Using MEMS inertial sensors to monitor and assess the performance of tennis serve

Abstract

In recent years, there has been a growing interest in using Micro Electro Mechanical Systems or MEMS inertial sensors (accelerometers and gyroscopes) in monitoring the performance of athletes. Although it is in experimental stage, the main features of the MEMS technology including miniaturized, light, inexpensive, real time environment monitoring, close to real time feedback were appealing for athletes, coaches, sports scientistic and engineers. The use of the MEMS inertial sensors gained popularity due to the ability to measure the linear and rotational movement of the body segments without hindering the performance of athletes in many sporting activities including swimming, golf, soccer and cricket. This thesis investigates the feasibility of using wearable inertial sensors to monitor and assess the tennis players’ first serve. Suitable high rate gyroscopes were not developed at the time of the PhD candidature, therefore a novel technique to simulate the behaviour of gyroscopes (virtual gyroscopes) to measure the angular velocity of body segments was developed. This virtual gyroscope can be attached on any segment to measure the angular velocity of that segment. The virtual gyroscope proposed in this dissertation allows for extraction of simulated angular velocity data from virtually any point on an athlete’s body or equipment. The novel one-dimensional virtual gyroscope can also measure angular velocities of any range, as opposed to real gyroscope sensors which suffer from limited measurement range.

This novel technique was applied to measure the upper arm internal rotation (one of the main contributors of the tennis serve with approximately 54% contribution) of athletes during the first tennis serve. Wrist flexion and shoulder rotation contribute approximately 31% and 10% respectively, were also measured using common vector based techniques. The results were verified against the real gyroscope and found to have a relationship and follow the same trends ((r = 0.9470, p < 0.0001), (r = 0.98914, p < 0.0001), (r = 0.8734, p < 0.0001) for the upper arm internal rotation, wrist flexion and shoulder rotation respectively).

Additionally, the use of accelerometers to detect the kinematic chain and virtual gyroscopes to assess the players from different skill levels was investigated. The peak values of the upper arm internal rotation, wrist flexion, and shoulder rotation just before impact were measured and used to categorize athletes in different levels from amateur to elite. It was shown that all the three parameters as well as the racquet head speed increased as the level of proficiency of the subjects increased. A line ( ) was fit to the scatter data containing the upper arm internal rotation, wrist flexion, and racquet head speed. The fit line is a function of upper arm rotation and wrist flexion. It was shown that the fit line can be used as a potential skill improvement tool to provide feedback on which variables (upper arm internal rotation, wrist flexion or shoulder rotation) needed to be improved. Therefore, it is envisaged that gyroscope sensors could be used for skill assessment and skill improvement for a first tennis serve. These results have great potential benefit for athletes to use non-invasive wearable inertial sensors to monitor and enhance their performance during training sessions on the tennis court.