market and its popularity, this project is a justified choice as it helps to explore different issues prevalent to this context and develop something new and efficient (James & Petrone, 2016). Right now wearable technologies are no doubt one of the most important part of the consumer electronics. Different innovations are being executed in this context and these innovations are being aimed at improving already available technologies which provides motivation for this project (Azam, Chatzi & Papadimitriou, 2015).

The advancement of technology has made it possible to develop high end devices for the sport industry (Kim, Chiu & Chow, 2018). Technologies are making it possible to improve the ways sport activities such as player monitoring for optimizing and improving the efficiency of the player performance and consideration of wearable tracking devices as part of this initiatives are increasing as well which has motivated to take this project to design an effective and performance oriented player monitoring system which is titled “Position tracking of an athlete using IMU, Magneto meter (GPS) and UWB”.

As already discussed there is a huge potential in the market for an efficient and improved wearable device that helps in accurate tracking of players both on the field and off the field as well (Halson, Peake & Sullivan, 2016). Hence there is a requirement for a wearable device that is not only accurate but at the same time less complicate and also easy to use because at the field and while players are in training of the field, coaches are less interested in technical complexity but demand superior performance from the device (Bailon et al., 2018). Hence the objectives of this project is not to make things technically complicated, but to increase efficiency and reliability so that the devices are easy to used even if the person is not technically highly advanced (James,  Lee  & Wheeler, 2019). Hence some of the important objectives of the project are:

• Design an affordable wearable devices for tracking athletes.
• Maintain high quality and ensure accuracy in the performance
• Ensure the product is easy to setup and easy to work with less technical complexity
• Maintain high security for the devices as it collects personal data about the athletes for tracking and improving performance by analysing training related data
• Ensure high ethical standard for the project with adherence to the ethical code of standard for conducting research for the project. 

• Significance

Application of tracking device is not a new concept in the sport. Sport professionals such as players, coaches are applying this technologies to analyse performance of the players so that it is easier for them to identify issues and plan accordingly to design strategy for improving those issues. GPS has been one of the standard technology that is being used for years for tracking players. Although GPS is no doubt an excellent technology that helps in monitoring players, it still has some issues such as:

• It is only accurate within 3 to four meters
• It is not highly accurate in complex environment or there is congestion or signal frequency is low
• Results are sometime affected when change in position and velocity is random

Importance of Wearable Technology in the Consumer Electronics Market

However an internal measurement unit is an electronic devices that comprises of accelerometers, gyroscopes and magnetometer. As IMU contains of so many sensors, it provides more accurate result than GPS and suitable for 3d motion tracking of any objects which includes bothy linear and angular motion and hence provides enhanced monitoring facility and provides a detailed insight about player fitness, skills and also provides various performance related information (Yang, Shi & Chen, 2019). 

Although IMU is better than the GPS in certain aspects. It is still not sufficient for real time analysis of player position as it requires wideband support and this is where ultra wideband technology or the UWB is an important role to play. Due to the support for wideband application, UWB has been considered for this project for accurate position estimation ((Toth, Jozkow,  Koppanyi & Grejner-Brzezinska, 2017)).

However as each of these technologies have their own set of benefits and disadvantage as well, it is important to consider ways that combines all of these technologies for designing a device that is even more precise and accurate and hence this project is a significant one as it offers an excellent means to improve efficiency and accuracy to track athletes based on real time positioning for enhanced performance.

As already discussed, when GPS, IMU and UWB technologies are combined, it is more efficient than the individual technologies, especially when tracking needs to be conducted with precision. However combining these technologies are difficult and hence proper techniques are required in this context. 

• First literature review will be conducted to have proper knowledge about the sensor technologies and articles will be collected from the google scholar to identify relevant journals which are authenticate and peer reviewed
• Once proper understanding about the technologies are gained, the project will be started
• First calibration of the sensors will be done through magnetic calibration method and later through some calibration algorithms
• In order to integrate vision measurement and inertial measurement, extended kalman filter or the EKF is considered. The EKF filter is capable to process different sample rates along with different correspondences efficiently. It is compatible with the high data rate associated with the EMU sensor and the vision updates are only executed only there is correspondences available.  Hence a thorough experiment will be conducted with the EKF filter to reduce error in the sensor performance
• For this project, lose coupling will be considered. In this approach, measurement obtained from the UWB sensor is fed the position solver associated with the GPS sensor and then this measurement is processed with the sensor fusion algorithm and thus details about the athlete is obtained 

1. Timeline

Task Name	Duration	Start	Finish
acquire resources	10 days	Wed 27-02-19	Sat 09-03-19
requirement definitions	3 days	Mon 11-03-19	Wed 13-03-19
detailed design	10 days	Thu 14-03-19	Mon 25-03-19
review of literature	2 days	Tue 26-03-19	Wed 27-03-19
Acquire and Install System	4 days	Thu 28-03-19	Mon 01-04-19
Application Development	15 days	Tue 02-04-19	Thu 18-04-19
data migration	10 days	Fri 19-04-19	Tue 30-04-19
system documentation	2 days	Wed 01-05-19	Thu 02-05-19
testing	15 days	Fri 03-05-19	Mon 20-05-19
training	10 days	Tue 21-05-19	Fri 31-05-19
production implementation	10 days	Sat 01-06-19	Wed 12-06-19
close down	0 days	Wed 12-06-19	Wed 12-06-19

Risk Assessment

Step 1 – Identify the hazards and associated risks

• Equipment related hazard
• Technical hazards
• Health hazards
• Legal hazard
• Physical hazard 

Type of hazards	Associated risk	Description
Equipment related hazards	Issues in performance of the tracking device	Equipment related hazards include performance issue of the sensors including GPS, IMU and UWB which needs proper calibration to perform properly. This issue is even more significant when all of these sensors need to be integrated together for optimum position tracking. Hence lack of proper calibration leads to equipment related hazards
Technical hazards	Project design is not efficient	Integration of the required sensors for the project is not only complex, but technically advanced as it requires various filters for sensor filters which requires extensive technical knowledge. Hence this should be treated as technical hazard as without proper application of the techniques required for the project the design of the project will not be compatible with the requirement and that is to track the position of the athlete
Health hazards	Player might face health related issues	If not proper measure is taken for reducing the electromagnetic radiation and if the electromagnetic radiation is not within limit, it might create health issues for the athletes
Legal hazard	Player might sue trainers or clubs for data breach that includes important data about player performance which other clubs might want to gain advantage in the sport	If proper encryption is not followed, data will be hacked and it will create legal hazard for the training organization or the club for which the athlete play for
Physical hazard	Injuries due to extensive training	Sensor data might indicate that payers need to increase training time to improve skill and playing techniques. however, if athletes are forced to train without analysing physical endurance, it might lead to injury

Step 2 – Identify the current risk treatments 

Risk treatment describes the part of risk management in which decisions are made about how to treat risks that have been previously identified. In this step, efforts have been made to identify the existing risk treatments that are in place to mitigate the identified risks. Risk treatment is a process of implementing measures to reduce the risks associated with a hazard. 

Priority	risk included	Risk treatment	Example
1	Calibration issue makes data collection less sufficient which affects performance	Eliminate	Calibration issue is eliminated through proper calibration of the sensors such as gps, IMU and UWB through the magnetic calibration method
4	Technical complexity affects design process and sensor integration as well	Engineer	Training is provided to work with this technology so that it becomes easier to design the project for ensured benefits
2	Athlete might face health related issue	Substitute	Low performing high sensors that has radiation issues are replaced with sensors that are high performing and has less radiation issue
1	Data is hacked due to lack of security	Eliminate	Encryption is integrated while data is collected and processed in database
5	Players might face physical fatigue due to extensive training as suggested by sensor data	Administrative	Players should not trained beyond their physical endurance even if the sensor data suggest

Step 3 – Analyse the risk 

Current Level of Risk
Likelihood	Consequence	Risk Level	Ranking
Possible	Major	High	5
Moderate	Major	Medium	2
Moderate	Major	Medium	2
Possible	Major	High	5
Unlikely	Moderate	Low	1

Step 4 – Additional risk treatments and risk acceptance 

In this step, any additional risk treatments should be identified that will reduce the overall level of risk. The remaining level of risk (residual risk) should be of such a nature that the resulting level of likelihood and consequence are acceptable for the risk owner.

Risk description	Level of risk
Issues in performance of the tracking device	High Risk (Not controlled)
Project design is not efficient	Substantial Risk (Readily controlled)
Player might face health related issues	Moderate Risk (Routine assessment)
Player might sue trainers or clubs for data breach that includes important data about player performance which other clubs might want to gain advantage in the sport	High Risk (Not controlled)
Injuries due to extensive training	Low Risk

			Risk Acceptance Categories
	Level of Risk		Action Required
	Low Risk		Risks to be managed by routine procedures. This include periodic review of player endurance level and modify it accordingly to provide optimum training to the athletes
			The activity can proceed provided that:
	Moderate Risk (Routine assessment)		• The risks have been reduced to As Low As Reasonably Practicable.
		• the Risk Assessment has been reviewed and approved by the project
			Supervisor.
			Activity can proceed provided that:
			• The risks to the activity are reduced to As Low As Reasonably Practicable.
	Substantial Risk (Readily controlled)		• Risk minimisation treatments must be implemented and documented.
			• the Risk Assessment and documented risk treatments have been reviewed
			and approved by the project supervisor.
			The proposed activity can only proceed when:
			• The risks to the activity are reduced to As Low As Reasonably Practicable.
	High Risk (Not controlled)		• risk minimisation controls must be implemented and documented.
			• the Risk Assessment and documented controls have been reviewed and
			approved by the Head of School. In order to reduce chance of sensor poor calibration synchronization time between the sensors should be as low as possible To improve data security access to the database that contains athlete performance data should be restricted to authorized users which reduces chance for database exploitation

Progress to Date

Task Name	Duration	Start	Finish
acquire resources	10 days	Wed 27-02-19	Sat 09-03-19
requirement definitions	3 days	Mon 11-03-19	Wed 13-03-19
detailed design	10 days	Thu 14-03-19	Mon 25-03-19
review of literature	2 days	Tue 26-03-19	Wed 27-03-19

So from the timeline it is clear that some of the tasks of the project has already been completed such as resource acquirement, defining requirement definitions, detailed design description and literature review as well. The other activities as listed in the timeline are hence required to be completed to finish the project.

Advancement of Technology in the Sports Industry

Conclusion 

Advancement in consumer electronics is enhancing the consumer electronics innovation and wearable device has been an integrated part of this innovation. The demand for wearable device is increasing in the sport section as well. Although GPS based tracking has long been an standard in this context, it has certain issues as well like low latency, signal interference, low tracking ability in congested areas and indoors, it provides way for even advanced technology to be applied for improving the performance of the tracking devices. However, lack of support for bandwidth makes these two te3chnoogy less efficient for developing real time monitoring device which is required for effective monitoring. Hence in this project application of UWEB technology has also been considered. Although these three technologies when combined provides excellent way to measure player position in real time for analysing player movement, speed required for fitness and skill analysis of the player, combining this technologies are technically complex and requires extensive application of sensor fusion technologies for effective analysis of sensor data. Although design and integration is complex, but focus is provided to the objective of the project that is to make the devices less complicated and easy to execute. Extensive care has also been taken to develop design plan that ensures higher accuracy of the devices. Extensive research has also been conducted for this research, but ethical standard has also been maintained while conducting the research. 

References 

Abyarjoo, F., Barreto, A., Cofino, J., & Ortega, F. R. (2015). Implementing a sensor fusion algorithm for 3D orientation detection with inertial/magnetic sensors. In Innovations and advances in computing, informatics, systems sciences, networking and engineering (pp. 305-310). Springer, Cham. 

Kim, S., Kim, H., Yoo, W., & Huh, K. (2016). Sensor fusion algorithm design in detecting vehicles using laser scanner and stereo vision. IEEE Transactions on Intelligent Transportation Systems, 17(4), 1072-1084. 

Garcia, F., Martin, D., De La Escalera, A., & Armingol, J. M. (2017). Sensor fusion methodology for vehicle detection. IEEE Intelligent Transportation Systems Magazine, 9(1), 123-133. 

Zhang, R., Candra, S. A., Vetter, K., & Zakhor, A. (2015, May). Sensor fusion for semantic segmentation of urban scenes. In 2015 IEEE International Conference on Robotics and Automation (ICRA) (pp. 1850-1857). IEEE. 

Gravina, R., Alinia, P., Ghasemzadeh, H., & Fortino, G. (2017). Multi-sensor fusion in body sensor networks: State-of-the-art and research challenges. Information Fusion, 35, 68-80. 

Ligorio, G., Bergamini, E., Pasciuto, I., Vannozzi, G., Cappozzo, A., & Sabatini, A. (2016). Assessing the performance of sensor fusion methods: Application to magnetic-inertial-based human body tracking. Sensors, 16(2), 153. 

Requirements of an Efficient Wearable Device

Fourati, H. (2015). Heterogeneous data fusion algorithm for pedestrian navigation via foot-mounted inertial measurement unit and complementary filter. IEEE Transactions on Instrumentation and Measurement, 64(1), 221-229.

Korpilo, S., Virtanen, T., & Lehvävirta, S. (2017). Smartphone GPS tracking—Inexpensive and efficient data collection on recreational movement. Landscape and Urban Planning, 157, 608-617.

Zhao, Y. (2016). Performance evaluation of cubature Kalman filter in a GPS/IMU tightly-coupled navigation system. Signal Processing, 119, 67-79.

Zihajehzadeh, S., Loh, D., Lee, T. J., Hoskinson, R., & Park, E. J. (2015). A cascaded Kalman filter-based GPS/MEMS-IMU integration for sports applications. Measurement, 73, 200-210.

Li, Z., Chang, G., Gao, J., Wang, J., & Hernandez, A. (2016). GPS/UWB/MEMS-IMU tightly coupled navigation with improved robust Kalman filter. Advances in Space Research, 58(11), 2424-2434.

Moore, T., & Stouch, D. (2016). A generalized extended kalman filter implementation for the robot operating system. In Intelligent Autonomous Systems 13 (pp. 335-348). Springer, Cham.

Yang, C., Shi, W., & Chen, W. (2019). Robust M–M unscented Kalman filtering for GPS/IMU navigation. Journal of Geodesy, 1-12.

Zhong, M., Guo, J., & Cao, Q. (2015). On designing PMI Kalman filter for INS/GPS integrated systems with unknown sensor errors. IEEE Sensors Journal, 15(1), 535-544.

Azam, S. E., Chatzi, E., & Papadimitriou, C. (2015). A dual Kalman filter approach for state estimation via output-only acceleration measurements. Mechanical Systems and Signal Processing, 60, 866-886.

Ibrahim, Z., Aziz, N. A., Aziz, N. A. A., Razali, S., Shapiai, M. I., Nawawi, S. W., & Mohamad, M. S. (2015). A Kalman filter approach for solving unimodal optimization problems. ICIC Express Lett, 9(12), 3415-3422.

Hong, S., Lee, C., Borrelli, F., & Hedrick, J. K. (2015). A novel approach for vehicle inertial parameter identification using a dual Kalman filter. IEEE Transactions on Intelligent Transportation Systems, 16(1), 151-161.

Chui, C. K., & Chen, G. (2017). Kalman filtering (pp. 19-26). Springer International Publishing.

Wang, J., Gao, Y., Li, Z., Meng, X., & Hancock, C. (2016). A tightly-coupled GPS/INS/UWB cooperative positioning sensors system supported by V2I communication. Sensors, 16(7), 944.

Zhang, K., Shen, C., Zhou, Q., Wang, H., Gao, Q., & Chen, Y. (2018). A combined GPS UWB and MARG locationing algorithm for indoor and outdoor mixed scenario. Cluster Computing, 1-10.

Guanke, L., Jianan, Y., Wen, Y., & Yanling, Z. (2018, August). Research on Seamless Positioning of Power Wearables Based on GPS/UWB Combination. In 2018 IEEE International Conference on Computer and Communication Engineering Technology (CCET) (pp. 123-127). IEEE.

Toth, C. K., Jozkow, G., Koppanyi, Z., & Grejner-Brzezinska, D. (2017). Positioning slow-moving platforms by UWB technology in GPS-challenged areas. Journal of Surveying Engineering, 143(4), 04017011.

Shi, G., & Ming, Y. (2016). Survey of indoor positioning systems based on ultra-wideband (UWB) technology. In Wireless Communications, Networking and Applications (pp. 1269-1278). Springer, New Delhi.

James, D. A., & Petrone, N. (2016). Sensors and Wearable Technologies in Sport: Technologies, Trends and Approaches for Implementation. Berlin, Germany:: Springer.

Kim, T., Chiu, W., & Chow, M. K. F. (2018). Sport technology consumers: Segmenting users of sports wearable devices based on technology readiness. Sport, Business and Management: An International Journal.

Halson, S. L., Peake, J. M., & Sullivan, J. P. (2016). Wearable technology for athletes: information overload and pseudoscience?.

Bailon, C., Damas, M., Pomares, H., Sanabria, D., Perakakis, P., Goicoechea, C., & Banos, O. (2018). Intelligent Monitoring of Affective Factors Underlying Sport Performance by Means of Wearable and Mobile Technology. In Multidisciplinary Digital Publishing Institute Proceedings (Vol. 2, No. 19, p. 1202).

James, D., Lee, J., & Wheeler, K. (2019). Introduction to Wearable Sensors. In Wearable Sensors in Sport (pp. 1-6). Springer, Singapore.