Digital Twin Coaching for Physical Activities: A Survey

Abstract

Digital Twin technology has been rising in popularity thanks to the popularity of machine learning in the last decade. As the life expectancy of people around the world is increasing, so is the focus on physical activity to remain healthy especially in the current times where people are staying sedentary while in quarantine. This article aims to provide a survey on the field of Digital Twin technology focusing on machine learning and coaching techniques as they have not been explored yet. We also define what Digital Twin Coaching is and categorize the work done so far in terms of the objective of the physical activity. We also list common Digital Twin Coaching characteristics found in the articles reviewed in terms of concepts such as interactivity, privacy and security and also detail future perspectives in multimodal interaction and standardization, to name a few, that can guide researchers if they choose to work in this field. Finally, we provide a diagram for the Digital Twin Ecosystem showing the interaction between relevant entities and the information flow as well as an idealization of an ideal Digital Twin Ecosystem for team sports’ athlete tracking.

Precision strength training: Data-driven artificial intelligence approach to strength and conditioning

“…The majority of deep learning studies and startups have targeted recreational gym-goers [145,850]. For example, algorithms and models have been developed for exercise recognition [401,587,646,651,747], automatic training diary logging [359,403,774], repetition counting [27,639], adherence monitoring [103], and real-time exercise technique evaluation and correction [117,133,254,349,371,372,589,607,610,845,900].…”

“…In our simplied example, the athlete has a knee valgus [338], resulting from weak hip abductors, and we can follow longitudinally (V, see e.g. [254,461]) the ecacy of recommended exercises for strengthening. (B) The NMS model can be used to drive haptic feedback system, implemented for example in a smart compressive garment, and use to provide real-time haptic feedback for the athlete to correct movement execution.…”

“…[70,508,581,922]), with the NMS in idealised case would be able to detect automatically overall weakness in hip abductors, or sessionspecic issues for example due to poor warm-up or insucient recovery from previous session possibly leading to increased injury risk, complimentary to human expert assessment [316]. The same model could be then used to visualize the progression in hip abductor strength from acquired data, and display it on a dashboard system for the athlete and the coach [254]. Squat images from John Paul Cauchi's video on mitigating knee valgus https://youtu.be/Lt6OxhEHavQ.…”

“…ing context (see gure 1), the training progress can be interpreted as the disease progression trajectory, the changes in training program and active recovery interventions as medical intervention/individualised medical treatments, and the clustering/phenotyping athletes based on their response to training, e.g.machine learning recommender system for sports[254,574] with recommendations like you athletes with similar trunk:thigh ratios cannot squat as upright'. The personalisation in strength training, is typically based on subjective measures for daily preparedness and logged training loads, with training adjustments done subjectively by either human expert [210, 276, 392, //www.jtsstrength.com/product/powerliftinga-i/), which rst surveys basic athlete characteristics, then designs training periodisation and updates the program dynamically based on the logged progress of the athlete.…”

Deep Learning-Enabled Multitask System for Exercise Recognition and Counting

“…Technology innovation plays an increasingly significant role in improving exercise experience. In particular, Digital Twin coaching [1] is a promising area, which is starting to be explored. It allows us to provide individuals with a digital coach by utilizing advances in machine learning.…”