Patent Application
20250032857
In various embodiments, the present disclosure relates generally to exercise machines, and more specifically to exercise machines that dissipate energy through controllable loads, and to exercise machines having an interconnection capability that enables loads to be set responsively to the efforts of other athletes.
In many athletic activities, including some forms of racing, the power output (effort) required of an individual athlete at a given level of performance depends on choices made during the activity by that athlete and by other athletes, among other factors. For example, the aerodynamic drag encountered by an athlete can depend on the proximity of other athletes. One activity in which this is the case is bicycle racing: e.g., when one or more cyclists ride directly behind a leader, aerodynamic drag on the drafting cyclists can be reduced to only 30%-70% of the drag that would be encountered by an isolated rider. In bicycle racing, cyclists therefore typically form a peloton, a group of cyclists riding closely together to reduce aerodynamic drag and energy expenditure. Cyclists in a peloton experience less drag and therefore exert less effort to ride at a given speed than they would if they were riding separately. In a large peloton (e.g., 120 riders), drag on individuals at the mid-rear of the group can be reduced to only 3%-10% of the drag on an isolated rider (Blocken B. et al., “Aerodynamic drag in cycling pelotons: New insights by CFD simulation and wind tunnel testing,” Journal of Wind Engineering and Industrial Aerodynamics, 179 (2018): 319-337). However, athletes typically seek to achieve outstanding performance, not simply to travel efficiently as part of a group. In the case of bicycle racing, a cyclist's choices about when and where to be part of the peloton, and thus to realize or sacrifice aerodynamic advantage, are strategically important. For example, a rider might ride in the midst of the peloton for much of a race, experiencing low drag and conserving their strength, then attempt to sprint forward against heightened drag near the end of the race. The decisions of other riders can affect the drag encountered by each rider: e.g., a decision by one or more riders to break away from a peloton changes the drags encountered both by the riders breaking away and by the riders remaining in the peloton, and for each rider, this change depends on their position with respect to other riders. In another example, if a peloton accelerates, its members all experience greater drag even without changing their positions. In general, a racing cyclist must dynamically manage their power output and strength reserves under conditions where (a) drag is a major determinant of their required effort and (b) drag is significantly influenced both by the cyclist's own behavior and by the behaviors of other cyclists, among other factors. In yet another example, in team riding, a cyclist may deliberately take a higher-drag position in front of a teammate for a time in order to help the teammate conserve energy. Context-dependent variations on these and other strategic decisions are, it will be clear, essentially endless and can be crucial to competitive outcomes. Herein, management of power output and strength reserves by an individual athlete or athletes comprising a team is termed strategic effort management.
Strategic effort management is a crucial aspect of some competitive group activities, but extant stationary-training methods do not simulate it. Teams can train for competition by actually riding on roads, rowing on water, or the like; however, due to seasonal, weather, cost, convenience, and other limitations on field exercise opportunities, in practice athletes prepare for and supplement field exercise by working out extensively on trainers, i.e., stationary machines whose mechanics simulate one or more aspects of the sport in question. A typical trainer comprises one or more mechanisms upon which the user's body rests and/or acts (e.g., pedals, oars, seats, skis) and one or more dissipative mechanisms (e.g., air fins, friction pads), typically adjustable, which place energetic loads on the user. A typical trainer also comprises an inertial mechanism (e.g., flywheel) that simulates the inertia of one or more athletes as well as that of a watercraft, bicycle, or other gear.
The prior art enables individual training with a variable load: e.g., for bicycling, stationary systems exist that project a video recording or simulation of a road on a screen before the cyclist and then vary load (pedaling resistance) according to the slope of the virtual road. Systems also exist that reduce load to simulate the drafting effect of a leading cyclist upon an operator. The prior art also provides for the coordination of variable loads of single-user and/or multiple-user training machines that may be located anywhere and are linked through a network such as the Internet, where the said load coordination enables trainees to experience virtual cooperation and competition, e.g. in virtual rowing or cycling: such methods are disclosed in U.S. Pat. No. 10,610,725B2, “APPARATUS AND METHOD FOR INCREASED REALISM OF TRAINING ON EXERCISE MACHINES,” the whole of which is incorporated herein by reference. However, the prior art does not enable users of a multiplicity of stationary training machines to experience individually varying loads that realistically simulate the complex, mutual, physical interactions of group racing. In particular, the prior art does not provide for load variations that reflect the relative virtual positions of a multiplicity of operators. In short, the prior art does not enable training in strategic effort management.
It is preferable that training methods for a variety of sports mimic the challenges of real-world activity, including strategic effort management, as realistically as possible. It is also preferable that such training be viable for persons who are not physically co-located and that it enable novel forms of beneficial non-realistic training and realistic competition by non-colocated athletes, whether human, simulated, or a mixture of the two. Techniques are therefore desired by means of which exercise machines can enable one or more athletes to receive energetically realistic training in multi-user settings.
Exercise machines constructed according to the prior art are generally provided with energy dissipation (i.e., load) devices that employ frictional, gaseous, liquid damping, and/or electromechanical mechanisms. Such exercise machines (herein also termed “trainers”) can, as described hereinabove, be linked with each other and with ancillary computing and other equipment via a telecommunications network in order simulate some aspects of real-time cooperation and/or competition. However, trainers constructed according to the prior art cannot realistically vary individual trainer load to reflect the physics of real-world racing, including but not limited to dynamic changes in drag forces experienced by athletes as a consequence of their own and other athletes' behaviors. For these and other reasons, improvements to exercise machines are desired.
Various embodiments of the disclosed apparatus and systems transcend the limitations of the prior art by enabling athletes on separate exercise machines, which may be far distant from each other, to exercise jointly in a manner that includes but is not limited to dynamic changes in drag forces experienced by athletes as a consequence of their own and other athletes' behaviors. Moreover, various embodiments of the disclosure enable athletes to train with or against simulated athletes. Some other advantages offered by embodiments of the disclosure shall be made clear hereinbelow both descriptively and with reference to the Figures.
Herein, the terms “user,” “operator,” “trainee,” and “athlete” are used interchangeably. Herein also, the term “position” is used to denote not static position but relative position in a moving group. Herein also, reference is made in the text and Figures to “racing,” “competitors,” and other like, but it will be understood that both competitive and noncompetitive activities are contemplated and within the scope of the disclosure. Herein also, reference is made in the text and Figures to aspects of bicycling, but it will be understood that the range of activities contemplated and within the scope of the disclosure is not limited to bicycling. There is no restriction of activities with respect to equipment type, number of participants, solo or team activity, competition or noncompetition, indoor or outdoor activity, or other features.
In various embodiments, an exercise machine comprises an electrical machine (generator) whose power output is dissipated largely by a resistive electrical load. Various embodiments comprise additional computational, communicative, and other aspects. In various embodiments, one or more computational aspects or portions of the apparatus collect measurement information telemetrically from portions of the exercise machine; issue commands to controllable aspects of the exercise machine (e.g., electrical machine, resistive load, audiovisual displays, haptic feedbacks); and communicate informatically (e.g., through a network) with various devices that can include, but are not necessarily limited to, one or more of (1) other, similarly equipped exercise machines, (2) a server that can gather and store data pertaining to multiple exercise machines and their operators and coordinate the behaviors of multiple exercise machines, (3) other computing devices, including devices operated by one or more coaches and/or by team participants with distinctive roles (e.g., leaders, coxswains), and (4) sources of physiometric information such as wearable athletic monitors or activity trackers.
The computational portions of various embodiments include software capabilities for (1) algorithmically modeling combinations of one or more operators, who may be at disparate physical locations, into one or more virtual groups whose members' behaviors affect the loads experienced by other members, (2) numerically simulating the effects of the behaviors of one or more operators, real and/or simulated, on the loads experienced both by those operators and by other operators, and causing the calculated effects to be felt by the one or more operators in the load presented by their training machine, (3) numerically simulating the effects of aspects of a virtual environment upon individual operators (e.g. slopes or elevation changes, turns, winds), including the modulation of such effects by the behaviors of one or more operators, (6) rating and otherwise analyzing operator and team performance, and (4) calculating the performance of combinations of real and simulated operators so that individual operators may train as individuals or as part of a complete team, or a partial team may train as part of a complete team, or one virtual team (entirely real or partly or entirely simulated) may compete against one or more other teams (entirely real or partly or entirely simulated). The performance characteristics of a simulated operator constitute a set of tunable parameters that may be based on the measured characteristics of a real operator, selected from a library, custom-specified, randomly generated, or otherwise specified; the behaviors of a simulated operator constitute a set of numerical outputs representing effort, position, fluid-dynamic characteristics, and other physical variables.
Also, various embodiments comprise devices that offer audiovisual and/or haptic feedback to operators that can supplement the feedback supplied by the mechanical load of the exercise machine: for example, a stationary-bike operator may face a device that gives visual and/or aural cues such as an audiovisual representation of (a) a landscape to provide a visual indication of motion, (b) other cyclists (real and/or simulated), and (c) performance metrics (e.g., velocity, time, mileage, power output) for the operator and/or for one or more other operators. Preferably the audiovisual feedbacks offered to multiple operators training as a group are coordinated by a computational device so that operators are offered consistent information. Audiovisual feedback is in some embodiments provided to the operator by a virtual reality device (e.g., Oculus Rift) to endow the training experience with a high degree of psychophysical realism.
In various embodiments, the disclosed apparatus comprises an electrical machine (e.g., a linear or rotary electrical generator) that is motivated by one or more operators and supplies power to a load (e.g., a bank of resistors) and that provides load for the operator. In various embodiments, the load that dissipates power generated by the electrical machine can comprise one or more resistors that dissipate energy as heat. The one or more resistors of the electrical load are herein collectively termed “the electrical load bank.” In one example, the net resistance of the electrical load bank is fixed and the current through the load bank is varied in proportion to the required load. Alternatively, the net resistance of the electrical load bank is adjustable by means of signals transmitted from the system controller: e.g., relays may connect or disconnect resistors in the electrical load bank, thus increasing or decreasing the mechanical load presented to the operator. Additionally or alternatively, the electrical load bank may comprise continuously variable resistive elements (e.g., potentiometers). Non-electrical loads such as friction brakes and fluid-stirring mechanisms may be comprised by various embodiments, additionally or alternatively to resistive and other electrical loads.
In one example, the computational portion of the apparatus maintains in its memory numerical representations of the state of a virtual environment and of one or more virtual riders in that environment. The virtual environment's state representation can include a three-dimensional physical map, a time-varying wind field, the virtual riders, and other data. A virtual environment may reflect the character of a wholly real place, a partly imaginary place, or a wholly imaginary place. A virtual rider's state representation can include mass, position, velocity vector, acceleration, drag, drag coefficient, sensed steering input, sensed braking input, sensed power output of the operator, and other data. The computational portion of the apparatus updates these state representations, and any audiovisual and/or haptic feedbacks to operators based upon them, frequently enough (e.g., 15 times per second) to create an illusion of continuity for human operators. Updates are based on a computational model of the physics of the virtual environment and of virtual riders, including calculated inputs (e.g., gravity, winds) and measured inputs such as the efforts exerted by trainer operators associated with specific virtual riders. Calculated rider inputs maybe substituted for measured inputs to create a simulated virtual rider. Since each virtual rider's state updates (e.g., changes in position and velocity) are influenced by the state of the virtual environment, which contains all virtual riders, each virtual rider's state (e.g., effort, position, velocity) affects their own evolving state and potentially (though not always significantly) that of every other rider. The computational portion of the apparatus adjusts the load mechanism of each trainer to reflect the computed load of the virtual rider corresponding to that trainer's operator. Trainer loads for operators will, for example, tend to vary as virtual riders change relative position in the virtual environment, altering each other's aerodynamic conditions.
Additionally or alternatively in various embodiments, the trainer comprises braking controls that mimic the operation of brakes on an actual bicycle or other machine, and sensors whose outputs convey the state of operation of the braking controls to the computational portion of the apparatus. The computational portion, taking into account sensed braking and other inputs (e.g., slope or elevation change, effort, drag), then calculates appropriate alterations in the velocity, position, drag, and other aspects of numerical representations of that operator and of other operators and commands corresponding changes in audiovisual representations delivered to that operator and other operators, as well as in various operators' training apparatus loads. In one example, a cyclist who brakes to avoid hitting a cyclist who cuts them off in the simulation drops back in audiovisual representations presented to them and to the other cyclists in the group, and the braking cyclist and other cyclists may experience altered drag in the form of increased or decreased trainer load.
Additionally or alternatively in various embodiments, the training apparatus comprises steering controls (e.g., handlebars) that mimic the operation of steering controls on a bicycle or other machine, and sensors convey the state of operation of the steering controls to the computational portion of the apparatus, which calculates appropriate alterations in the velocity, position, drag, and other aspects of a numerical representation of the operator, and commands corresponding changes in audiovisual representations delivered to that operator and other operators, and to trainer loads. In one example, a cyclist who steers left to separate from a peloton, but pedals hard enough to maintain constant velocity, will experience increased drag; also, their changed position will appear in audiovisual representations presented to them and to the other cyclists in the group, and other cyclists in the group may also experience altered drag in the form of increased or decreased trainer load. Steering controls may incorporate haptic feedback so that the “feel” of steering (e.g., a bicycle) at a certain speed will be reproduced in the controls.
In one example, cyclists in a virtual race—every one of whom is physically hundreds of kilometers away from all the others—share a virtual reality in which each cyclist occupies a particular position on a virtual roadway at each moment of time. In this example, the virtual reality of each cyclist is consistent with that of every other cyclist, i.e., the point of view of each cyclist agrees with the representation of that cyclist in the virtual reality of every other cyclist (e.g., the leader of a peloton is seen by herself and all other cyclists to be occupying the lead position). Moreover, the movements of each cyclist in the virtual environment are determined algorithmically from the numerically represented physical efforts, forces acting upon, and other aspects of the cyclist and the virtual environment. Moreover, the load and other feedbacks presented by each trainer to its operator are determined algorithmically from factors that include, but are not limited to, the aerodynamic effects of other riders on that rider. In this example, each cyclist has a group-interactive training experience that realistically reflects the dynamic, mutually influential loads and other constraints of actual road riding.
Various embodiments of the disclosure combine networked communications and model-based control of trainer loads to transcend shortcomings of the prior art. The ability of various embodiments to place trainer operators in a shared virtual environment that simulates a physical setting, where each operator experiences a varying trainer load that depends realistically on their own behaviors and the behaviors of other operators and/or simulated athletes, overcomes the prior art's inability to train athletes in strategic effort management, which is a crucial aspect of real-world competition in bicycle racing and some other activities. Various embodiments enable operators of stationary trainers, who may be at widely separated physical locations, to convene virtually for races whose athletic realism exceeds that made possible by the prior art. Such races may be held for training purposes or as a distinctive form of competition.
Certain functions offered by some embodiments of the disclosure are entirely novel as compared to the prior art. In an example, in various embodiments one or more operators can train with a physically unrealistic interactive goal, such as a specified total group power output, constant or time-varying. Each operator can, in this example, be encouraged to increase or decrease their effort in response to increased or decreased effort from other team members, where such feedback may be based on an algorithm that takes into account performance profiles of participating operators. Or, each operator may simply be shown whether the team as a whole is achieving, exceeding, or falling short of its target output, and be allowed to adjust their own effort at will. In this example, each operator may experience a trainer load that is identical to the loads of other operators, or an individualized load; trainer loads may be constant or time-varying; and trainer loads may be set algorithmically in a manner that takes into account the efforts of various operators. It will be clear that variations on such novel, physically non-realistic training, preferably designed to provide amusement and/or strengthen specific aspects of athletic ability, can be readily multiplied. In yet another example, operators may practice as a member of a group, some or all of whose other members are simulated, where simulated operators can be adjusted to challenge human operators in ways that a coach deems appropriate for training needs of those human operators. A simulated operator's characteristics may be based upon those of a prominent human athlete, enabling operators to train against competition similar to that offered by expert athletes.
In other examples, embodiments of the equipment and the methods of operation described herein, applied to bicycle racing training, can enable simulated races of groups of bicycles consisting of bicyclists from multiple geographic locations. This simulated event can simulate a traditional bicycle race where multiple bicycles permitting teams of bicyclists race at a single location. The simulated event includes the ability to simulate certain natural conditions during a bicycle race (e.g., drafting effects, slope of a road track, wind, etc.). The simulated event can also use different bicycles (e.g., bicycles having different mass values, gearing arrangements, etc.) Course conditions can also be simulated at the discretion of those setting up the simulated event or the participant bicyclists.
These and other objects, along with advantages and features of the disclosure, will become more apparent through reference to the following description, the accompanying drawings, and the claims. Furthermore, it is to be understood that the features of the various embodiments described herein are not mutually exclusive and can exist in various combinations and permutations. Furthermore, the particular features, structures, routines, steps, or characteristics may be combined in any suitable manner in one or more examples of the technology. Also, although single-user trainers are frequently referenced herein, multi-user trainers may be similarly incorporated in embodiments of the disclosure. All such variations are contemplated and within the scope of the disclosure.
The foregoing and other aspects of the present disclosure will become apparent to those skilled in the art to which the present disclosure relates upon reading the following description with reference to the following figures:
In the Figures and discussion thereof, systems and methods are disclosed that enable the construction of an exercise apparatus which improves aspects of individual and team training. These systems and methods can provide networked communication between multiple exercise machines to provide machine users with a common exercise experience that simulates the energetic interactions which occur when mobile athletic apparatuses are operated in a common space, e.g., in a real bicycle race. The types of exercise machine to which these systems and methods apply include, but are not limited to, stationary bicycles, rowing machines, treadmills, elliptical machines, and cross-country skiing machines. This disclosure primarily describes illustrative cases in which the exercise machine is a stationary bicycle, but no restriction is intended by this usage. In the Figures, for the sake of clarity, certain features are omitted whose necessity or utility would be clear to persons familiar with the design and operation of exercise machines and other relevant devices; for example, detailed provisions for wiring an alternator or for plugging into main power are not depicted, and force transmission mechanisms standard to various exercise machines are not depicted. The emphasis of the Figures is on features that clarify embodiments of the disclosure.
The load 112 comprises an electrical load which serves to dissipate or absorb electrical energy produced by the electrical machine 110.
The user interface 116 comprises one or more of audible, visual, and haptic means of conveying information to the athlete 102, where such information can comprise metrics of athlete performance (e.g., stroke rate, power output), athlete biometrics (e.g., heart rate), audiovisual representations or simulations (e.g., of a virtual reality), audio (e.g., voice, rhythm cues), and others. The user interface 116 also comprises one or more means of information input from the athlete 102 (e.g., voice input, keyboard input, touchscreen input, eye-movement based interaction, etc.).
The computer 114 comprises a data-gathering capability, computational capability, control capability, communications capability, and memory capability. The data-gathering capability of the computer 114 receives signals from sensors (not shown) communicating with various portions of the trainer 104. In
The communications capability of the computer 114 enables it to exchange information with a network 130. The communications capability is capable of information exchange through one or more wired channels and protocols, one or more wireless channels and protocols, or both. In an example, the network 130 comprises M—1 exercise machines (M≥1) that are similar to trainer 104 and are interconnected by cabled or wireless channels, where machine 104 and the machines with which it is in communication act as communicative nodes in a network topology. In another example, the network 130 is the Internet. Through the network 130, the computer 114 can be in informatic communication with machines similar to trainer 104, general computing devices, and other devices capable of informatic exchange through the network 130. In an example, the trainer 104 communicates through the network 130 with a wearable sensor device worn by the athlete, acquiring biometric information (e.g., heart rate) and utilizing such information in the computation and memory capabilities of the computer 114.
All data available to the computer 114 may be made available via the network 130 to some other computer or to more than one computer, and all relevant calculations performable by the computer 114 may be performed by some other computer or more than one computer linked to the network 130. All forms of local, remote, central, edge, cloud, and distributed computing, including all forms of distributed or local control and data storage, are contemplated and within the scope of the disclosure. For simplicity, this description of an illustrative embodiment treats all computation and data storage as being performed by the local computer 114.
The trainer 104 is in communication via the network 130 with M−1 other, preferably similar exercise machines, which are preferably also in communication with each other. Together, the trainer 104 and the M−1 exercise machines with which it is in networked communication constitute a networked group of M exercise machines. Such a networked group of M exercise machines is similar to that shown and described in U.S. Pat. No. 10,610,725B2. The quantity of members of a networked group may vary from occasion to occasion or from time to time on any given occasion.
Reference is now made to various embodiments of the disclosed apparatus and systems. These various illustrative embodiments comprise one or more exercise machines similar to the machine 104 of
As shall be made clear with reference to illustrative embodiments hereinbelow, the computational capability of the computer 114 in various embodiments implements a computational algorithm, herein termed the “race algorithm.” The race algorithm accepts as numerical inputs measured electrical and mechanical quantities from portions of the machine 104 (e.g., rotational velocity of a flywheel, acceleration of a flywheel, voltage across a resistive load, current in a generator winding). These measured quantities are such as enable estimation of the real-time locomotive effort exerted by the athlete 102 and, potentially, of other quantities, such as braking or steering effort exerted by the athlete 102 on appropriate controls (not depicted) comprised by the trainer 104. The race algorithm also accepts as inputs a number of numerical parameters stored in the memory capability of the computer 114. These parameters can express physical properties of a hypothetical athletic apparatus (e.g., drag coefficient and mass of a particular type of bicycle), physical characteristics of a physical trainer (e.g., the moment of inertia of a flywheel), physical characteristics of real or simulated athletes (e.g., drag coefficient and mass), behavioral and physiological characteristics of real or simulated athletes (e.g., strategic habits, conditioning level, performance data), characteristics of a virtual environment (e.g., a race-course in a virtual landscape and a wind field of that landscape), and other variables. The race algorithm can also accept as input real-time data representing the activities of N real and/or simulated athletes, N≥M≥1, one of whom may be the operator 102 of the trainer 104. The N athletes are herein termed a “virtual group.” Activity data may be derived from the activities of real human athletes or numerically generated to represent the activities of simulated (virtual) athletes: that is, some or all of the N athletes in a virtual group may be real athletes and some or all may be simulated athletes, although preferably at least one real athlete participates in a virtual group. Simulation of an athlete is performed by code computed by the computer 114. Simulation can be based on parameters derived by measurement from real athletes or otherwise derived, and may include a machine-learning aspect (e.g., a simulated athlete may learn from experience) and/or a random aspect (e.g., the efforts of a simulated athlete may vary slightly from moment to moment in a realistically nondeterministic fashion). If M real athletes on M real machines are participating in an N-member virtual group, then N−M team members are simulated.
Data received by the computer 114 via the network 130 during computation of the race algorithm typically include real-time data on the activities of the M−1 real athletes other than the real local athlete 102 in the virtual group. Real-time data on the activities of the real local athlete 102 are gathered directly by the computer 114 from machine 104, and are similarly gathered by computers comprised by other networked trainers. Also, the computer 114 typically transmits activity data on the local athlete 102 via the network 130 to the M−1 machines from which the machine 104 is receiving athlete activity data. Data on the activity of simulated athletes on a virtual team may be produced locally by the computer devices of exercise machines (e.g., computer 114), or communicated to or among exercise machines or computers via the network 130, or both.
The race algorithm computed by computer 114 produces commands that are communicated to various controllable mechanisms of the machine 104 (e.g., aspects of the electrical machine 110 and load 112), ultimately altering the mechanical load experienced by the athlete 102. The M−1 other networked machines similarly compute the race algorithm to calculate commands for their own mechanisms, thus affecting the experiences of their own operators in a manner coordinated with that of machine 104. That is, the M machines of the M real athletes on an N-member virtual team all possess or receive activity information for N athletes and compute load adjustments for the N athletes. For the M machines of the virtual team operated by real athletes, physical load adjustments are made; for the N−M simulated athletes, adjustments are made to the simulation calculations, appropriately altering the effort data corresponding to each simulated athlete. The method of measurement, calculation, and apparatus adjustment herein described constitutes a form of closed-loop control.
In various embodiments, the race algorithm, operating on real activity data from M real athletes and simulated activity data from N−M simulated athletes, and consequently modifying the loads specified for both real and simulated athletes on an N-member team, is designed to approximate the performance of an athletic apparatus (e.g., bicycle) operated by a single athlete or of an athletic apparatus (e.g., rowboat) operated jointly by K athletes (K≥1). By altering the parameters of the race algorithm, the physical responses of various apparatuses may be simulated. In the case of an athletic apparatus operated jointly by K athletes, (e.g., 2-rower craft of a first type, 2-rower craft of a second type, 8-rower craft), the athletes experience time-varying resistance from their exercise machines that reflects the efforts of other team members, real and simulated, in a manner approximating joint team operation of a real physical apparatus, even though other real team members are operating physically separate machines that may be geographically distant. In an example, the trainer 104 is a stationary bicycle and the athlete 102 is jointly riding a virtual 2-person tandem bicycle with another athlete, real or simulated. The resistance presented by the pedal mechanism of the trainer 104 is set by the racing algorithm in a manner that depends partly on the timing and power of the pedal-strokes exerted by the other athlete riding the virtual tandem bicycle, partly on aerodynamic drag (which influenced by the virtual positions of other virtual riders), and partly on other factors. In another example, the trainer 104 is a rowing machine and the athlete 102 is a rower participating in a virtual team rowing a virtual four-person scull. The resistance presented by the handle or oar to athlete 102 will vary throughout each stroke and from stroke to stroke in a manner that depends via the race algorithm on the timing, power, and other features of the strokes of athlete 102, on the strokes of the other three athletes on the team, and on the characteristics of the watercraft model chosen as the virtual apparatus (e.g., four-person scull).
In various embodiments of the disclosed apparatus and systems, the term “simulation,” used with reference to virtual athletes, simulated athletes, landscapes, wind forces, friction, machine behaviors, and the like, denotes any form of calculation that produces useful approximations of quantities which pertain to the operator experience, as for example simulated rider behaviors, landscape views, and wind forces. Relatively simple parametrized calculations, high-resolution physics-based simulations, and methods of mixed or intermediate character are all contemplated and within the scope of the disclosure. In particular, fully immersive virtual realities with realistic physics and high resolution are contemplated.
In various embodiments, quantitative data on individual performance, team performance, competitor performance, drag forces, and other variables can be made selectively available (e.g., visually) to individual athletes, coaches, teams, and others. Audio, video, and other data gathered from athletes and other parties (e.g., coaches, onlookers) may be integrated variously with the outputs of the operator interface 116 to produce virtual settings of varying character, interactivity, and realism, enabling the training of athletes in the psychosocial as well as physical aspects of a sport. Sport onlookers may be linked to the system through virtual-reality headsets, enabling audiences to be virtually present at virtual races rowed by real and/or simulated athletes, where all onlookers and real participants may be separated geographically to any degree. Other forms of interface coordination, e.g., coach audio shared simultaneously to all athletes on a virtual team, are also contemplated and within the scope of the disclosure. All such applications, however elaborate, depend on the capability of various embodiments of the disclosure to coordinate the experiences of athletes participating in a virtual group, where this experience includes the mechanical production for each individual athlete of a trainer exercise experience that reflects that athlete's efforts, the simultaneous efforts of other athletes, real and/or simulated, and the computed physical interactions (e.g., aerodynamic interactions) of the individual athlete and of other athletes in the virtual group. In short, various embodiments of the disclosure enable athletes to jointly participate in an augmented reality that realistically simulates interactive physical aspects of a group activity such as racing.
It is possible to apply the described apparatus and methods to other types of exercise machines. In an example, the trainer 104 of
Reference is now made to
The trainer 200 is illustrative: other configurations, which will be known to persons having skill in the art, are possible. In the illustrative embodiment of
Referring again to
Various exercise machines according to the disclosure include mechanical energy storage devices, e.g., flywheel 216 of machine 200. (Mechanical energy is stored in all moving components of an exercise system, including the athlete, but herein the phrase “mechanical energy storage device” refers to a device whose primary function is to store mechanical energy.) In the example of
It is to be understood that the mechanical energy storage device can include structures other than the flywheel 216. For example, an electrical machine that has sufficient inertia may act as both the flywheel 216 and as the generator 218. Other mechanical energy storage devices are also contemplated. Alternatively or additionally, appropriate time-variable adjustment of the electromechanical properties of and energy dissipation by an electrical machine (e.g., generator 216) can mimic for an operator the sensations of being coupled via the cyclical actuator to a mechanical energy storage device: there is no absolute requirement that any component of the exercise machine be dedicated to the storage of inertia. However, for simplicity, any device that either stores inertia or mimics for an operator the experience of interacting with stored inertia is herein termed a “mechanical energy storage device.”
In various embodiments, this relationship of parts (cyclic actuator, connective structure, mechanical energy storage device) can be realized by various mechanisms. In an example, any suitable connective structure can be used (e.g., strap, cord, chain, lever, friction wheel, pedal crank arm) that provides a physical connection between the cyclical actuator (e.g., handle, pedal, ski) and the mechanical energy storage device (e.g., flywheel, spring, moveable weight, moving fluid). Regardless of the physical make-up of the connective structure, the connective structure transfers and/or transforms a force generated by an operator of the exercise machine in such a manner that motion of the cyclical actuator urges motion of the mechanical energy storage device (e.g., rotation of a flywheel).
In this example and in various other embodiments, the electrical generator 218 can be any suitable device including, but not limited to, a separately excited electric machine, alternating current (AC) induction, permanent-magnet alternating current, brushless direct current motor, etc. Additionally, the described components are but one example of a drivetrain, and any suitable means of transferring motion can be employed by exercise machines according to various embodiments of the disclosure.
For simplicity, the illustrative exercise machine 200 of
Continuing discussion of the foregoing example, an exercise machine (e.g., machine 200 of
In an exemplary member of the illustrative class, the resistive load bank 226 is configured to supplement the load resistance of the flywheel 216. The resistive load bank 226 is in electrical communication with the electrical generator 218. The resistive load bank 226 can be considered part of the “armature circuit.” In another exemplary member of the illustrative class of machines, a wire harness delivers the electrical signal from the electrical generator 218 to the electrical load bank 226 and the electrical signal is dissipated at the electrical load bank 226, typically by generating heat. In one example, heat generated in the electrical load bank 226 can be dissipated using at least one fan 230. The rate of fan speed can be proportional to the average electrical load through the electrical load bank 226.
Additionally, the electrical load bank 226 can comprise various different structures to achieve the goal of dissipating the electrical energy created by physical work by the operator 202 input into the electrical generator 218. In one example, the electrical load bank 226 can comprise a series of resistors that dissipate at least a portion of the electrical signal created by the electrical generator 218. In another example, the electrical load bank 226 can comprise a combination of resistance elements and capacitance elements. In yet another example, the electrical load bank 226 can comprise thermo-electric generators. The thermo-electric generators can beneficially decrease the overall size of the electrical load bank 226 and provide electrical cooling to the electrical load bank 226.
The operative organization of cycling machine 200, which is typical of a number of cycling machines according to various embodiments of the disclosure, is schematically clarified in
Referring again to the illustrative machine 200 of
Discussion hereinbelow will first focus on the provision of a specific load profile to a user of isolated machine 200, that is, on a state of operation not incorporating activity data from other trainers or from simulated operators. Although this discussion refers for the sake of specificity and clarity to machine 200 of
First, it is to be noted that the effort produced by the operator 202 at any moment can be characterized by the instantaneous torque Tathlete exerted by the operator 202 on the flywheel 216 via the connective structure 210. The torque Tathlete may be considered under two aspects, i.e., actual or measured Tathlete and targeted or desired Tload. Actual Tathlete is produced by the operator 202; targeted Tload is a numerically calculated quantity which the exercise machine 200 will, in typical operation, proceed to produce in response to a changing state of the exercise machine 200. In general, one goal of a cyclist is to move at a certain speed, e.g., to produce a certain rate of acceleration, as during startup, or to maintain a certain speed, as during a cruising phase. Also, in the exercise machine 200, the rotational velocity ω of the flywheel 216 is analogous to bicycle speed: i.e., the angular momentum of the flywheel 216 turning at a given ω is analogous to the linear and angular momentum of a rider-bearing bicycle moving at a given velocity. Similarly, the effort (Tathlete) required to increase or maintain the rotational velocity w of the flywheel 216 is determined by the moment of inertia J of the flywheel 216 and by any torque loads on the flywheel 216, and this effort is analogous to that required to increase or maintain the velocity of a bicycle, which is determined by the inertia of the bicycle and rider and by any aerodynamic drag on the bicycle. The function of the controller 232 can, in this context, be stated as follows: To require of the operator 202, as the operator 202 produces a certain power output, an actual Tload that matches a calculated, target Tload reflecting hypothetical physical conditions. These hypothetical physical conditions are determined by the calculated behaviors of a hypothetical apparatus, also herein termed a “virtual rider” and similar terms (e.g., a hypothetical bicycle with rider), moving in a hypothetical physical environment that can include virtual representations of other, similar apparatuses, each corresponding to a real trainer operator or to a simulated athlete. Herein, we refer to a numerical characterization of load imposed on the hypothetical apparatus by the properties and behaviors of the hypothetical apparatus, its virtual environment, and other hypothetical apparatuses in the virtual environment as a “load profile.” Thus, targeted Tload is in general a function both of a load profile and the state of operation of the exercise machine 200, including actual Tathlete, ω, and all settable and/or intrinsic loads that contribute to the physical load experienced by the operator 202. The numerical values used to set settable loads in the exercise machine 200 may be influenced by the measured activities of both the operator 202 and other operators on other machines, real or simulated, hence the ability of various embodiments of the disclosure to produce a joint, mutually influential training experience for operators on physically separate exercise machines. These general considerations, with other considerations discussed with reference to the illustrative exercise machine 200 shown as a stationary bicycle, will be understood to apply also, with appropriate modifications, to other forms of exercise machines and athletic apparatus. This disclosure now turns to portions of the closed-loop control method employed by the illustrative exercise machine 200.
As will be clear to persons familiar with electrical machines, the excitation of an alternator, e.g., alternator 218, can be controlled by pulse-width modulation of the excitation current of the field winding, that is, by switching the field-winding voltage on and off at a fixed frequency but with a variable duty cycle. The exercise machine 200 can thus adjust, by altering the duty cycle of a pulse-width-modulated voltage source, the average excitation current of the alternator 218, which in turn affects the torque load placed on the flywheel 216 by the alternator 218 and thus the load experienced by the operator 202. To accomplish this, the controller 232 calculates an estimate of a torque value, Tathlete, that is applied by the operator 202 to the flywheel 216. The calculation of Tathlete is based on several measured variables of machine operation along with a set of pre-recorded variables representing physical characteristics of the exercise machine 200. Calculations can be performed, in various embodiments, using various algorithmic models. In one example, sensors monitor armature voltage Varm of the alternator 218, a field current Ifld in the field circuit of the alternator, and a rotational velocity ω of the flywheel 216. The rotational acceleration α of the flywheel 216 can be estimated from repeated measurements of the flywheel rotational velocity ω. Additionally, an armature current Iarm of the alternator 218 can be calculated based upon the sensed value of the armature voltage Varm. In an example, the power output of the alternator 218 can be between zero (0) and one (1) kilowatt.
The programmed physical characteristics of the exercise machine 200 can include resistance of the electrical load bank, Rload (which may in, various embodiments, be a controllable quantity); inductance of the field circuit, Lfld; resistance of the field Rfld; resistance of the armature, Rarm; inductance of the armature, Larm; mutual inductance between the armature and the field circuit, Laf; moment of inertia of the flywheel 216, J; and a number of drivetrain damping coefficients, e.g., b0, b1, and b2, so called because they appear in torque terms proportional to powers of ω. The values for Laf, J, Rarm, Rfld, Rload, b0, b1, and b2 are system characteristics initially identified during design of the exercise machine 200 and can be refined for each individual exercise machine 200 during a calibration process at or near the end of the manufacturing process, or at a later time.
In an example, the controller 232 can use the described values to calculate an estimate of the applied torque value Tathlete, which can be a sum of a mechanical torque, Tmech, and an electrical torque, Telec, using the following equations:
where Tmech is the sum of an inertial term and several drag terms, i.e.,
Note that Telec is proportional to Ifld, where Ifld is a readily controllable quantity, as explained above. Also, Iarm may be varied by changing the net resistance of the electrical load bank.
It is to be understood that EQUATIONS 1-3 are illustrative only, and that additional or other variables and equations can be employed to estimate Tathlete, and that other or additional variables can be sensed to accomplishing the same purpose without departing from the spirit of this disclosure. For example, the current of the armature, Iarm, can be sensed or measured and used directly in the above Telec equation without sensing or measuring Varm first and then calculating Iarm using Ohm's law. Sensing or measuring any number of variables is anticipated by the present disclosure. Persons having skill in the art of electrical engineering will readily understand the above calculations, and also that it is possible to measure a variety of variables to use in various calculations to accomplish the same purpose.
The calculated value for Tathlete (e.g., athlete activity; torque actually applied by the athlete) is applied to a dynamic model of a desired load profile to arrive at the appropriate load that the operator should experience. A dynamic model of the exercise machine 200 itself is then referenced for converting the desired load to an appropriate actuation command. For the purposes of this disclosure, an appropriate actuation command can be any number of actions taken by the controller 232 to selectively modify the load experienced by the operator 202.
In one example, the controller 232 can change at least one value used in one or both of the expressions for Tmech and Telec shown above. Changing at least one of the values in these equations changes the load experienced by the operator 202. For example, the controller 232 can alter the value of one or more of the values J, b0, b1, or b2 of the Tmech equation so that the exercise machine 200 feels, with respect to load, like an actual mobile bicycle; or, as noted above, Ifld and/or Iarm may be altered. The closed-loop methods of load control employed in various embodiments of the disclosure allow alteration of operator load at electronic speeds and thus, advantageously, the simulation of constant, slowly shifting, and rapidly shifting real-world loads.
Moreover, the apparatus and methods of various embodiments enable the controller 232 to alter the torque load experienced by the operator to match a selected simulated-apparatus profile. In an example, the calculated Tmech has to make up any difference between Telec and the desired torque load Tathlete based on the selected profile and state of trainer operation. As shown in EQUATION 2, the value of Tmech is a function of velocity and acceleration of the flywheel 216. One method of providing a different load for the exercise machine operator (e.g., operator 202) is to change at least one of the damping coefficients for a given velocity of the flywheel 216 and then change at least one of Ifld and Iarm to alter Telec so that the actual value of Tathlete is equal to or is substantially equal to the desired value of Tload.
Moreover, the controller 232 can be programmed to replicate the loads felt by an operator on any number of actual bicycles, watercraft, or exercise machines (depending on trainer type). The exercise machine 200 can mimic any number of athletic apparatuses, with each mimicked device being represented by a different profile that can be stored in the memory of the controller. Each profile can include changes to any number of the J, b0, b1, and b2 values.
In one example, the process for mimicking a particular device can be described as follows, in an example where the exercise machine 200 of
For example, the flywheel 216 can be specified, designed, and/or constructed to have particular inertia value J. In some examples, the b0, b1, and b2 damping coefficients are almost negligible. Additionally, in some examples, there can be additional damping coefficients; however, these terms are often not significant enough to substantially affect the calculation result. The controller 232 will then calculate a value for Tathlete (the load felt by the operator) using the constants for Laf, J, b0, b1, and b2.
The controller 232 then accesses a desired torque value for the particular desired profile (e.g., a four-person scull) selected by the operator. Because Telec is controlled, the controller 232 will conduct calculations to augment the Telec value with a new Tmech value such that the Tathlete is equal to or is substantially close to the desired torque value for the desired profile. In one example, the same Tmech and Telec equations are used by the controller 232, except that new values for the inertia and damping coefficients replace the previous ones, for example the equation can use J′, b0′, b1′, b2′ rather than J, b0, b1, and b2 to calculate a value for Tmech. The controller 232 will then add the Telec and the new Tmech torque values to ascertain whether actual Tathlete is equal to or is substantially close to the desired Tathlete. If not, the controller 232 can re-calculate Tmech using yet another set of inertia and damping coefficients. This process can continue within the controller until an appropriate Tathlete value is attained.
The controller 232 then applies the known inertia and damping coefficients to the Tmech equation to make the exercise machine 200 “feel like” the selected apparatus (e.g., a four-person scull). Each actual apparatus moves very differently on the water; e.g., it is to be appreciated that a one-person apparatus can exhibit relatively fast acceleration values and have a relatively low top speed on the water. Another apparatus, such as an eight-person apparatus, can exhibit relatively slow acceleration and have a relatively high top speed. The Tmech equation shown above can mimic each of the apparatus and their various characteristics with the proper values for J, b0, b1, and b2.
It is to be appreciated from the above equations that the torque load the operator experiences (Tathlete) is a function of the current at the armature (Iarm), which can be calculated after measuring or sensing Varm, and of the current through the field circuit (Ifld), which is a closed loop control variable. The controller 232 constantly measures and adjusts the Ifld modulator to produce the desired Tathlete. In one example, if Ifld is higher than the value required to replicate the selected profile, the controller 232 can decrease the duty cycle of Ifld to reduce the average (effective) Ifld. Similarly, the controller 232 can increase the duty cycle if the value of Ifld is too low. The controller 232 can monitor and adjust Ifld at relatively short intervals such that Ifld is adjusted as needed. In this way, Ifld is controlled such that the exercise machine 200 can approximate real-world conditions of various rowing apparatus as described above.
Referring again to
The exercise machine 200 can preferably communicate with at least one additional associated exercise machine via the network 130 of
Various suitable algorithms can incorporate various data items, including activity data from multiple machines, to achieve desired closed-loop control characteristics with the apparatus and methods of the present disclosure. In an example where machine 200 is one of P comparable exercise machines (e.g., with similar flywheels) combined virtually in a group-training fashion, as on a multi-person scull, using the inertia J of the flywheel 216 and the desired acceleration αdes of the flywheel 216, one can calculate a net torque, Tnet, acting on the flywheel 216 using
Solving for desired rotational acceleration αdes, one obtains:
If the desired rotational velocity of the flywheels of the P machines is ωdes, then, integrating with respect to time,
ωdes=∫αdes
This can be combined with all known applied torque(s) from each of the associated exercise machines to determine a desired rotational velocity, ωdes, for the flywheels using desired damping coefficients b0des, b1des, and b2des:
In EQUATION 4, Jdes is the desired flywheel inertia, ω is the actual rotational velocity of the flywheel, and
where Tathlete(i) is the torque applied by the ith of the P athletes.
Tnet in the above description is the desired Tload, and, the Tcrew term can be calculated for an arbitrary number of athletes as appropriate.
EQUATION 4 can be used to directly perform closed loop speed control. Any number of closed loop control methodologies can be applied to achieve desired closed loop control characteristics. Examples of closed-loop control methodologies can include, but are not limited to, proportional-integral-derivative control, lag-compensation, h-infinity, and state-space.
Reference is now made to
For example, in
The server 418 comprises software programs that implement various functional aspects of the system 400. These programs can include a database app 430, which maintains the contents of the database layer 420 and retrieves information for serving to trainers and other devices as needed; a simulation app 432, which calculates the team algorithm, calculates the activities of simulated operators, and performs other calculative tasks; an administrative app 434, which enables a master user to act at an operations management level; a developer app 436, which enables access to the application programming interfaces of the system for application development; and a root app 438, which enables master control over other user categories and access to everything contained in the database layer 420. In various embodiments, the functions realized in the illustrative system 400 by the database layer 420 and the apps 430, 432, 434, 436, and 438 are realized by a differently organized set of applications or software modules. Moreover, the system 400 can comprise one or more additional computing devices, e.g., a coach device 440 supplying authorized access to a “coach,” i.e., user having coordinative, administrative, or oversight powers. The coach device 440 may in various embodiments or modes of operation of system 400 be the computer device of one of the trainers (e.g., trainer 402), a laptop or desktop, or a mobile computing device. The network 416 may also communicate with other networks and with devices connected thereto. The collection, analysis, presentation, and transmission of athlete performance data is carried out on the server 418. In alternative embodiments, some, or all, of such actions may be carried out on a server or processor located on one or more of the trainer devices, e.g. trainer 402.
In an illustrative mode of operation of system 400, a coach device 440, communicating with the server 418, is authorized to work with some subset of the N trainers logged on to the system 400. For example, the coach device 440 may be one of a limited number of coach devices at a university authorized to access the system 400 as part of a paid subscription service. The user of coach device 440, employing a software capability running on their computer device, assigns up to N operators to be members of one or more virtual teams. The user of the coach device 440 also specifies conditions that will influence the load profile of the run (e.g., race topology, wind conditions, race duration). The server 418 sets up a computational model (e.g., race algorithm), to be executed by the simulation app 432, with parameters set and/or updated during simulation to reflect the choices transmitted by the coach device 440 and by the participating trainers, as well as other pertinent variables (e.g., trainer-specific mechanical characteristics, virtual landscape data, aerodynamic interactions of virtual athletes) and employing also as inputs activity data from the participating operators, including data on effort, steering, and braking. The run begins on a signal from the coach device 440 or at a set time, whereupon activity data from the trainers begin to be transmitted to the server 418 through the network 416. The simulation app 432 of the server 418 computationally models the behavior of the operators' virtual apparatuses based on its various parameters and inputs, and transmits instructions for each of the participating operators' trainers accordingly to modify the loads experienced by the operators (e.g., by increasing or decreasing the current to a generator winding). The run terminates at another signal or time. The server 418 records in its database layer 420 all data received or generated by the server 418 during the course of the run, which may include activity data from the trainers, operators' physiometric data that may have been transmitted through the network 416 from activity monitors, race outcomes, and the like.
The topology of
In another illustrative mode of operation of system 400, one or more virtual athletes may be simulated by the server 418. At one operative extreme, all participating athletes are real and no athletes are simulated; in various mixed cases, one or more real athletes and one or more simulated athletes are employed; and at another extreme, all athletes are simulated. The latter mode may be used for training of coaches, investigation of various styles of team formation and competition tactics, and other purposes.
Reference is now made to
The previously described communication pathway enables the server to communicate with each exercise machine (e.g., each trainer) to communicate a virtual position of each exercise machine relative to every other exercise machine. In many examples, the virtual position of each exercise machine can be a virtual measurement of a virtual distance between each exercise machine and every other exercise machine. The virtual measurement can be measured in a direction at least one of parallel to a direction of a race event or perpendicular to the direction of the race event. In other words, the virtual measurement can be generally along a main front-to-back axis of the bicycle, or the virtual measurement can be perpendicular to the main front-to-back axis of the bicycle (i.e., side-to-side distance between exercise machines). In some examples, the virtual position includes a virtual measurement of a virtual distance between each exercise machine and every other exercise machine while including a component (e.g. a vector component) measured in a direction perpendicular to a direction of a race event and a component measured in a direction parallel to the direction of a race event.
As such, the server receives data from each exercise machine and extracts performance information from the data. The server then transmits the performance information to each exercise machine to alter the resistance of the mechanical energy storage device based upon the relative position of each exercise machine to every other exercise machine.
The lower part of
Effort curves in the lower part of
In a first time interval 500, Riders 1-9 ride at constant, common velocity in a peloton with constant relative positions. During this interval, all athletes exert constant average effort (power output) on their stationary bikes: that is, each stationary bike presents its rider with a constant resistance (load). Referring to
As depicted for timepoint A, Rider 1 is in the lead of the virtual group, Rider 2 is just behind and to the left of Rider 1, and Rider 5 is in the center of the peloton. Consequently, Rider 1 experiences more aerodynamic drag than any other Rider, Rider 2 experiences slightly less drag than Rider 1, and Rider 5 experiences significantly less drag than Rider 1 or Rider 2. Accordingly, to maintain constant velocity, Rider 1 exerts a level of effort 502 that is higher than Rider 2's level of effort 504, and Rider 5 exerts a level of effort 506 that is lower than that of the Rider 1 or Rider 2.
In some examples, the computer-implemented system can simulate a situation in which a portion of the peloton, or group of riders, move faster than another part of the group. This phenomenon can be termed a break away event. In such situations, perhaps when at least one live rider or even at least one simulated rider begins the break away event, a first number of exercise machines breaks away from a second number of exercise machines. At this time, the server transmits a signal to increase the resistance of the mechanical energy storage devices of the second number of exercise machines of the plurality of exercise machines as the virtual location of the first number of exercise machines increases in distance from the virtual location of the second number of exercise machines in order to replicate real world conditions of break away events
After timepoint A, Riders 1, 2, and 3 break away from the peloton, moving in unison as the peloton continues at constant velocity. To do so, these three Riders (the breakaway group) must accelerate to a higher velocity than that of the peloton for some interval of time. In practice, the three operators accelerate by increasing their power output—pedaling more vigorously. This change in power is detected by each stationary bike (e.g., by measuring torque applied to and rotational speed of the flywheel 216 of
The acceleration of Riders 1-3, followed by a period of increased velocity and then by deceleration to the original velocity (i.e., velocity of the peloton), occurs during interval 508, as depicted. By timepoint B, Riders 1-3 have formed a cluster ahead of the peloton that is moving at the velocity of the peloton. During interval 508, Rider 1 exerts a constant level of effort 510 and Rider 2 exerts a constant level of effort 512. These levels of effort 510, 512 reflect the higher power needed to accelerate and to travel at some higher velocity (with higher drag). When Riders 1-3 reach a desired position ahead of the peloton, as estimated by them based on experience or communicated explicitly to them via audiovisual feedback, they decelerate to the (still unchanging) velocity of the peloton. The operators accomplish this deceleration by decreasing their effort (pedaling more slowly), which is sensed by the stationary bike and communicated to the Simulation App as described hereinabove for their earlier acceleration. Stationary bike mechanical resistance is adjusted smoothly and realistically during these maneuvers to reflect acceleration, changing aerodynamic drag, and other pertinent variables.
It will be clear that in the remainder of this discussion of
Reference is again made to
During interval 522, which includes timepoint B, all riders' velocities are equal again and their efforts have stabilized. Rider 1 now exerts a level of effort 516, still the highest of any rider as an aerodynamic consequence of their leading position. Rider 1's effort 516 during interval 522 is higher than their effort 502 during interval 500 because he drag on Riders 1-3 is no longer reduced by the following group, Riders 4-9. (A cyclist's drag is reduced both by following and by being followed, although following typically reduces drag more.) During interval 522, Rider 2 exerts an effort 518 slightly lower Rider 1's effort 516, experiencing some reduction in drag by riding behind and slightly to the left of Rider 1. During interval 522, Rider 5's effort 520 is higher than their effort 506 during interval 500 because during interval 522, Rider 5's drag is no longer reduced (or is reduced less) by following Riders 1-3. The efforts of Rider 4 and Rider 6, not depicted in
During interval 522, all riders maintain constant velocity and effort in the positions shown for timepoint B. At the end of interval 522, Riders 1-3 then form up in file, as shown for interval 528 (which includes timepoint C) and interval 530. To assume the new formation, at least Rider 2 and Rider 3 must use their trainers' steering controls to maneuver in the virtual environment. Shifting to a file formation results in a more significant drafting effect for Riders 1-3. Consequently, in interval 528 Rider 1's effort declines to a new level 524 and Rider 2's effort declines to a new level 526. The effort 520 of Rider 5 and the efforts of Riders 4-9 (not depicted) remain approximately unchanged. Rider 2 experiences enough drag reduction by riding directly behind Riders 1 and 3 that Rider 2's effort 526 is slightly less than Rider 5's effort 520.
During interval 528, all riders ride at constant velocity with constant effort. At the end of interval 528, the simulation app 432 of the training simulation system 400 of
Having made the flexible, simulative nature of various embodiments clear, mention is now made of three novel, advantageous approaches to off-road training for competitive sports, or to personal entertainment, made possible by some embodiments of the disclosure. There is no restriction to these three training approaches; it will be clear many more may be readily imagined, all depending on the novel capabilities of the disclosure.
In a first approach, freestyle, exemplified by the virtual race of
In a second approach, assignment, a pre-set training profile or algorithm for the virtual race, or a coach supervising the race, or a supervised algorithm, sets goals for one or more operators (e.g., athletes, cyclists, etc.) which they then strive to achieve. In an example, a group of simulated riders breaks away from a peloton comprising human athletes participating in the virtual race, and feedback to individual athletes on their trainers tells them what performance to strive for to achieve a certain goal, e.g., for the whole peloton to catch up with the breakaway group, or for selected riders to catch up with the breakaway group while others remain with the peloton or draft behind certain other riders. Performance may be specified as power output in Watts, velocity achieved within a certain number of seconds, pedal revolutions per minute at a specified trainer resistance level, or in other terms. The rider may be presented with options for achieving the target performance, and choose between them. In the assignment approach, athletes are challenged to achieve specific performance levels: they learn what it takes to be part of a breakaway group, or to be the lead rider in a peloton, or the like.
In a third approach, emulation, the athlete chooses or is assigned a performance profile (quantitative specification of effort, speed, technique) that corresponds to a model athlete, e.g. a winner of the Tour de France or other professional contests, or to a simulated model athlete. In examples of the third approach, the server extracts a virtual location of a lead exercise machine relative to every other exercise machine from the data. The server then transmits a race profile of the lead exercise machine and the virtual location of the lead exercise machine to every other exercise machine in order to alter the resistance of the mechanical energy storage device. This alteration of resistance occurs as the virtual location of the lead exercise machine breaks away from every other exercise machine in order to urge an athlete riding an individual exercise machine to match at least one of an exercise tempo or a resistance level of the lead exercise machine in order to maintain a separation distance between the lead exercise machine and the individual exercise machine.
For example, the profile of a model athlete is a numerical model that specifies approximately how that athlete makes performance decisions in competition (e.g., how fast to pedal, how much power to produce at what times and under what conditions, when and how to break away). A profile may be derived from records of the model athlete's performances over a number of real-world races. Persons familiar with modeling and control methods will be aware of a number of alternative computational methods for deriving such a profile (e.g., applying deep learning methods of artificial intelligence to videos and other records of model athlete performance) and specifying the profile's outputs so that, in effect, a virtual rider simulates the human model athlete. In the emulation approach, the athlete in training specifically seeks to emulate the performance profile of the model athlete. Since model athletes have distinguishable styles, some model athletes may be more or less challenging, appropriate, or informative for a given trainee to emulate (“Do I ride more like Lance Armstrong or Gino Bartali?”): this can be discovered using the third training approach repeatedly with various race conditions, model athletes, and other factors. In this approach, the athlete discovers or develops their own most competitive style by performatively “studying” the styles of model athletes.
In yet further examples, the described computer-implemented system can enable a plurality of athletes to select a group-oriented goal such as a total group power output (e.g., a team load value) for a particular ride. In this way, a group of athletes can still include athletes who may have a relatively low experience level or a relatively low athletic performance expectation. For example, a plurality of athletes can use a plurality of exercise machines to produce a plurality of work output values (e.g., ride at a common time to produce work output values). The plurality of work output values are added at the server in order to compare a total work output to the team load value. At least one exercise machine receives an altered resistance value as haptic feedback based upon the difference between the total work output and the team load value to achieve the team load value.
Having described the foregoing embodiments of the disclosure, it will be apparent to those of ordinary skill in the art that other embodiments incorporating the concepts disclosed herein may be used without departing from the spirit and scope of the disclosure. The described embodiments are to be considered in all respects as only illustrative and not restrictive.
This application claims priority to both PCT application PCT/US2022/13434, filed on Jan. 22, 2022 which claimed priority to U.S. Provisional Patent Application No. 63/140,538, filed on Jan. 22, 2021, both entitled “APPARATUS AND METHODS FOR INTERACTIVE MULTI-USER RESISTANCE TRAINING ON EXERCISE MACHINES,” both of which are hereby incorporated by reference herein.
