Monitoring and Control of Power Preparedness of Athletes in Flatwater
Rowing and Canoeing Using Strain Gauges
Eugene Alooeff
1 a
, Daniil Guseinov
2 b
, Dmitry Lukashevich
2 c
, Dmitry Bykov
2 d
and Alexander Minchenya
3 e
1
R&D Department, ATEK, Wroclaw, Poland
2
Department of Innovative Sports Technologies, Belarusian State University of Physical Education,
pr.Pobeditelej 105, Minsk, Belarus
3
Integrated Devices LLC, 9 Kulman Street, Minsk, Belarus
Keywords:
Flatwater Rowing and Canoeing, Strain Gauges, Data Processing, Power Preparedness.
Abstract:
Control over the preparedness of flatwater rowers and canoeists is realized in the process of solving a variety
of particular problems related to the organization of training activities, planning and dosing of loads, selection
of training tools and methods for assessing various aspects of readiness and competent interpretation of the
results obtained for carrying out corrective measures. Solving these problems is facilitated by strain gauges,
which make it possible to record dynamic and some kinetic parameters in natural rowing conditions (on water
recording) as part of training and monitoring activities. The article presents the developed strain gauge sensors,
describes the features of their calibration and mounting on an athlete’s paddle, and also proposes software for
automated processing of recorded data. The article is based on a practice-oriented study on experimental
testing of the developed sensors as part of the training and monitoring process.
1 INTRODUCTION
Monitoring technical and power preparedness in flat-
water rowing and canoeing is an integral part of the
training process, the results of which are purposefully
used in the selection of training tools, planning and
rationing of external loads, as well as in assessing the
effectiveness of training sessions. The main criteria
that determine the possibility of including certain in-
dicators in the control program are their information
content and reliability (Kolumbet, 2017).
Flatwater rowing and canoeing are sports with a
predominant manifestation of power abilities of ath-
letes that place high demands on the anaerobic mech-
anisms of energy supply for athletes (Rosdahl, 2019).
Accordingly, the external manifestation of an ath-
lete’s strength and speed preparedness is the power
of movements realized by explosive muscle efforts
in a minimum period of time (Kvashuk, 2021). The
a
https://orcid.org/0009-0009-2958-2431
b
https://orcid.org/0000-0003-4812-1832
c
https://orcid.org/0000-0003-3506-9430
d
https://orcid.org/0000-0002-9516-5080
e
https://orcid.org/0009-0001-3489-4597
performance of each stroke and the advancement of
the boat directly depends on the power of the move-
ments (Wainwright, 2014; Wainwright, 2015). The
efficiency of the rower’s movements can be assessed
by various methods, starting with a simple measure-
ment of the time it takes to cover the distance. Vari-
ables such as boat speed over the course and pace are
indicators of an athlete’s performance and can be used
as a fairly simple method for comparing an athlete’s
performance with competitors, as well as with one’s
own previously demonstrated results (Gomes, 2022;
Redwood-Brown, 2021).
However, these variables do not allow us to as-
sess the level of speed-strength readiness, establish
cause-and-effect relationships in achieving high per-
formance of movements and understand how the
rower achieves his results (Oronova, 2018; Brown,
2010). This gave impetus to the emergence of stud-
ies aimed at complex biomechanical control of the
flatwater rower’s motor actions, which would take
into account objective data that comprehensively re-
flect the performance and efficiency of movements
in terms of kinematic, dynamic, energetic and phys-
iological parameters. In particular, one of the most
relevant areas of research work today is the develop-
Alooeff, E., Guseinov, D., Lukashevich, D., Bykov, D. and Minchenya, A.
Monitoring and Control of Power Preparedness of Athletes in Flatwater Rowing and Canoeing Using Strain Gauges.
DOI: 10.5220/0012893700003828
Paper published under CC license (CC BY-NC-ND 4.0)
In Proceedings of the 12th International Conference on Sport Sciences Research and Technology Support (icSPORTS 2024), pages 15-21
ISBN: 978-989-758-719-1; ISSN: 2184-3201
Proceedings Copyright © 2024 by SCITEPRESS – Science and Technology Publications, Lda.
15

ment, optimization and implementation of fully au-
tonomous devices for measuring the dynamic param-
eters of rowing in natural conditions into the structure
of the training process of professional rowers and am-
ateur athletes (Galipeau, 2018).
2 STUDY METHODS
It is known that strength and force impulse reflect
the realization of an athlete’s speed-strength potential,
however, to correctly assess the efficiency of a stroke,
it is necessary to have data characterizing the move-
ment of the boat for each stroke (Baker, 2012). To do
this, in addition to the already indicated force and its
impulse of force, in order to assess the effectiveness
of interaction with the aquatic environment, it is nec-
essary to have data reflecting the frequency of strokes
(tempo), the length of the drive in the support part
(amplitude), the area of the conditionally fixed sup-
port, the power of the stroke (characterizes the per-
formance of movements athlete) and the length of the
boat rental for each stroke (Gomes, 2022). To solve
this problem, many researchers resort to the method
of simulating racing conditions in non-competitive or
training conditions, which makes it possible to ana-
lyze the relationship of various parameters (Bertozzi,
2022; Bonaiuto et al., 2020b). However, at present,
this problem has not been fully solved, since it re-
quires the use of several measuring systems simulta-
neously, which significantly complicates the process
of data recording and has a negative impact on the
biomechanical structure of the athlete’s movements.
Therefore, this direction can be considered promis-
ing with the need to overcome all kinds of technologi-
cal limitations of existing and already used measuring
systems.
Currently, there are no commercially available de-
vices with the functionality to record and analyze the
parameters of the force and impulse of the stroke
force in natural conditions. However, the high rele-
vance of research in this direction, including in solv-
ing the problem of developing methods and algo-
rithms for monitoring and assessing speed-strength
readiness in kayaking and canoeing in natural condi-
tions, are confirmed by analysis of a significant num-
ber of publications (Bonaiuto et al., 2020a; Bonaiuto
et al., 2022).
The most promising direction in harnessless row-
ing to solve this problem seems to be the use of
portable wearable strain gauge sensors, circuitry im-
plemented on the basis of MEMS technology. This
makes it possible to create ergonomic designs (with
small overall dimensions and weight), with function-
ality that allows for high-frequency data recording
with high accuracy. Such sensors provide not only
registration and conversion of the relative magnitudes
of the mechanical impact on the sensitive element into
an electrical signal, but also its primary processing
and conversion into a discrete numerical and graphic
signal in real time.
The purpose of the study is experimental testing of
the developed strain gauge sensors in real conditions
of the training and monitoring process for a long pe-
riod of time.
2.1 Research Methods
In this study, we used strain gauge sensors developed
by us, structurally implemented in the format of a
two-section measuring device. Sensor section is con-
nected to the data recording section via connecting
PDC (Power and Data cable) plugged into the appro-
priate connectors (Figure 1).
Figure 1: Strain gauge on the oar to record the dynamic
characteristics of the stroke.
Sensor section is mounted on the paddle shaft us-
ing metal or plastic mounting clamps. The sensitive
element, represented by a strain gauge, is glued to
the elastic thin metal plate of the sensor section. The
plate fits tightly directly to the paddle shaft. The pro-
file of the metal plate provides improvised grooves for
mounting clamps. Microcircuits, controls and indica-
tors of operation and mode changes are placed in the
data recording unit.
The data recording unit has the necessary hard-
ware and software capabilities for simultaneous
recording of signals from 4 sections with strain gauge
sensors. The initial recorded data is the time and the
resulting external load acting on the paddle during
rowing and expressed in Newtons. To ensure correct
use of the strain gauge sensors, calibration must first
be carried out. The calibration functionality is prein-
stalled in the software part of the data recording sec-
tion.
icSPORTS 2024 - 12th International Conference on Sport Sciences Research and Technology Support
16

2.1.1 Strain Gauge Calibration
The calibration process consists of several required
steps and can only be performed on one sensor at a
time. First you need to correctly configure the system
file. In particular, it is necessary to indicate the weight
of the calibration weight in newtons. Next, it is nec-
essary to secure the paddle in a horizontal position
on two fixed supports, the location of which corre-
sponds to the athlete’s grip (Figure 2); A – diagram of
the calibration of sensors installed on the paddle for
canoeing; B – diagram of the calibration of sensors
installed on the paddle for kayaking).
Figure 2: Schemes for calibrating strain gauge sensors on
different paddles.
The support points must have a cylindrical cross-
section, the radius of which should not exceed the
radius of the paddle shaft. This is really necessary
to create the point interaction under conditions of the
two-point deformation. Then you need to change the
operating mode of the device by pressing a certain
combination of buttons. The transition to calibration
mode will be accompanied by a corresponding indi-
cation.
The calibration process itself consists of two
mandatory procedures: fixing the “zero indicator” of
the calibration - an unloaded state, when the paddle is
not affected by any external load, and it is located in
a horizontal position on two fixed supports; fixation
of the “load indicator” of calibration - a state when a
calibration weight is applied to the paddle blade, and
the paddle is located in a horizontal position on two
fixed supports. These two states are fixed by pressing
one of the sensor control buttons.
To verify the accuracy and reliability of the values,
which were recorded by the sensor, a study was previ-
ously conducted using a specially designed stand and
a universal electronic testing machine MTS Criterion
43 (limits of permissible relative error of force mea-
surements – no more than 1%) (Guseinov D.I., 2024).
2.1.2 Starting the Data Registration Process
To record biomechanical rowing data, an paddle with
attached strain gauge sensors and a recording unit is
transferred to the athlete into the boat. The wires are
fixed on the forearms and shoulders with elastic ban-
dages (Figure 3).
Figure 3: Strain gauges attached to the athlete’s paddle.
It should be noted that the orientation of the sensor
sections coincided with the orientation of the paddle
blades. The sections themselves were connected to
the data recording unit via PDC wires.
Next, the conditions for performing test tasks are
announced and explained to the athlete, data record-
ing is activated and the recording unit is placed in the
boat (Figure 4).
Figure 4: Activate data recording and place the recording
unit in the boat.
2.1.3 Testing
One athlete aged 17 years, with experience of per-
forming at regional competitions, took part in the
study. He is included in the roster of the national
team (trainee variable team composition). The ath-
lete was asked to perform 3 accelerations with maxi-
mum intensity over a 100 m distance. The study was
carried out in the equipped rowing channel. Special-
ized floats (buoys) installed along the distance every
10 meters acted as reference points for athlete when
passing control segments. Between accelerations, the
athlete rested all the time during the next 100 m. He
covered the recovery 100 m with low intensity.
Monitoring and Control of Power Preparedness of Athletes in Flatwater Rowing and Canoeing Using Strain Gauges
17

2.1.4 Data Processing
The processing of the obtained data was carried out
using specially developed software, which allows you
to automate the basic manipulative procedures usually
performed with such data (Figure 5).
Figure 5: Software screenshot for automated processing of
sensor data.
The software allows us to interactively designate
the boundaries of the recording segment of interest,
filter using a digital filter with a moving average (the
size of the filter window is also customizable and se-
lected by the user), and automatically designate the
boundaries of the beginning and end of the supporting
part of each stroke included within the previously des-
ignated limits recording segment, and also calculate
all the necessary rowing parameters. Current software
can process up to 4 signals simultaneously, and the
computing functionality allows it to be used to pro-
cess recordings of kayaking and canoeing. In partic-
ular, for canoeing it is necessary to calculate parame-
ters for each stroke, and for kayaking it is necessary
to calculate only for target strokes (Figure 6).
Figure 6: Interactive designation of the boundaries of the
supporting part of the strokes.
The software is implemented using the Python
programming language based on the public libraries
Pandas, Numpy and Matplotlib.
This study calculated the numerical values of the
average force (F
mean
) and peak force (F
max
) within
each stroke, as well as the ratio of average force to
peak (F
mean
/F
max
), which can quantitatively charac-
terize the density of the stroke. The numerical value
of this parameter ranges from 0 to 1. The higher the
value, the denser and better quality the stroke is in
terms of propulsive efficiency. In addition, some other
biomechanical parameters of rowing have been calcu-
lated. In particular, the time of the supporting part of
the stroke (t
sup
), cycle time (t
c
), as well as the tempo
of rowing (T). Also, for a better understanding of the
speed-force nature of rowing, the values of the im-
pulse (I) are calculated.
3 RESEARCH RESULTS
The results of each test task are presented in Table 1.
For each parameter, the standard deviation (SD) was
also calculated, which quantitatively characterizes the
stability of the rowing process.
According to the recorded data, the athlete has a
pronounced force asymmetry. In particular, the forces
developed by the left hand are greater than those de-
veloped by the right. This is a certain kind of motor
dysfunction, since it provokes the boat to turn to the
left, which the athlete is forced to compensate for by
steering through the steering mechanism of the boat.
This circumstance reduces the propulsive efficiency
of rowing. In addition, one can notice that the degree
of asymmetry of movements decreases with each sub-
sequent attempt, which indicates a slight increase in
the propulsive efficiency of rowing against the back-
ground of fatigue. Analyzing the numerical indicators
of the standard deviation, there is reason to assert that
the athlete is distinguished by high rowing stability,
since the standard deviation values do not exceed 5%
of the target indicator.
To assess the trustworthiness of the recorded and
calculated data, their statistical processing was car-
ried out by means of two-factor analysis of variance
with repetitions (ANOVA). Such an analysis makes it
possible to determine whether the differences in the
analyzed values are random. The numerical values of
the analysis are presented in Table 2.
The results of the analysis indicate statistically
significant differences in 4 of the 6 registered and
calculated parameters (p < 0.05). The p-values for
t
sup
and I parameters are statistically insignificant (the
probability of accidental differences is 56% and 23%,
respectively). Such a phenomenon, it seems to us, is
caused by the imperfection of the algorithm for au-
tomated marking of the boundaries of the beginning
and end of the supporting part of each stroke. The
improvement of this algorithm is a priority task in the
framework of future research activities.
icSPORTS 2024 - 12th International Conference on Sport Sciences Research and Technology Support
18

Table 1: Results of test tasks.
Trial Sensor
F
mean
, N F
max
, N F
mean
/F
max
t
sup
, s t
c
, T, min
−1
I, Ns
SD SD SD SD SD SD SD
1
Left
87.09 188.45 0.46 0.43 0.95 55.01 37.78
4.62 9.32 0.02 0.04 0.04 0.32 3.99
Right
114.37 211.56 0.54 0.43 0.94 52.81 47.35
7.30 12.92 0.04 0.06 0.05 0.40 6.78
2
Left
91.28 195.94 0.47 0.42 0.91 59.05 38.44
4.11 9.08 0.02 0.03 0.05 0.27 3.28
Right
112.73 204.11 0.55 0.44 0.93 56.84 47.76
6.94 11.70 0.04 0.04 0.04 0.36 5.50
3
Left
89.41 173.96 0.51 0.43 0.95 55.29 39.14
3.93 8.72 0.02 0.03 0.02 0.62 3.81
Right
107.62 199.15 0.54 0.44 0.94 54.84 45.43
4.24 9.13 0.03 0.04 0.04 0.73 4.27
Table 2: Results of test tasks.
Metrics F
mean
, N F
max
, N F
mean
/F
max
t
sip
, s t
c
, s I, Ns
p-value 0.0002 0.00007 0.000003 0.56 0.006 0.23
4 THE DISCUSSION OF THE
RESULTS
The development and implementation of various
wearable and mounted sensors and measuring sys-
tems as means of diagnosing and monitoring the pre-
paredness of athletes is a relevant and popular area
of research and inventive activity, as evidenced by
a large number of thematic publications (Warmen-
hoven, 2018; V
˙
e
ˇ
zys et al., 2020; L
¨
opp
¨
onen et al.,
2022; Annino et al., 2023; Cristian Romagnoli and
Gatta, 2022). However, the process of introducing
such sensors into sports, in which training and com-
petitive activities are carried out at the junction of air
and water environments, is accompanied by certain
difficulties.
In particular, there are problems of adapting the
design of sensors to the amplitude deformations of
equipment, as well as ensuring hardware resistance
to conditions of high humidity and immersion in wa-
ter (Cruz et al., 2023). However, it is noted that the
use of strain gauge elements in the base of the sensor
is the most preferable option for circuit implementa-
tion, since the electrical behavior of the strain gauge
remains stable both in dry conditions and in condi-
tions of high humidity (Laaraibi et al., 2024). In addi-
tion, there is information that the numerical data ob-
tained using strain gauges, provided that the measures
for their installation and calibration are followed, are
reliable and make it possible to record with a suffi-
cient degree of accuracy the mechanical stresses aris-
ing during rowing, as well as to objectively determine
the indicators characterizing power of movements of
athletes when interacting with the surface of the wa-
ter in the supporting part of the stroke (V
˙
e
ˇ
zys et al.,
2020).
It should be noted that the design of such sen-
sors and measurement systems must also be accom-
panied by necessary and sufficient ergonomic mea-
sures, which will improve mechanical strength and
electrical insulation, as well as eliminate various po-
tential movement restrictions that can reduce the re-
liability of the recorded data (Laaraibi et al., 2024;
Rana and Mittal, 2021). Based on the results of the
analysis of thematic publications, it was established
that in practice, within the framework of diagnos-
tics and monitoring of the readiness of rowing ath-
letes, both wired and wireless configurations of sen-
sors are applicable and useful. However, it should be
noted that for research work, as well as when working
with two or more devices, the most preferable method
is a wired data recording method, which allows for
high-frequency data recording without loss. A wire-
less configuration, in turn, may be more preferable
for providing prompt feedback, while this informa-
tion will be useful to the coach, but not to the athlete
himself, who is fully concentrated on performing the
motor task.
As mentioned earlier, in order to achieve high effi-
ciency in diagnostics and control of rowers’ readiness
at various stages of the training process, it is neces-
sary to determine the most informative control param-
eters. Such parameters that would allow one to judge,
to the necessary and sufficient extent, the technical
Monitoring and Control of Power Preparedness of Athletes in Flatwater Rowing and Canoeing Using Strain Gauges
19

and speed-strength readiness of athletes. It is known
that such parameters are the forces developed by the
athlete when interacting with the surface of the wa-
ter, as well as their derivatives, including the power
of movements, force impulse, force gradient and oth-
ers (Bonaiuto et al., 2020b; Bonaiuto et al., 2020a).
The use of sensors based on strain-resistive circuit el-
ements makes it possible to register and calculate the
numerical values of these and many other parameters
that quantitatively characterize the temporal features
of rowing.
Strictly speaking, to ensure regular monitoring, it
is very important to have the ability to digitally repre-
sent each stroke when an athlete performs target train-
ing and training-diagnostic tasks. This will allow us
to form an objective idea of the athlete’s level of pre-
paredness, as well as track the dynamics of his/her
results.
5 CONCLUSION
The work demonstrates an experimental device de-
veloped for technical equipment of diagnostic proce-
dures and monitoring of technical and speed-strength
readiness of athletes specializing in kayaking and ca-
noeing. An experimental testing of the device was
carried out under the conditions of the training pro-
cess. Empirical data recorded using the developed
device and processed using specialized software are
presented.
The results of the experimental testing of the de-
veloped strain gauge sensors and automated process-
ing software can be considered positive, since the sen-
sors themselves worked properly, the athlete did not
feel discomfort during the test tasks, and the dura-
tion of the data collection process within the training
and diagnostic process did not exceed the wishes of
the coach. Obviously, an additional series of experi-
ments is required to test devices and software on vari-
ous data. However, there are already grounds to assert
that such devices are necessary to ensure an effective
training process, especially for professional athletes.
To summarize, it should be noted that the effec-
tiveness of flatwater rowing and canoeing technique
can be determined quite accurately and objectively by
comparing individual movements with references val-
ues, establishing the relationship between individual
indicators of technique and sports results, as well as
regular monitoring of the dynamics of indicators. All
of the listed tasks in the field of kayaking and canoe-
ing can be solved through the use of similar technical
devices and appropriate software.
REFERENCES
Annino, G., Boatto, P., Bonaiuto, V., Campoli, F., Capri-
oli, L., Edriss, S., Lanotte, N., Padua, E., Panichi, E.,
and Romagnoli, C. (2023). A daq system suited for
olympic sprint canoeing performances monitoring. In
2023 IEEE International Workshop on Sport, Technol-
ogy and Research (STAR), pages 81–84.
Baker, J. (2012). Biomechanics of paddling.
Bertozzi, F., P. S. M. M. P. A. M. G. M. S. C. . Z. M. (2022).
Whole-body kinematics during a simulated sprint in
flat-water kayakers, volume 22.
Bonaiuto, V., Annino, G., Boatto, P., Lanotte, N., Capri-
oli, L., Padua, E., and Romagnoli, C. (2022). System
for Performance Assessment of K2 Crews in Flatwater
Sprint Kayak.
Bonaiuto, V., Gatta, G., Romagnoli, C., Boatto, P., Lan-
otte, N., and Annino, G. (2020a). A New Measurement
System for Performance Analysis in Flatwater Sprint
Kayaking, volume 49.
Bonaiuto, V., Gatta, G., Romagnoli, C., Boatto, P., Lan-
otte, N., and Annino, G. (2020b). A Pilot Study on
the e-Kayak System: A Wireless DAQ Suited for Per-
formance Analysis in Flatwater Sprint Kayaks, vol-
ume 20.
Brown, M., L. M. . D. R. (2010). Activation and contribu-
tion of trunk and leg musculature to force production
during on-water sprint kayak performance.
Cristian Romagnoli, Massimiliano Ditroilo, V. B. G. A. and
Gatta, G. (2022). Paddle propulsive force and power
balance: a new approach to performance assessment
in flatwater kayaking. Sports Biomechanics, 0(0):1–
14.
Cruz, M., Gomes, B., Silva, M., Amaro, A. M., and Ro-
seiro, L. (2023). Use of the paddle and oar instru-
mented as a structural element for quantification of the
exercised force—pre-study of strain gauge behavior
with simulation in dry and wet environments. In Mar-
tins Amaro, A., Roseiro, L., Messias, A. L., Gomes,
B., Almeida, H., Ant
´
onio Castro, M., Neto, M. A.,
de F
´
atima Paulino, M., and Maranha, V., editors, Pro-
ceedings of the 10th Congress of the Portuguese Soci-
ety of Biomechanics, pages 453–462, Cham. Springer
Nature Switzerland.
Galipeau, C. (2018). The On-water Instrumentation of a
Sprint Canoe Paddle.
Gomes, B. B., R. N. V. C. F. S. R. V. M. . V.-B. J. P.
(2022). Paddling time parameters and paddling ef-
ficiency with the increase in stroke rate in kayaking,
volume 21.
Guseinov D.I., Permyakov T.V., N. A. L. D. M. A. (2024).
Technologies for measuring the dynamic parameters
of rowing based on strain gauge systems. Russian
Journal of Biomechanics, 28(2):95–104.
Kolumbet, A. N. (2017). Dynamic of kayak rowing tech-
nique in the process of competition activity. Num-
ber 4. London, 2nd edition.
Kvashuk, P. V., V. A. V. S. G. N. . M. I. N. (2021). Bene-
fits of specific strength training model with water re-
icSPORTS 2024 - 12th International Conference on Sport Sciences Research and Technology Support
20

sistance control gear for rowing and canoeing sports
elite. Number 9.
Laaraibi, A.-R. A., Jodin, G., Depontailler, C., Bideau, N.,
and Razan, F. (2024). Design and characterization
of piezoresistive sensors for non-planar surfaces and
pressure mapping: A case study on kayak paddle. Sen-
sors, 24(1).
L
¨
opp
¨
onen, A., V
¨
anttinen, T., Haverinen, M., and Linnamo,
V. (2022). The Effect of Paddle Stroke Variables Mea-
sured by Trainesense SmartPaddle® on the Velocity of
the Kayak, volume 22.
Oronova, D., H. O. G. N. . H. R. (2018). Research of speed-
strength qualities of specific muscle groups in rowers,
volume 2.
Rana, M. and Mittal, V. (2021). Wearable sensors for real-
time kinematics analysis in sports: A review. IEEE
Sensors Journal, 21(2):1187–1207.
Redwood-Brown, A. J., B. H. L. O. B. . F. P. J. (2021).
Determinants of boat velocity during a 200 m race in
elite paralympic sprint kayakers, volume 21.
Rosdahl, H., C. J. S. A. W. . B. R. (2019). Physiology of
canoeing. International Olympic Committee.
V
˙
e
ˇ
zys, J., Lukashevich, D., Huseynov, D., Minchenya, A.,
and Bubulis, A. (2020). Smart sensors for estimation
of power interaction of an athlete with water surface
when paddling in the cycle of rowing locomotions,
volume 3. JVE International Ltd.
Wainwright, B., C. C. . L. C. (2015). Performance related
technique factors in Olympic Sprint kayaking.
Wainwright, B., C. C. B. . L. C. (2014). A deterministic
model for Olympic Sprint kayaking. Number 32.
Warmenhoven, J., C. S. D. C. . S. R. (2018). Over 50 years
of researching force profiles in rowing: what do we
know?, volume 48.
Monitoring and Control of Power Preparedness of Athletes in Flatwater Rowing and Canoeing Using Strain Gauges
21