This thesis discusses the potential of the RHIDE system, a device designed to measure
in-field wheelchair handrim push characteristics. The research question is if the RHIDE
can accurately discriminate propulsion/recovery and determine contact/release angles. This
research question is answered by determining push time (τ ), contact angle (α), release angle
(β) and push time (∆) for each push. To validate the RHIDE, these biomechanics are
measured with two different systems, one being the RHIDE and the other being the MarkerLess and Machine Learning Vision System (MLVS). Validation is conducted using data
from an experiment comprising 12 wheelchair propulsion-related tests on an ergometer, with
variations in speed (S1 and S2) and resistance (R1 and R2). For every push, τ , α, β and
∆ are extracted with both systems and the reliability of the RHIDE is calculated using an
ICC(2,1) on absolute agreement of the biomechanics for all pushes. For τ , ICC values are
0.65 for S1R1, 0.88 for S1R2, 0.98 S2R1 and 0.96 for S2R2 therefore reliability is considered
moderate, good, excellent and excellent respectively. For α, ICC values are 0.67 for S1R1,
0.59 for S1R2, 0.56 S2R1 and 0.51 for S2R2 giving moderate reliability for all tests. The
β ICC values are 0.57 for S1R1, 0.34 for S1R2, 0.35 S2R1 and 0.26 for S2R2 giving poor
reliability for all tests except S1R1. At last, ∆ ICC values are 0.17 for S1R1, 0.52 for S1R2,
0.25 S2R1 and 0.14 for S2R2, giving poor reliability for all tests except S1R2. Overall,
the RHIDE is able to accurately discriminate propulsion/recovery phases. Touch/release
angles are not reliable between systems. There are two potential causes for this. Firstly,
the RHIDE could improve its accuracy by increasing sample rate, streamline data pathways
and the introduction of a calibration protocol. Secondly, both systems measure different
contact/release angles and are therefore not comparable in this matter.
Overall, RHIDE has the potential to occupy a unique position in wheelchair propulsion
analysis by enabling the measurement of daily activities. With RHIDE, it would be feasible
to (1) provide information about propulsion technique by supplying push/release angles and
push times and (2) offer insights into daily wheelchair use over extended periods.

Propulsion power is an important metric in wheelchair racing. For a flat surface, it can be estimated from the sum of friction power and kinetic power. Usually, to determine friction power, the rolling and air drag coefficient first need to be determined with coast-down tests or other time-consuming methods. The aim of this paper was to investigate whether friction power could be estimated from Inertial Measurement Unit (IMU) data during wheelchair propulsion without the need for previously determining these coefficients. Two approaches were investigated using the kinematic data of a wheelchair athlete measured by a wheel, frame and trunk IMU. Firstly, an approach was used that considers the recovery phase of the propulsion cycle to be a coast-down period. Secondly, a machine-learning approach (Random Forest Regressor) was implemented. Coast-down tests were used to calculate a reference power with which the results from the two approaches could be compared. Results indicate that the machine-learning approach is more promising than the recovery phase analysis. However, whether the current machine-learning model can predict friction power for unseen subjects and surfaces should still be determined with inter-subject validation. Otherwise, it is recommended that a machine-learning model is trained for multiple subjects and a variation in conditions affecting friction force (surface, tyre pressure, wind, slope) to achieve a more robust model.

Insight into in-field mechanical power estimation in wheelchair racing is useful for athletes and coaches. A non-invasive method to estimate mechanical power is by using inertial measurement units (IMUs) to estimate the power lost to resistive forces during wheelchair propulsion. During the recovery phase, no propulsive force acts on the athlete/wheelchair combination and therefore, the deceleration of the athlete/wheelchair combination during this phase is caused by the power lost to resistive forces. The aim of this study was to investigate whether using deceleration in the recovery phase for estimation of resistance using IMUs is applicable in wheelchair racing. To approach the instantaneous velocity of the athlete/wheelchair combination and therefore the deceleration during the recovery phase, the kinematics of the wheelchair and the upper body were measured and used for three different methods. The simplest method is to use the velocity of the wheelchair (v_wc), the second method takes the influence of the trunk movement into account (v_com,2seg) and the last method takes the influence of all upper body segments into account (v_com,tot). The results of this study indicate that using v_wc, v_com,2seg or v_com,tot for estimating the deceleration during the recovery phase is not yet suitable to provide accurate estimation of the total resistance compared to drag test based results. This indicates that application of this method of estimating resistance forces in the recovery phase is not as straightforward as initially anticipated. Despite the potential benefits of this approach, the results suggest that refinement of this method or development of a new method is necessary to obtain an accurate estimation of the total resistance.

The first aim of this master thesis is to explore whether the normal force on the castor wheels can be estimated from IMU data using a machine learning approach. The second aim is to evaluate whether incorporating the changing load distribution due to trunk movement could improve the friction power estimation compared with neglecting changes in load distribution. Twenty-five subjects performed forward handrim wheelchair propulsions with no trunk, moderate and fast trunk movement on a treadmill in a wheelchair with six conditions regarding wheelchair mass and tire pressure. Two IMUs were placed on the trunk and wheelchair and two load pins in each castor wheel measured the normal force. After feature, model and hyperparameter selection, a model was trained to estimate the normal force on the castor wheels in percentage of the total weight from the IMU data. Accordingly, the predicted instantaneous normal force is used to calculate the friction power including changing mass distribution. When using the linear velocity and acceleration of the wheelchair and the, linear acceleration of forward movement of the trunk, adequate estimations (MAE of 3.69% total weight) of the normal force from an LSTM model can be obtained for unseen subjects. This model is robust for wheelchair settings regarding wheelchair mass and tire pressure and for propulsions with no, moderate and fast trunk movement. The instantaneous friction power prediction incorporating the changing load distribution is proven to more accurate during propulsions with moderate and especially fast trunk movement. Coaches, sport scientists, and athletes may find this model useful for analysing the effect of different propulsion techniques or wheelchair conditions on the friction power. As part of a larger context, this research will contribute to the process of filling the technological gap of in-field monitoring mechanical power. Future research must validate the robustness of the model during game situations