Computational intelligence for qualitative coaching diagnostics: Automated assessment of tennis swings to improve performance and safety

Abstract

Coaching technology, wearables and exergames can provide quantitative feedback based on measured activity, but there is little evidence of qualitative feedback to aid technique improvement. To achieve personalised qualitative feedback, we demonstrated a proof-of-concept prototype combining kinesiology and computational intelligence that could help improving tennis swing technique. Three-dimensional tennis motion data were acquired from multi-camera video (22 backhands and 21 forehands, including common errors). Selected coaching rules were transferred to adaptive assessment modules able to learn from data, evolve their internal structures and produce autonomous personalised feedback. The generated qualitative assessment for each swing sample was compared with the expert. The prototype demonstrated autonomous assessment on future data based on learning from prior examples, aligned with skill level, coaching scenarios and coaching rules. The generated intuitive diagnostic feedback consisted of elements of safety and performance for tennis swing technique. For safety aspects of the relative swing width, the prototype showed improved assessment (from 81% to 91%) when taking into account occluded parts of the pelvis. The next-generation augmented coaching, personalised rehabilitation and exergaming systems will learn from small data, adapt and provide personalised intuitive autonomous assessment of motion data aligned with specified coaching programme and feedback requirements.