Method for generating a velocity-slope-time model characterizing a physical activity of a participant and associated method of use

Abstract

Title: Method for generating a speed-slope-time model characterizing a physical activity of a participant and associated method of use. According to a first aspect, the invention provides a method for generating a speed-slope-time model characterizing, for a given participant and a human-locomotion endurance physical activity, such as trail running, and a method for using the speed-slope-time model thus generated. The technical effect of the invention is to allow the individualization and use of the concept of critical speed for a participant in a physical activity, taking into account, at each instant, the slope, altitude, and type, or even conditions, of terrain, to indicate to the participant, in real time, during the physical activity, the critical speed not to be exceeded in order to optimize their performance. Figure for the abstract: Fig. 9.

Description

Title of the invention: Method for generating a speed-slope-time model characterizing a physical activity of a practitioner and associated method of use Technical field

[0001] The present invention relates to the field of endurance physical activities involving human locomotion. It relates more particularly to a method for generating a speed-slope-time model characterizing a physical activity of a practitioner and an associated method of use. STATE OF THE ART

[0002] Trail running (or 'mountain running') is one of the endurance activities involving human locomotion, such as cycling, rowing, cross-country skiing, ski touring or even swimming for example. These activities are closely linked to the ability of its practitioner to produce and maintain power, mechanically defined as the product of force by speed, in order to perform a movement. In cycling for example, power is obtained by applying a torque at the crankset at a certain angular speed (i.e. pedaling cadence). A change in gear ratio (ratio between chainrings and sprockets) makes it possible to maintain an optimal pedaling cadence whether on the flat or on a slope depending on the desired speed or power. The same goes for rowing where the length of the oars can be modified to vary the rowing cadence.On the other hand, in trail running, the slope has a greater impact since the runner has less room to maneuver regarding the adaptation of his cadence and stride length. For a given speed, an increase in force is therefore necessary when the slope increases.

[0003] Each practitioner of a physical activity, with human locomotion, is characterized by a force-speed-time relationship which presents a great inter-individual variability. Note that this variability manifests itself when the mechanical conditions of locomotion are variable (eg by the slope in trail) while it remains little visible for flat running, such as a marathon.

[0004] In the field of human locomotion endurance activities, including flat running, the concept of critical power or speed is widely used. It is both an indicator of a participant's endurance capabilities, but also useful information for race management strategy, in particular by making it possible to avoid acute fatigue phenomena by continuously managing the speed of movement during physical activity.

[0005] However, although methods and technologies are today effective for the analysis of running on the flat, they do not take into account the specificities linked to other physical activities of endurance with human locomotion, such as trail running. These specificities include the slope, the altitude, and the terrain conditions (technicality and humidity for example), which are particularly, but not necessarily only, linked to trail running, and which can vary from one section of the course to another.

[0006] Some solutions use a correction factor for the slope, referring to the equivalent flat speed. However, this type of correction is often based on an average of data taken from a set of individuals, which, given the high inter-individual variability mentioned above, does not work (or works poorly) for most users.

[0007] The present invention aims precisely to allow these specificities (slope, altitude, type / condition(s) of terrain) to be taken into account, in order to allow each practitioner of an endurance activity involving human locomotion to benefit from the concept of “critical speed” for each force condition imposed by the environment (for example the slope in trail running).

[0008] There are several solutions, but these all have limitations that the present invention aims to overcome.

[0009] For example, the solution developed by Garmin® and called PacePro™ has consisted, since November 2019, of a feature integrated into certain GPS watch models. User feedback seems mixed regarding the paces indicated by the watch and that the runner is asked to respect. Indeed, segments of the course are automatically defined by the watch, or by the user, and are used to calculate an average pace to respect on these segments. From the same GPS watch manufacturer, a feature called Stamina is known which has been integrated into certain Garmin® watch models since January 2022. It offers the runner an indicator which is similar to an energy reserve. The total capacity of this reserve is estimated based on the runner's training data.The latter can then manage his running pace so as not to completely exhaust his reserve before the end of the race, such exhaustion often being synonymous with a significant slowdown in speed, or even abandonment, but no indication of the pace to be respected is offered; and the user must therefore manage his pace and evaluate whether what he is told remains of his energy reserve will be enough to reach the finish line. A priori, these solutions do not allow for taking into account either the altitude, or the type or conditions of terrain.

[0010] For another example, the company Coros®, another manufacturer of GPS watches, has improved its “Effort Pace®” functionality for estimating equivalent flat speed. The latter can now take into account the individual profile of each runner based on their training data. This allows for real-time, depending on of the slope, the equivalent speed on the flat of the runner. However, this functionality, even improved, does not allow for the management of certain specificities such as the type or conditions of terrain or altitude. In addition, the concept of critical speed is not applied; the runner therefore benefits from an indication of equivalent speed on the flat, but must know, by himself, at what speed he must run, relative to this indication, depending on the conditions of the terrain and his fatigue during the race, to manage his strategy.

[0011] For another example, the company RunMotion® offers an application for road and mountain running allowing the generation of personalized training plans. However, if this application allows the management of the training aspect which precedes a race, it does not aim to build a race plan or to guide the participant, in real time, with, among other things, a recommendation of pace to be respected depending on the type of terrain.Part of the application seems to aim to offer the runner a performance prediction, but only for road races. The mathematical model used to predict a runner's performance has been the subject of a scientific publication ("A minimal power model for human running performance", Matthew Mulligan, Guillaume Adam, Thorsten Emig, November 16, 2018), with one of the founders of RunMotion®, Guillaume Adam, as co-author.

[0012] It should also be noted that the STRAVA© platform offers users a feature that estimates an equivalent flat speed in post-analysis of a human-powered endurance physical activity, and in particular a trail. However, this feature does not offer real-time guidance of the equivalent flat speed, nor does it take into account the altitude and the type or conditions of terrain. In addition, to estimate the equivalent flat speed, an average of the physical capacities of the runners registered on the platform is used rather than the individual physical capacities of each runner.

[0013] An objective of the present invention consists in providing, in real time, to a practitioner of a physical activity of endurance with human locomotion, such as a trail runner (or 'mountain running'), a speed of movement to be respected, thus allowing him to optimize his race management according to his physical capacities, and the route presented to him, this route being characterized by a slope, an altitude and a type of terrain, possibly terrain conditions, and thus limit the appearance of fatigue. SUMMARY

[0014] To achieve this objective, according to a first aspect of the invention, a method is provided for generating, or calibrating, or adjusting, a speed-slope-time model characterizing a human locomotion endurance physical activity of a practitioner. The method according to the first aspect of the invention is essentially as it includes the following steps: a. receive a set of geolocation data representative of the latitude, longitude and altitude of the practitioner at different times during at least one session of said physical activity, b. process at least part of all the geolocation data received to calculate representative data, per unit of length chosen, of a slope profile covered by the user during each session, c. based on at least part of all the geolocation data received and the data representative of the slopes previously calculated, calculate an average slope and an average speed of the practitioner for each of a plurality of predefined target durations, d. based on the calculated average slopes and average speeds, deriving (or selecting) a maximum average speed for each target duration from the plurality of predefined target durations and each from a plurality of predefined target slopes, and e. based on the deduced maximum average speeds, generate the speed-slope-time model of the practitioner by determining a plurality of parameters on which the speed-slope-time model depends, by adjusting said model to the maximum average speeds

[0015] According to a second aspect of the invention, a method is provided for using a speed-slope-time model characterizing a human locomotion endurance physical activity of a practitioner. The method according to the second aspect of the invention is essentially such that it comprises the following steps: a. receive coefficients defining, for the practitioner, his speed-slope-time model, b. receive a matrix of tiles comprising a route that the practitioner proposes to follow during a practice session of said upcoming physical activity, then c. at a plurality of times, preferably at each time, during the session: i. receive geolocation data representative of a latitude, longitude and altitude of the practitioner, ii. based on at least part of the geolocation data received, calculate a slope traveled by the user, iii. based on at least part of the geolocation data received, calculate a speed of the user, iv. based on at least part of the geolocation data received, estimate the type of terrain covered by the user, or even the terrain conditions during the session, v. based on the calculated slope and speed, determine using the speed-slope-time model and the coefficients received, a critical speed not to be exceeded by the practitioner, vi. depending on the estimated type of terrain, or even the terrain conditions encountered, apply a correction coefficient to the determined critical speed, if, for a period greater than a predefined threshold value, the calculated speed of the practitioner is greater than the critical speed corrected by the application of the correction coefficient, indicate this to the practitioner.

[0016] A third aspect of the invention relates to a computer program product, preferably recorded on a non-transitory medium, comprising instructions, which when carried out by at least one of a processor and a computer, cause the at least one of the processor and the computer to execute the generation method according to the first aspect of the invention and / or the use method according to the second aspect of the invention.

[0017] The invention as introduced above has the technical effect of allowing the individualization and use of the concept of critical force-speed of a practitioner (individual capacity) of a physical activity, taking into account, at each instant, the environmental conditions of slope, altitude and the type, or even the conditions (for example wet or dry), of terrain, to indicate to the practitioner, in real time, during the physical activity, the speed which will optimize performance and limit the appearance of fatigue.

[0018] The main advantages resulting from this include: a. the generation and visualization of the speed-slope-time relationship characterizing the practitioner, allowing the identification of areas for improvement with a view to adapting one's training or choice of races, for example. Weaknesses may be a lack of endurance over long activities, or a lack of strength that is penalizing on the steepest slopes; and / or b. personalized and real-time indication, during physical activity, of the speed to be respected, and the characteristics of the race (slope, type of terrain (or technicality), altitude, or even terrain conditions (rain or good weather)); the “real-time” aspect is particularly interesting given the highly variable terrain conditions during certain physical activities, such as trail running; and / or c. the ability to warn the participant of exceeding their target speed which, after only a few minutes of activity, could generate significant fatigue and lead to a sharp decrease in the participant's physical performance, sometimes synonymous with a longer race time, or even a abandonment in certain cases; and / or d. prevention of injury to the practitioner which may be linked to an effort which would be greater than that corresponding to his target speed and which would be sustained for too long a period, such an injury being synonymous with abandonment in certain cases; and / or e. the reduction of the dropout rate during a race; this aspect is interesting for the organizers of a sporting event since it reduces repatriation logistics and increases the satisfaction rate of participants. BRIEF DESCRIPTION OF THE FIGURES

[0019] The aims, objects, as well as the characteristics and advantages of the invention will emerge more clearly from the detailed description of an embodiment thereof which is illustrated by the following accompanying drawings in which:

[0020] [Fig.l] [Fig.l] graphically represents an example of a speed-slope-time model adjusted to a practitioner by an implementation of the method according to the first aspect of the invention and / or used by a practitioner during his physical activity of endurance with human locomotion by an implementation of the method according to the second aspect of the invention.

[0021] [Fig.2] [Fig.2] graphically represents the measurement points illustrated in [Fig.l] so as to better illustrate the influence of the duration of the effort provided on the speed of movement of the practitioner.

[0022] [Fig.3] [Fig.3] graphically represents the measurement points illustrated in [Fig.l] so as to better illustrate the influence of the slope on the speed of movement of the practitioner.

[0023] [Fig.4] [Fig.4] graphically represents the known relationship between the level of a practitioner's ability to exert effort and the altitude at which the practitioner is located.

[0024] [Fig.5] [Fig.5] represents a capture of a screen via which the practitioner can be kept informed of an exceeding of his critical speed by an implementation of the method according to the second aspect of the invention.

[0025] [Fig.6] [Fig.6] graphically represents an example of the evolution of the speed of movement of a practitioner and that of his critical speed as a function of time, during a physical activity of endurance with human locomotion.

[0026] [Fig.7] [Fig.7], to be read in conjunction with [Fig.6], graphically represents when, by an implementation of the method according to the second aspect of the invention, an indication of exceeding its critical speed is indicated to the practitioner during a physical activity of endurance with human locomotion, and to which of two levels of importance (noted "light" and "strong") this indication is associated.

[0027] [Fig-8] [Fig.8] is a flowchart illustrating a mode of implementation of the method according to the first aspect of the invention.

[0028] [Fig.9] [Fig.9] is a flowchart illustrating an embodiment of the method according to the second aspect of the invention.

[0029] The drawings are given as examples and are not limiting of the invention. They constitute schematic representations of principle intended to facilitate the understanding of the invention and are not necessarily on the scale of practical applications. In particular, the values ​​indicated on the different axes of the illustrated graphs are not necessarily representative of reality. DETAILED DESCRIPTION

[0030] Before beginning a detailed review of embodiments of the invention, optional features which may possibly be used in combination or alternatively are set out below:

[0031] According to an example of the first aspect of the invention, the human locomotion endurance physical activity comprises at least one of running, and in particular mountain running (or trail running), cycling, cross-country skiing, ski touring, etc. Where appropriate, the slope criterion is then replaced by a criterion adapted to the physical activity considered. For example, for cycling, the criterion may be defined as a function of the ratios between chainrings and sprockets and / or the pedaling speed.

[0032] According to another example of the first aspect of the invention, the geolocation data was collected using a GPS watch, a smartphone, or a third-party application, for example offered via an application platform, the geolocation data is where appropriate received from the GPS watch, the smartphone, or a server hosting the third-party application having collected the data.

[0033] According to another example of the first aspect of the invention, the generation method according to the first aspect of the invention is implemented by a computer server.

[0034] According to another example of the first aspect of the invention, before the processing step, the geolocation data are received in JSON format, GPX format or FIT format, then converted into matrix form, or are directly received in matrix form. According to this example, the processing of the geolocation data in the following steps of the generation method according to the first aspect of the invention is advantageously facilitated.

[0035] According to another example of the first aspect of the invention, the data representative of the slope profile traveled by the practitioner during each session are calculated as a derivative per unit length of the geolocation data representative of the altitude.

[0036] According to another example of the first aspect of the invention, at least part of the set of geolocation data used to calculate the average slopes and speeds of the user for the predefined target durations includes geolocation data representing the latitude and longitude of the user at different times during each session.

[0037] According to another example of the first aspect of the invention, the unit of length is chosen to be between 20 cm and 1 km, and is for example substantially equal to 1 m.

[0038] According to another example of the first aspect of the invention, the geolocation data being defined in a spherical reference frame, the step of processing the geolocation data considered comprises, before the calculation of the data representative of the slope profile: a. a transformation of the geolocation data by a change of reference from the spherical reference to a Cartesian reference, then b. a first interpolation, for example linear, on the transformed geolocation data, at a sampling frequency of between 0.02 and 10 Hz, preferably substantially equal to 1 Hz, to obtain temporally sampled data, then c. a second interpolation, for example linear, on the transformed data defined on the time base, over a sampling distance of between 0.1 and 500 m, preferably substantially equal to 1 m, to obtain geographically sampled data, so that the calculation of the representative data of the slope profile is a function of the geographically sampled data. According to this example, the calculation of the representative data of the slope profile is facilitated at least in terms of numerical implementation.

[0039] According to another example of the first aspect of the invention, the method further comprises, after the first interpolation and before the second interpolation, a deletion, from the temporally sampled data, of those which do not exist in the data sampled before the first interpolation, and for which the elapsed time is greater than a threshold value of between 1 second and 1 minute, preferably substantially equal to 10 seconds. According to this example, it is advantageous to avoid taking into account, for the subsequent calculations, geolocation data which reveal a (micro-)pause of the practitioner during a session or which are to be differentiated from (micro-)pauses and which are part of an optimization of the size of the data recorded by the watch and the consumption of its battery.

[0040] According to another example of the first aspect of the invention, the method further comprises, after the first interpolation and before the second interpolation, the application of a low-pass filter to the geolocation data representative of the altitude, in order to smooth out any measurement inaccuracies. According to this example, this avoids to use, for subsequent calculations, measurements which could distort them.

[0041] According to another example of the first aspect of the invention, the method further comprises, after the calculation of the data representative of the slope profile, a third interpolation, for example linear, on the data representative of the slope profile, at a sampling frequency of between 0.02 and 10 Hz, preferably substantially equal to 1 Hz, so that the calculation of the average speeds is a function of the data representative of the slope profile calculated and sampled in time. According to this example, the subsequent calculations are facilitated at least in terms of digital implementation.

[0042] According to another example of the first aspect of the invention, the calculation of the data representative of the slope profile is followed, preferably immediately, by the application of a low-pass filter to said data representative of the slope profile. According to this example, it is thus avoided to consider, for the subsequent calculations, slope measurements, potentially representative of aberrations, which could distort the generation of the speed-slope-time model.

[0043] According to another example of the first aspect of the invention, the target durations are predefined and included in a set of intervals ranging from a minimum value between 1 and 600 seconds, preferably substantially equal to 120 seconds, to a maximum value between 6 minutes and 6 hours, preferably substantially equal to 14 minutes, the intervals of the set, preferably regular and / or adjacent to each other, taking a value between 1 second and 10 minutes, preferably substantially equal to 2 minutes.

[0044] According to another example of the first aspect of the invention, the target slopes are predefined and included in a set of intervals ranging from a minimum value between -10% and -60%, preferably substantially equal to -50%, to a maximum value between 10% and 60%, preferably substantially equal to 50%, the intervals of the set, preferably regular and / or adjacent to each other, taking a value between 0.1 and 10%, preferably substantially equal to 2%.

[0045] According to the previous example, the step of calculating the average slopes and speeds is implemented, for each target duration, by sliding averages. This saves on calculation time and central processing unit (CPU) resources.

[0046] According to another example of the first aspect of the invention, the step of deducing the maximum average speed for each target duration is preceded by a deletion, for each target duration and each target slope, from the calculated average slopes and average speeds, of data representative of slopes not having a majority of values, for example at least 80% of values, located in the corresponding target slope, preferably with a tolerance substantially equal to + / - 2.5%.

[0047] According to another example of the first aspect of the invention, the step of deducing the maximum average speed for each target duration and each target slope further comprises a deletion, from the deduced maximum average speeds, of maximum average speeds substantially less than 0.5 m / s and substantially greater than 7 m / s. According to this example, it is advantageously avoided to take into account, for the remainder of the generation method according to the first aspect of the invention, maximum average speeds whose validity is uncertain.

[0048] According to another example of the first aspect of the invention, the generation method comprises, after the step of deducing the maximum average speed for each target duration and each target slope, a recording on a non-transitory medium of the deduced maximum average speeds with the corresponding target durations and slopes.

[0049] According to another example of the first aspect of the invention, the step of generating the speed-slope-time model of the practitioner comprises at least the step of constructing a two-dimensional matrix containing the maximum average speeds per target duration and target slope, so that the adjustment of the speed-slope-time model is carried out on the basis of the matrix thus constructed.

[0050] According to another example of the first aspect of the invention, the adjustment of the speed-slope-time model is implemented by implementing a method of minimizing the residuals of an equation p) defining the speed-slope-time model, where V is the maximum average speed, f is the time and P is the slope.

[0051] According to another example of the first aspect of the invention, the velocity-slope-time model is defined by an equation V(t, p) taking the following form: V(t, p) = D / 1+V£p), where D is a reserve of distance achievable above the critical speed (in meters), f is the duration of an effort made (in seconds), P is the slope (in %), and whose limit, when the duration of the effort made t tends towards infinity, defines a critical speed-slope relationship, noted VL(p), characterizing the practitioner.

[0052] According to another example of the first aspect of the invention, the speed-slope-time model is defined in part by a critical speed-slope relationship Vc(p) taking the form of the inverse of a polynomial of at least the second degree, having as variable the slope p and whose parameters, noted for example a, b and c, are determinable by adjustment of the equation V(t, p) defining the speed-slope-time model on the basis of the deduced maximum average speeds.

[0053] According to another example of the first aspect of the invention, the critical speed-slope relationship V^p) takes the form of the inverse of a second-degree polynomial: Vc(p) = 1 / (ax p2 + bx p + c) where the parameters a, b and c are determinable (or determined) by adjusting the critical speed-slope relationship V / p) on the basis of the maximum average speeds deduced.

[0054] According to an example of the second aspect of the invention, the speed-slope-time model was generated by implementing a method for generating a speed-slope-time model characterizing a human locomotion endurance physical activity of a practitioner according to the first aspect of the invention.

[0055] According to another example of the second aspect of the invention, each tile is defined by a surface, for example square, of the geographical area in which the session of practicing said physical activity takes place or is intended to take place.

[0056] According to another example of the second aspect of the invention, the geolocation data is received at a frequency substantially equal to 1 Hz.

[0057] According to another example of the second aspect of the invention, further comprising, before the steps of calculating the slope and the speed of the practitioner, the application of an infinite impulse response (IIR) filter on the received geolocation data. According to this example, the method of use according to the second aspect of the invention makes it possible to process all of the geolocation data without a size being imposed on this data or whatever the size of this data, and therefore makes it possible to filter the geolocation data without knowing their size a priori.

[0058] According to another example of the second aspect of the invention, the estimation of the type of terrain covered comprises: a. the identification of the tile corresponding to the geolocation of the practitioner, b. for each of the paths listed on the identified tile, the calculation of the distance between the geolocation of the practitioner and said path, and c. the recovery of characterization data of the path closest to the geolocation of the practitioner.

[0059] According to another example of the second aspect of the invention, the correction coefficient to be applied is determined as a function of the recovered characterization data.

[0060] According to another example of the second aspect of the invention, the correction coefficient to be applied is substantially between 0.5 and 1, the higher the technicality of the type of terrain traveled, the lower the correction coefficient, and / or in which the application of the correction coefficient consists of multiplying the critical speed determined by the correction coefficient.

[0061] According to another example of the second aspect of the invention, the determination, using the speed-slope-time model and the received coefficients, of the critical speed not to be exceeded comprises the calculation of the critical speed according to the following equation: Vc = ( 1 - Coeffvo2max xh) / (axp^ + bx p+where: a. h is the altitude of the practitioner, in meters; b. P is the slope traveled, in %; c. a, b, and c are the received coefficients, and d. Coej j vo2max is a coefficient reflecting the reduction in the physical capacities of the practitioner as a function of altitude, this coefficient is for example approximately equal to 6x105.

[0062] According to another example of the second aspect of the invention, the indication given to the practitioner is of a first level of importance if the calculated speed of the practitioner is greater than the critical speed corrected by the application of the correction coefficient for a duration greater than a first predefined threshold value and less than a second predefined threshold value, and of a second level of importance, greater than the first, if the calculated speed of the practitioner is greater than the critical speed corrected by the application of the correction coefficient for a duration greater than the second predefined threshold value, the second predefined threshold value being greater, for example substantially twice greater, than the first predefined threshold value.

[0063] A parameter “substantially equal / greater / less than” a given value means that this parameter is equal / greater / less than the given value, plus or minus 20%, or even 10%, close to this value. A parameter “substantially between” two given values ​​means that this parameter is at least equal to the smallest given value, plus or minus 20%, or even 10%, close to this value, and at most equal to the largest given value, plus or minus 20%, or even 10%, close to this value.

[0064] A preferred embodiment of the invention is described below, with reference to the appended figures. More particularly, the embodiment described below constitutes a sports monitoring system applied to trail running (mountain running), it being understood that the different aspects of the invention are not limited to an application to this physical activity.

[0065] With reference to Figures 8 and 9, the generation 100 and use 200 methods according to the first two aspects of the invention can be developed in the form of an application for GPS watch and smartphone. They allow a trail runner to provide him, in real time, with a speed to be respected to limit his fatigue, to optimize his race management according to his physical capacities, and the route presented to him, characterized by a slope, an altitude, and a type or conditions of terrain.

[0066] The generation 100 and use 200 methods described below in the context of trail running can be considered as based on the use of the concept of a runner's critical speed applied to trail running, taking into account the slope, the altitude and the type or conditions of terrain, for each instant of time during an activity.

[0067] The main advantages of applying this concept of critical speed in trail running include at least one of the following: a. Obtain an indication, in real time during the race, of the critical speed to be respected according to the physical condition of the runner which can evolve over the course of his training or within an activity itself, and the characteristics of the race (slope, type of terrain (or technicality), altitude, terrain conditions). The real-time aspect is particularly interesting given the very variable terrain conditions in trail running; b. Avoid exceeding your critical speed, which would generate significant fatigue after only a few minutes and lead to a significant decrease in physical performance, resulting in a longer race time or even abandonment in some cases; c. Prevent injuries that may be linked to an effort above its critical speed sustained for too long a period, also synonymous with abandonment in certain cases; and d. Reduce the dropout rate during a race, an interesting aspect for organizers since this reduces repatriation logistics and increases the satisfaction rate of runners

[0068] The generation method 100 according to the first aspect of the invention is notably illustrated by the flowchart of [Fig.8]. This flowchart translates an algorithm capable of determining the speed-slope-time relationship of a trail runner. The input data of the algorithm are GPS trail training data. They are not necessarily measured / recorded by the invention, but can be recovered from third-party software or devices, for example those mentioned in the introduction. This first aspect of the invention makes it possible, from these GPS data, to establish, for the runner, their individual speed-slope-time profile.

[0069] To do this, the speed of movement at each instant of the practitioner is calculated from the derivative of the position with respect to time. In addition, the type of terrain (trail, carriageway, asphalt), the ground conditions (wet or dry for example), the slope, as well as the altitude are associated with each GPS position of the practitioner during each of his training sessions.

[0070] GPS data from several months, for example the last 12 months, are considered in order to calculate the maximum average speed for each time condition (between 10 seconds and 1 hour; or even several hours) and slope (between -50% and +50%).

[0071] A mathematical model of a speed-slope-time model is then adjusted with the maximum average speeds recovered for each of the conditions, slope and duration. From this data and the files of the track of a competition (defined in GPX format for example), the method of using 200 the second aspect of the invention, notably illustrated by the flowchart of [Fig.9], determines the maximum speed at which the participant must run on each portion of the race in order to maximize his performance and limit his fatigue.

[0072] This model is also scalable in order to correspond to the physical fitness of the runner which evolves over time, whether upwards or downwards. When new training data is received 110, the model can be recalculated / readjusted.

[0073] Once adjusted, the practitioner's speed-slope-time model can then be integrated into an application for GPS watches and smartphones. The application retrieves real-time data from GPS positions, including altitude, during an activity in order to calculate, in real time, the movement speed 206 and the slope 205, from the derivative of the position and altitude with respect to time.

[0074] The critical speed not to be exceeded under current conditions (slope, type of terrain (or technicality), terrain conditions, altitude) is then calculated 208 using the model.

[0075] Indications for respecting this running speed are delivered 210 to the runner from notifications emitted by his GPS watch or his smartphone, the notifications taking, for example, the form of sound and / or light signals, vibrations, etc. The screen of the GPS watch or the smartphone can also indicate the current speed and target speed, and / or their deviation.

[0076] Method for generating the model using training data

[0077] The method 100 for generating the speed-slope-time model 1 (Cf. [Fig.l]) characterizing, for the practitioner, his mountain race, according to an embodiment of the first aspect of the invention, can take place on a computer server, for example when a future user registers on the platform. His speed-slope-time model will be modeled using his previously recorded training data. The generated model 100 can then be used by the smartphone application and GPS watch during a trail activity to guide the user on the speed to be respected.

[0078] Retrieving input data

[0079] The geolocation data (or GNSS data, for example GPS data) comprising longitude, latitude and altitude and time data of a trail activity are received 110 and extracted from input files originating for example from a GPS watch, a smartphone, or a third-party application already used by the practitioner.

[0080] Only files less than a year old can be kept, so that the profile is representative of the practitioner's fitness level.

[0081] The GNSS data contained in a structured computer file, such as .fit, .gpx or .json files for example, can advantageously be converted into the form of matrices, so as to facilitate their processing 120 in the following steps.

[0082] Data processing

[0083] Once received 110, the GPS data are processed 120. This processing can take different forms including that described below which, although preferred, is not a priori limiting of the generation method 100 according to the first aspect of the invention. In particular, in the flowcharts illustrated in Figures 8 and 9, the steps indicated by a dotted box are not considered essential.

[0084] Treatment 120 according to its preferred version comprises: a. The transformation 121 of GPS data defined in a spherical reference frame into Cartesian data (x,y,z; metric); b. A linear (re-)interpolation 122 of the data resulting from the transformation 121 at a frequency substantially between 0.02 and 10 Hz, preferably substantially equal to 1 Hz; c. The deletion 123 of the interpolated data 122 when a pause in the activity is detected. The data resulting from the transformation 121 may, before linear (re-)interpolation 122, be sampled at a frequency lower than 1 Hz, in particular for the sake of saving resources (battery or storage capacity on non-transitory media of the device used). A lack of data over a duration greater than a predefined threshold is then considered as a pause d. Applying a low-pass filter to the altitude data to smooth out inaccuracies in measuring devices that may distort speed and slope calculations in subsequent steps; e. The linear (re-)interpolation 125 of the data from the previous step 124, every meter in order to have data on a distance basis rather than a time basis for the following calculations; f. Calculation of the slope 126 traveled as the derivative of altitude with respect to distance; g. Applying 127 a low-pass filter to the slope data from the previous step; and h. Linear (re-)interpolation 128 of the data at a frequency of 1Hz.

[0085] Identification of record speeds by slope and duration

[0086] The target durations and slopes considered by the algorithm are respectively between: a. 10 seconds and 1 hour, or even several hours, in regular increments of seconds, and b. -50% and 50%, in regular increments of 0.5%.

[0087] For each target duration, the calculation 130 of the average slope and the average speed at each second of the activity is advantageously applied to the data from the step by implementing a sliding average (to save processing time or CPU time, and benefit from the non-necessarily finite side of the set of input data of this type of average, hence the evolutionary aspect of the algorithm).

[0088] Then, for each duration and target slope, the calculation 130 of the average slope and the average speed at each second of the activity can be immediately followed by a deletion 131 of the data which are not included in the predefined tolerances: a majority, for example 80%, of the slope included in the target slope with a tolerance of a few percent, for example a tolerance of + / -2.5%.

[0089] This is followed by a step, carried out on the basis of the calculated average slopes and average speeds 130, of deduction 140, in particular by selection, of a maximum average speed 11 (see figures 1, 2 and 3) for each target duration from among the plurality of predefined target durations and each from among a plurality of predefined target slopes.

[0090] When recovering 130 the maximum average speed for each slope and duration condition, only a speed slightly greater than Om / s, for example substantially greater than or equal to 0.5m / s, and substantially less than or equal to 10m / s is considered valid for the following steps. The maximum average speed values ​​which do not meet these conditions are advantageously deleted 141.

[0091] Identification of record speeds by slope and duration

[0092] The simplified example below illustrates the deduction step 140 by selecting the maximum average speeds 11 as a function of three target durations (2, 6 and 12 minutes) and three target slopes (10, 0 and -10%), for data from two activities, noted A and B, with the record speeds for the different duration and slope conditions:

[0093] Activity A:

[0094] [Tables 1] 2 min 6 min 12 min 10% 3 m / s 2 m / s 1 m / s 0% 5 m / s 4 m / s 4 m / s -10% 6 m / s 5 m / s 4 m / s

[0095] Activity B:

[0096] [Tables2] 2 min 6 min 12 min 10% 4 m / s 3 m / s 2 m / s 0% 6 m / s 5 m / s 3 m / s -10% 5 m / s 4 m / s 3 m / s

[0097] The deduction step 140 by selection of the maximum average speeds 11 consists of extracting the record speeds for each condition and results, from the example below, in the result presented in the following table, where the best records of each of the activities A and B have been selected for each condition of duration and slope:

[0098] [Tables3] 2 min 6 min 12 min 10% 4 m / s 3 m / s 2 m / s 0% 6 m / s 5 m / s 4 m / s -10% 6 m / s 5 m / s 4 m / s

[0099] The generation method 100 according to the first aspect of the invention can then comprise, after the step of deducing the maximum average speed 11 for each target duration and each target slope, a recording on a non-transitory medium of the maximum average speeds 11 deduced with the corresponding target durations and slopes.

[0100] It is also therefore possible to graphically project the maximum average speeds 11 deduced in a three-axis reference frame: slope P, time f and critical speed Vc, in the manner illustrated in [Fig.2], so as to better illustrate the influence of the duration of the effort provided on the speed of movement of the practitioner, or in the manner illustrated in [Fig.3], so as to better illustrate the influence of the slope on the speed of movement of the practitioner. We observe, in [Fig.2], a significant decrease in the average speed with an increase in the duration of effort. We observe, in [Fig.3], a significant decrease in the average speed with the absolute value of the slope.

[0101] Construction of the model

[0102] The generation 100 of the speed-slope-time model 1 according to the embodiment of the invention described herein, can then comprise: a. Constructing 151 a two-dimensional matrix containing the maximum average speed 11 for each target duration and slope from all the provided training data; and b. Based on the previously constructed matrix 151, the adjustment 152 of the model defined by the following equation, by the method of minimizing the residuals of the model V(t, p), where V is the maximum average speed; f is the time; P is the slope, where [Math.l] / À- dI t+ Or : • D = reserve distance achievable above the critical speed in meters • 1 = duration of the effort made in seconds • a, b, and c = unitless coefficients of the second-degree polynomial describing the critical speed-slope relationship, obtained by fitting data 11, • P = slope in %.

[0103] The critical speed-slope relationship is then defined as the limit of the model V(t, p) for a time 1 which tends towards infinity; it is noted Vc(p).

[0104] The unit of slope is % for convenience for runners who are used to using this unit. However, it is easy to use radians with the following formula:

[0105] [Math.2] P radians = percent)

[0106] Note here that the critical speed-slope relationship is defined, in the embodiment currently described, as the inverse of a second-order polynomial.

[0107] Once the adjustment 152 has been carried out, its parameters a, b, etc. are determined and it is possible to graphically draw the corresponding model. This takes the form of a finite surface in a three-axis reference frame: slope P, time f and critical speed Vc, as illustrated in [Fig.l]. We observe, in [Fig.l], a reduction in the average speed linked to an increase in the duration of effort and the slope (in absolute value).

[0108] Algorithm for real-time calculation of speed during an activity

[0109] An embodiment of the method of using the model as generated above is described below, in particular with reference to [Fig.9]. It can be implemented on a smartphone or a GPS watch so that the practitioner, having previously generated his speed-slope-time model 1, begins a trail activity.

[0110] He launches the application and retrieves 201, on his GPS watch or his mobile, the speed-slope-time model 1 generated 200 for example by a server, or at the very least the coefficients a, b, etc. of the critical speed-slope relationship which corresponds. Via the application, the user traces or imports 202 a route, preferably defined by GPS coordinates, which he intends to follow during his next trail. The route is thus preloaded in the application.

[0111] The retrieval 202 of the route may more particularly comprise the loading of a matrix of tiles, for example those implemented by the third-party application OpenStreetMap®, over which the preloaded route will pass. The tiles are defined by a square surface of the mapped geographical area, containing among other things the listed paths and roads.

[0112] These steps that can be described as preparatory for the method of use 200 according to the second aspect of the invention can be carried out by the user via an application interface, an illustration of which is given in [Fig.5].

[0113] During the race, the method of use according to the second aspect of the invention provides for the acquisition 203 at a frequency of 1 Hz of the GPS position (latitude / longitude) and altitude data. Note here that the altitude data are not necessarily accessible by the GPS data, or only by these, whether for the implementation of the generation method 100 according to the first aspect of the invention or that of the method of use 200 according to the second aspect of the invention; they can be given by the GPS watch or the smartphone, or any other device capable of determining the altitude at which the participant is geolocated at each moment.

[0114] Preferably, a filtering 204 of the altitude and / or of a determined distance between two positions of the practitioner between two consecutive measurement times is carried out with an infinite impulse response (IIR) filter. It is thus possible to process the entirety of the data of a signal without a size being imposed on this signal or whatever the size of the signal, and therefore to filter 204 the data without knowing a priori their size. In addition, in the event of loss of communication network, in particular between the geolocation system, for example GPS, and the GPS watch or the smartphone of the practitioner, a reconnection of 3 seconds is advantageously sufficient to regain the benefit of the delivered indications 210.

[0115] The method of use according to the second aspect of the invention then provides for the calculation 205 of the slope traveled at each instant as the derivative of the altitude with respect to the distance and the calculation 206 of the speed as the derivative of the distance with respect to time.

[0116] Possibly concomitantly with steps 205 and 206, the method of use according to the second aspect of the invention provides for the estimation 207 of the type of terrain encountered, as well as, where appropriate, the terrain conditions encountered. This estimation may more particularly comprise: a. An identification of the tile corresponding to the GPS position of the practitioner, b. For each path listed on this tile, calculating the distance between the GPS position and said path, then c. Retrieving characterization data for the closest path, these characterization data allow the type of terrain covered to be estimated.

[0117] This is followed by the calculation 208 of the critical speed to be respected as a function of the slope, the altitude, and the user model, with the following equation:

[0118] [MATH 3] y — 1-6x10 ^x / v where ; c a'^p^+h'^p+c a. h is the altitude of the practitioner in meters, b. P is the slope in %, and c. b, and c are the coefficients of the second-degree polynomial describing the critical speed-slope relationship.

[0119] Note that the coefficient “6 x 10'” corresponds to the decrease in physical capacities as a function of the increase in altitude, which is illustrated by the graph in [Fig.4],

[0120] The calculated speed 206 of the practitioner as well as the calculated critical speed 208 can be represented on the same graph in the manner illustrated by [Fig.6], which makes it possible, in relation to [Fig.7], to illustrate a way in which the indications of exceeding the critical speed can be managed.

[0121] But, preferably, before being compared to the speed at each instant of the practitioner, the critical speed is corrected by the application 209 of a correction coefficient to the critical speed to be respected according to the type of terrain, or even the terrain conditions. The correction coefficient to be applied 209 is more particularly determined according to the recovered characterization data. This coefficient, noted below coef, is more particularly applied in the following manner:

[0122] [Math.4] corrected — COCf Vcritique

[0123] It is typically between 0.5 (trail) and 1 (asphalt).

[0124] Then, if, for a duration greater than a predefined threshold value, the calculated speed 206 of the practitioner is greater than the critical speed corrected by the application 209 of the correction coefficient, the method of use 200 according to the second aspect of the invention provides for indicating it 210 to the practitioner, for example using his GPS watch or his smartphone.

[0125] More particularly, and with reference to [Fig.7], the indication 210 given to the practitioner may be of a first level of importance, noted “LIGHT” in [Fig.7], if the calculated speed 206 of the practitioner is greater than the critical speed corrected by the application 209 of the correction coefficient for a duration greater than a first predefined threshold value and less than a second threshold value predefined, and a second level of importance, higher than the first, noted "STRONG" in [Fig.7], if the calculated speed 206 of the practitioner is higher than the critical speed corrected by the application 209 of the correction coefficient for a duration greater than the second predefined threshold value, the second predefined threshold value being higher, for example substantially twice higher, than the first predefined threshold value. The notification of the higher level of importance can be more visible and / or more audible than that of a lower level of importance.

[0126] Thus, each plateau in [Fig.7] which takes a value greater than 0 m / s corresponds to sending a notification to the practitioner to reduce his running speed, preferably until he returns to the critical speed defined by his model. Only one of the notifications illustrated is of the second level of importance.

[0127] The invention is not limited to the embodiments previously described and extends to all embodiments covered by the invention.

[0128] For example, if the user has too little data, whether in terms of quantity or variability (slopes, durations and intensities of effort), model 1 may not be able to be generated 100, or with too low a confidence index. A simplified test protocol could be proposed to the user in order to quickly circumvent this limitation. This protocol would not require any measuring device other than the GPS watch or the practitioner's smartphone, and could only require the practitioner to record the route of a chosen portion of road over a known distance, such as the route of an athletics track, or a hilly route with several slope conditions for example, to complete the training data received 110. Thanks to this protocol, which the user could carry out alone, a model 1 could then be generated 100, with, that said, potential limitations as to its relevance compared to complete training data.

[0129] Another improvement of the various aspects of the present invention concerns the characterization of the type of terrain covered: path, carriageway, or asphalt. Today, the method of use 200 according to the first aspect of the invention is based on the data provided by the OpenStreetMap® mapping service. For each road or path listed, metadata makes it possible to characterize portions thereof. However, there are limitations to this way of proceeding, including: a. A portion of path is characterized by a “type,” or even by terrain conditions, for the entire portion. However, this portion of path may include sufficiently significant variations in terrain to make running more or less difficult within the same portion; and / or b. The metadata provided by the OpenStreetMap® mapping service comes from official country classifications of roads and paths. A The same classification may have different characteristics depending on the country. Furthermore, since officially unclassified roads and paths are provided by the OpenStreetMap® community of contributors, a subjective part of the contributor regarding the classification of the path, as well as the understanding of the numerous classification possibilities, can have a significant impact on it; and / or c. Generally speaking, the metadata provided by the OpenStreetMap® mapping service may not represent the current state of a path or road. For example, the quality of a path may have deteriorated since its classification: the appearance of stones due to erosion, growth of vegetation, etc.

[0130] As already mentioned above, the various aspects of the present invention could be transposed to other physical activities of endurance with human locomotion, and in particular to cross-country skiing, cycling, or ski touring, for example, which are distinguished from running only in that they are physical activities involving equipment that interacts between man and his environment. These endurance activities with human locomotion are thus very good candidates for the application of the speed-slope-time model described above. The application of the invention to these other physical activities would make available to the general public the performance optimization offered by the present invention, based solely on their training data.

[0131] Let us also note that the notion of "time" has been used throughout the above, but that this could be replaced by a notion linked to it, namely the notion of "endurance", where appropriate these two notions being linked together by a proportionality, or more generally a determined function. These two notions can therefore be considered as relatively equivalent to each other.

Claims

Claims

1. Method for generating (100) a speed-slope-time model (1) characterizing, for a given practitioner, a physical activity of endurance with human locomotion, the method comprising the following steps: • receive (110) a set of geolocation data representative of a latitude, a longitude and an altitude of the practitioner at different times during at least one session of practicing said physical activity, • process (120) at least part of all the geolocation data received (110) to calculate (126) representative data, per unit of length chosen, of a slope profile covered by the practitioner during each session, • on the basis of at least part of all the geolocation data received (110) and the data representative of the previously calculated slopes (126), calculate (130) an average slope and an average speed of the practitioner for each of a plurality of predefined target durations, • based on the calculated average slopes and average speeds (130), deducing (140) a maximum average speed (11) for each target duration among the plurality of predefined target durations and each among a plurality of predefined target slopes, and • on the basis of the maximum average speeds (11) deduced (140), generate (150) the speed-slope-time model (1) of the practitioner by determining a plurality of parameters on which the speed-slope-time model depends, by adjusting (152) said model to the maximum average speeds (11).

2. Generation method (100) according to the preceding claim, in which, the geolocation data being defined in a spherical reference frame, the step of processing (120) the geolocation data considered comprises, before the calculation (126) of the data representative of the slope profile: • a transformation (121) of the geolocation data by a change of reference from the spherical reference to a Cartesian reference frame, then • a first interpolation (122), for example linear, on the transformed geolocation data, at a sampling frequency of between 0.02 and 10 Hz, preferably substantially equal to 1 Hz, to obtain temporally sampled data, then • a second interpolation (125), for example linear, on the transformed data defined on the time base, over a sampling distance of between 0.1 and 500 m, preferably substantially equal to 1 m, to obtain geographically sampled data, so that the calculation (126) of the data representative of the slope profile is a function of the geographically sampled data.

3. Generation method (100) according to the preceding claim, further comprising, after the first interpolation (122) and before the second interpolation (125), a deletion (123), from the temporally sampled data of those which do not exist in the data sampled before the first interpolation (122), and for which the elapsed time is greater than a threshold value between 1 second and 1 minute, preferably substantially equal to 10 seconds.

4. Generation method (100) according to any one of the two preceding claims, further comprising, after the first interpolation (122) and before the second interpolation (125), the application of a low-pass filter (124) to the geolocation data representative of the altitude, in order to smooth out any measurement inaccuracies.

5. Generation method (100) according to any one of the three preceding claims, further comprising, after the calculation (126) of the data representative of the slope profile, a third interpolation (128), for example linear, on the data representative of the slope profile, at a sampling frequency of between 0.02 and 10 Hz, preferably substantially equal to 1 Hz, so that the calculation (130) of the average speeds is a function of the calculated (126) and temporally sampled data representative of the slope profile.

6. A generation method (100) according to any one of claims previous, wherein the calculation (126) of the data representative of the slope profile is followed, preferably immediately, by the application of a low-pass filter (127) to said data representative of the slope profile.

7. Generation method (100) according to any one of the preceding claims, in which the target durations are predefined and included in a set of intervals ranging from a minimum value between 1 and 600 seconds, preferably substantially equal to 120 seconds, to a maximum value between 6 minutes and 6 hours, preferably substantially equal to 14 minutes, the intervals of the set, preferably regular and / or adjacent to each other, taking a value between 1 second and 10 minutes, preferably substantially equal to 2 minutes.

8. Generation method (100) according to any one of the preceding claims, in which the target slopes are predefined and included in a set of intervals ranging from a minimum value between -10% and -60%, preferably substantially equal to -50%, to a maximum value between 10% and 60%, preferably substantially equal to 50%, the intervals of the set, preferably regular and / or adjacent to each other, taking a value between 0.1 and 10%, preferably substantially equal to 2%.

9. A generation method (100) according to any preceding claim, wherein the step of deducing (140) the maximum average speed for each target duration is preceded by a deletion (131), for each target duration and each target slope, from the calculated average slopes and average speeds (130), of data representative of slopes not having a majority of values, for example at least 80% of values, located in the corresponding target slope, preferably with a tolerance substantially equal to + / -2.5%.

10. A generation method (100) according to any preceding claim, wherein the step of deriving (140) the maximum average speed for each target duration and each target slope further comprises removing (141), from the deduced maximum average speeds (140), maximum average speeds substantially less than 0.5 m / s and substantially greater than 7 m / s.

11. A generation method (100) according to any preceding claim, wherein the adjustment (152) of the speed-model slope-time is implemented by implementing a method of minimizing the residuals of an equation V(t, p) defining the velocity-slope-time model, where V is the maximum average velocity,' is time and P is the slope.

12. Generation method (100) according to any one of the preceding claims, in which the speed-slope-time model (1) is defined by an equation V{t, p) taking the following form: p} — ^ + Ve(p)' • where D is a reserve of distance achievable above the critical speed (in meters), 1 is the duration of an effort made (in seconds), P is the slope (in %), and • the limit of which, when the duration of the effort made 1 tends towards infinity, defines a critical speed-slope relationship, noted Vc ( p ), characterizing the practitioner.

13. Generation method (100) according to any one of the preceding claims, in which the speed-slope-time model (1) is defined in part by a critical speed-slope relationship Vc{p} taking the form of the inverse of a polynomial of at least the second degree, having as variable the slope P and whose parameters, noted for example a, b and c, are determinable by adjustment of the equation V(t, p) defining the speed-slope-time model (1) on the basis of the maximum average speeds (11) deduced (140).

14. Generation method (100) according to the preceding claim, in which the critical speed-slope relationship Vc(p) takes the form of the inverse of a second-degree polynomial: Vc(p) = where the parameters a, b and c are determinable by adjusting the critical speed-slope relationship Vc(p) on the basis of the maximum average speeds (11) deduced (140).

15. Method of using (200) a determined speed-slope-time model (1) characterizing a human locomotion endurance physical activity of a practitioner, the method of using (200) comprising the following steps: • receiving (201) coefficients defining, for the practitioner, his speed-slope-time model (1), • receiving (202) a matrix of tiles comprising a route that the practitioner proposes to follow during a session of practice of said future physical activity, then • at a plurality of times, preferably at each time, during the session: i. receive (203) geolocation data representative of the latitude, longitude and altitude of the practitioner, ii. based on at least part of the geolocation data received (203), calculate (205) a slope traveled by the practitioner, iii. based on at least part of the geolocation data received (203), calculate (206) a speed of the practitioner, iv. based on at least part of the geolocation data received (203), estimate (207) a type of terrain covered by the practitioner, or even terrain conditions during the session, v. based on the calculated slope and speed (205 and 206), determine (208) using the speed-slope-time model (1) and the received coefficients (201), a critical speed not to be exceeded by the practitioner, vi. depending on the estimated type of terrain (207), or even the terrain conditions encountered, apply (209) a correction coefficient of the determined critical speed (208), vii. if, for a period greater than a predefined threshold value, the calculated speed (206) of the practitioner is greater than the critical speed corrected by the application (209) of the correction coefficient, indicate this (210) to the practitioner.

16. Method of use (200) according to the preceding claim, in which the speed-slope-time model (1) was generated by implementing a method of generating (100) a speed-slope-time model characterizing a physical activity of endurance with human locomotion of a practitioner according to any one of claims 1 to 14.

17. A method of use (200) according to any one of the two preceding claims, wherein each tile is defined by a surface area, for example square, of the geographical area in which the session of practicing said physical activity takes place or is intended to take place.

18. A method of use (200) according to any one of the three preceding claims, wherein the geolocation data is received (203) at a frequency substantially equal to 1 Hz.

19. Method of use (200) according to any one of the four preceding claims, further comprising, before the steps of calculating (205 and 206) the slope and the speed of the practitioner, the application (204) of an infinite impulse response (IIR) filter on the received geolocation data (203).

20. Method of use (200) according to any one of the five preceding claims, in which the estimation (207) of the type of terrain traveled comprises: • the identification of the tile corresponding to the geolocation of the practitioner, • for each of the paths listed on the identified tile, the calculation of the distance between the geolocation of the practitioner and said path, and • the recovery of characterization data of the path closest to the geolocation of the practitioner.

21. Method of use (200) according to the preceding claim, in which the correction coefficient to be applied (209) is determined as a function of the recovered characterization data.

22. Method of use (200) according to any one of the seven preceding claims, in which the correction coefficient to be applied (209) is substantially between 0.5 and 1 and / or in which the application (209) of the correction coefficient consists of multiplying the determined critical speed (208) by the correction coefficient.

23. A method of use (200) according to any one of the eight preceding claims, wherein the determination (208), using the speed-slope-time model (1) and the received coefficients (201), of the critical speed not to be exceeded comprises calculating the critical speed according to the following equation: y _ 1-Coeffy, WHERE: c crxp2+b'xp+c • h is the altitude of the practitioner, in meters; • P is the slope covered, in %; • a, / 2, etc. are the coefficients received (201), and • Coef fvo2wÆv is a coefficient reflecting the reduction in the physical capacities of the practitioner as a function of altitude, this coefficient is for example approximately equal to 6 x 10°-

24. Method of use (200) according to any one of the nine preceding claims, in which the indication (210) given to the practitioner is of a first level of importance if the calculated speed (206) of the practitioner is greater than the critical speed corrected by the application (209) of the correction coefficient for a duration greater than a first predefined threshold value and less than a second predefined threshold value, and of a second level of importance, greater than the first, if the calculated speed (206) of the practitioner is greater than the critical speed corrected by the application (209) of the correction coefficient for a duration greater than the second predefined threshold value, the second predefined threshold value being greater, for example substantially twice greater, than the first predefined threshold value.

25. A computer program product, preferably recorded on a non-transitory medium, comprising instructions, which when performed by at least one of a processor and a computer, cause the at least one of the processor and the computer to execute the generation method (100) according to any one of claims 1 to 14 and / or the use method (200) according to any one of claims 15 to 24.

Images