CLOSED-LOOP POWER DISSIPATION CONTROL FOR CARDIO-FITNESS EQUIPMENT | Patent Publication Number 20090118099

US 20090118099 A1
Patent Number-
Application Number11935080
Filled DateNov 5, 2007
Priority DateNov 5, 2007
Publication DateMay 7, 2009
Original Assignee
Current AssigneeTriplepoint Capital Llc
Inventor/ApplicantsJohn Fisher
Joel Jensen
Keith Thompson
Steve Anderes
International
1
A63B
National
1
482/5
Field of Search
0

Various embodiments of the present invention provide (a) an inexpensive apparatus enabling the measurement of power dissipated by the rider of a cardio-fitness station (or any other stationary bicycle) that does not depend on manufacturing tolerances or machine condition variations, and (b) a method of using the data measured by such an apparatus to improve the accuracy of exercise condition settings by implementing the invented apparatus into a closed-loop control system which improves the quality of the exercise experience and enhances the adoption of exercise on a cardio-fitness station employing this as a community activity.

  • 9. A method of power level control, the method comprising:nreceiving a request indicating a a target power level;determining a current power level based on identifying one or more of torque exerted on a flywheel and an angular velocity of the flywheel; andadjusting an electric current to change a resistance to movement of the flywheel;wherein the electric current is adjusted based on a difference between the target power level and the current power level; andrecomputing an updated current power level and adjusting the electric current such that the updated current power level substantially approaches the target power level.
  • 10. A machine-readable medium embodying instructions, the instructions, which when executed, causing a machine to perform a method comprising:nreceiving a request indicating a a target power level;determining a current power level based on identifying one or more of torque exerted on a flywheel and an angular velocity of the flywheel;adjusting an electric current to change a resistance to movement of the flywheel;wherein the electric current to be adjusted based on a difference between the target power level and the current power level; andrecomputing an updated current power level and adjusting the electric current such that the updated current power level substantially approaches the target power level.
See the invalidated claims, subscribe to our Concierge Program.
View Concierge Program
Subscription-Only
View Concierge Program
Subscription-Only
View Concierge Program
RELATED APPLICATIONS

This application claims priority to U.S. Utility patent application Ser. No. 11/766,312, filed Jun. 21, 2007, and entitled “Closed-Loop Power Dissipation Control For Cardio-Fitness Equipment and to U.S. Provisional Patent Application No. 60/817,657, filed Jun. 28, 2006, and entitled “Closed-Loop Power Dissipation Control For Cardio-Fitness Equipment,†by John Fisher et al., and are hereby incorporated herein by reference.

This application is related to and cross-references U.S. application Ser. No. 11/433,778, filed May 11, 2006, and entitled “Cardio-Fitness Station With Virtual-Reality Capability,†by John Fisher et al., the contents of which application are hereby incorporated by reference.

BACKGROUND

1. Field of Invention

This invention relates to stationary exercise equipment and power dissipation control used by such equipment. More specifically, the invention relates to closed-loop power dissipation control for cardio-fitness equipment.

2. Background of the Invention

A major sports equipment industry has developed over the last decades round providing fitness equipment for home and indoors exercises based on so-called stationary exercise equipments, which can be but are not limited to stationary exercise bicycles, in which the action of pedaling is used to dissipate power by the person (rider) exercising. The resistance to pedal rotation is allowed for power dissipation, which is an integral part of the exercise. State of the art exercise equipments often feature heart-rate monitoring, entertainment, and a varying degree of pedaling resistance, which is used to control the amount of power the rider dissipates while pedaling.

On many stationary exercise equipments, the power necessary to pedal can be set directly to a predetermined level by the rider, yet on some, the power can be set in terms of real-world parameters, such as slope of a hill, to give the rider the impression that he or she is riding a real bicycle up a hill. Cardio-fitness stations, the most advanced exercise tools, offer virtual reality capabilities in which the rider interacts with a virtual environment shown on a video monitor and experiences a virtual bicycle ride through a predetermined landscape with hills, valleys, and road obstacles. Such feature has given rise to competition between riders exercising on two cardio-fitness stations, i.e., the riders can operate separate cardio-fitness stations to ride jointly in a race through the same predetermined virtual landscape. Furthermore, with the advance of exercise equipments, many riders have increased their demands for accurate monitoring of their performance and performance history.

A fact not immediately apparent to an average rider of stationary equipments is that their performance, i.e., the resistance to pedal these cardio-fitness stations under a specified setting or virtual terrain slope, is not always consistent among the stations. This is noticeable when one rider is racing another rider riding another unit and the other rider may have an easier time making it to the finish line. Furthermore, for a given constant cadence and same resistance setting, stationary equipment will deliver pedal resistance that depends on the history of the cadence and torque in a practically unpredictable manner due to the cumulative effect of machine temperature and wear.

All these problems arise from the fact that pedal-rotation-resistance mechanism implemented in present-day exercise equipments is not intended for such precise setting and repeatability of pedal torque, which has been the choice of the manufacturers for cost reasons and the fact that it was not required by the riders. The source of the drift and unit-to-unit variation in the relationship between the setting of pedal resistance and the actual value of resistance experienced by the rider comes from the drift in the performance of mechanical and electrical elements, for non-limiting examples, manufacturing tolerance, mechanical wear, and heating effects on the equipment. Such variation ultimately yields unsatisfactory accuracy of power dissipation and an incorrect assessment of total amount of work that rider has performed during his or her exercise session, which makes it next to impossible to execute a fair race between riders on two separate cardio-fitness stations.

SUMMARY OF INVENTION

One embodiment of the present invention provides an inexpensive apparatus enabling measurement of power dissipated by the rider of a cardiofitness station (or any other stationary exercise equipment) that does not depend on the manufacturer, manufacturing tolerances, or machine condition. In addition, a method of using the data measured by such an apparatus to improve the quality of the exercise experience is provided.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention is illustrated by way of example in the accompanying drawings. The drawings should be understood as illustrative rather than limiting.

FIG. 1 illustrates an exemplary cardio-fitness station with virtual-reality capability in accordance with one embodiment of the present invention.

FIG. 2 illustrates an exemplary cardio-fitness station in block diagram form in accordance with one embodiment of the present invention.

FIG. 3 illustrates an exemplary pedal assembly function diagram in accordance with one embodiment of the present invention.

FIG. 4 illustrates an exemplary magnetic resistance device in accordance with one embodiment of the present invention.

FIG. 5 further illustrates an exemplary magnetic resistance device in accordance with one embodiment of the present invention.

FIG. 6 illustrates power dissipated by an exemplary flywheel under varying separation distances in accordance with one embodiment of the present invention.

FIG. 7 illustrates measured flywheel torque as a function of electromagnetic current for varying separations distances in accordance with one embodiment of the present invention.

FIG. 8 illustrates an exemplary process for control of pedaling resistance in accordance with one embodiment of the present invention.

DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION

A system, method and apparatus are provided for closed-loop power dissipation control for cardio-fitness equipment. The specific embodiments described in this document represent examples (e.g., stationary exercise bicycles) or embodiments of the present invention, and are illustrative in nature rather than restrictive. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the invention. It will be apparent, however, to one skilled in the art that the invention can be practiced without these specific details. In other instances, structures and devices are shown in block diagram form in order to avoid obscuring the invention.

Reference in the specification to “one embodiment†or “an embodiment†means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. The appearances of the phrase “in one embodiment†in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Features and aspects of various embodiments may be integrated into other embodiments, and embodiments illustrated in this document may be implemented without all of the features or aspects illustrated or described.

Some embodiments of the present invention provide (a) an inexpensive apparatus enabling the measurement of power dissipated by the rider of a cardio-fitness station (or any other stationary exercise equipment) that does not depend on manufacturing tolerances or machine condition variations, and (b) a method of using the data measured by such an apparatus to improve the accuracy of exercise condition settings by implementing the invented apparatus into a closed-loop control system which improves the quality of the exercise experience and enhances the adoption of exercise on a cardio-fitness station employing this as a community activity.

Some embodiments of the present invention provide (a) a device for measuring the output power of a stationary exercise equipment (b) use of the device to calibrate the resistive force applied to the pedals of a stationary exercise equipment according to a present or programmed value, and (c) use of the device to enhance the quality of exercise experience on cardio-fitness stations with virtual-reality capability. There are several ways known in the industry that enable the stationary exercise equipments to provide and control resistance to pedaling. For a non-limiting example, the rotational pedal motion may be transferred to a rotating flywheel whose rotation is slowed down by mechanical friction. The rotation of the flywheel may be converted to electrical energy using an alternator, and then the generated power is dissipated on an electrical load. Finally, the resistance to the rotation of the flywheel may be provided by a magneto-resistive device in which the eddy currents induced by an electromagnet give rise to magnetic fields that oppose the flywheel rotation, thereby slowing the flywheel down. The type of control of pedaling resistance may include discrete levels of resistance settings available as a switch or a level accessible to the rider, or is controlled by a computer program which is guiding the rider (person exercising) or is being guided by the rider, as in an exercise session on a cardio-fitness station with virtual reality capability. In order to determine the power output by the rider, one has to determine the product of the torque applied on the pedals and the angular velocity of the pedals, from now on referred to as cadence. Neither the torque exerted on the pedals nor the cadence is uniform in time—both depend on the pedal position (angle) with respect to the rider's legs (or ground) and/or the condition and the performance of the rider. The total energy (kcal) dissipated can be found by integrating (calculating the integral of) the torque and the instantaneous cadence. There are a number of ways of dissipating the rotational power delivered by the pedals practiced in the industry. Most of the ways offer options for adjusting the amount of resistance to pedal rotation. Three common ways for dissipating pedal power and the associated mechanisms for adjustment of resistance are presented here. The first example is by dissipating the pedal power on a flywheel which is being slowed down by a belt, wherein the resistance to rotation is adjusted by tightening or loosening the belt placed around the flywheel. Although options for adjusting the resistance are provided, there is no precise measurement of the torque induced with the belt, and hence no attempt is made to correct the tightness of the belt to meet the setting. The second and third examples of ways to dissipate pedal power commonly practiced today involve conversion of pedal rotational energy into electricity and then adjusting the dissipation of the electrical power. The second example is involved in stationary fitness bicycles that use an alternator to convert mechanical energy into electrical energy, and then dissipate this electrical energy on an electrical (resistive) load. The adjustment of dissipated power is achieved via adjustment of the magnitude of the electrical current through the resistive load where the generated electrical power is converted to heat. The third example of a way to dissipate power, recently more commonly used, is to use a metallic flywheel and adjust the strength of a magnetic field through which at least one part of the flywheel is passing as it rotates. The magnetic field established by an electromagnet induces eddy currents in the flywheel and the induced currents dissipate energy on the electrical resistance in the flywheel. The power dissipation heats the flywheel, while the eddy currents establish a magnetic field which opposes the rotation of the flywheel, thereby exerting resistance to rotation experienced (caused) by the rider. The adjustment of the pedal resistance (torque) is performed by adjusting the current flowing into the electromagnets. Fitness equipment that uses this type of power dissipation method is commonly referred to as equipment with a magnetic resistance device (MRD). The way above approaches are generally implemented is that for a particular design, the pedal resistance is experimentally evaluated in advance for every cadence and resistance setting, and used in the form of a look-up table or a formula based on a fit to the experimental data. This is generally done for every design, namely, an identical formula or look-up table for a specific model, but is not cost effective to evaluate on every unit a company ships. Even if this were done for every unit, the systematic variation and wear on the equipment could not be predicted, and therefore the look-up table would not be solving the entire problem—it would drift out of sync over time. In the case of an MRD, the variation in the size of the gap between the flywheel and the electromagnet produces most dramatic changes in the relationship between the electromagnet current and the resistive force. This is because the gap, which is air filled, dramatically impacts the magnetic circuit made up from the electromagnet and the flywheel. The gap between the magnet and the flywheel changes due to manufacturing tolerances and the temperature of the flywheel. As the temperature of the flywheel rises, it expands and closes the gap between the flywheel and the magnets, thereby increasing the strength of the resistance to rotation for a given current. The value of the flywheel resistance affects the rate of heating and varies from flywheel to flywheel. Due to the large thermal capacity of the flywheel, the temperature depends on a long history of pedaling at any time. These factors make the relationship between the pedaling resistance and the current energizing the electromagnets very difficult to predict and repeat. Consequently, the tracking between the pedal resistance setting and the actual value of pedal resistance is insufficient to produce consistent exercise results and/or fair competition done on two cardio-fitness stations of the same design. In all of the above methods for providing pedal resistance, the value of the resistance is set by the rider or a computer program, but it is not measured to check the accurate value of the resistance and no attempt is made to make correction to the quantity that controls the resistance (electrical current of the electromagnets in the MRD, for a non-limiting example). This is a potential disadvantage of all commercially available stationary bicycles. Today, there are many options for measuring power and torque accurately and researchers have gone through development of experiments and tools to provide such experiments. See, for a non-limiting example, Bicycle Science by David Gordon Wilson (3rd edition, The MIT Press, Cambridge, Mass., 2004), which is incorporated herein by reference. However, these tools are used in research environments for monitoring and have not been manufactured in a form suitable for commercial products. The primary reasons for this are cost and complexity needed to implement a sophisticated power monitoring system. In addition, it has never become apparent that an accurate calibration of exercise equipment would be needed.

Except for bicycles used for bicycling-science research and research in human power, the control of pedaling resistance described is typically done in so-called open-loop control. Open-loop control is a control architecture that involves setting a process parameter to a particular predetermined value depending on another process parameter without asking or taking into account the result. A control mechanism that does exactly the same, but also takes into account the result to make fine adjustment of the parameter set is called closed-loop control. A non-limiting example of open-loop control is night lights that come on when the sky darkens. In the case of a pedaling resistance, it means that a setting may be applied for a predetermined amount of resistance torque, but the torque that is actually experienced is not measured in real time and no correction to the setting is available. What distinguishes the open-loop control from closed-loop control is the presence of feedback in closed-loop control. Open-loop control is used because of its simplicity and lower cost of implementation. It is a practical solution for applications in which output accuracy is not important and where the system can function well without the guarantee that the output will track the input. Simple commercially available stationary bicycles typically satisfy these two conditions. Open-loop control implemented in present-day exercise bicycles cannot correct for the uncertainties in the performance of mechanical and electrical elements, such as, manufacturing tolerance and heating effects on the equipment. This variation results in unsatisfactory accuracy of power dissipation.

The approach presented in the present invention is to implement closed-loop control of power dissipation on any type of stationary exercise equipment and more specifically on cardio-fitness stations with virtual reality capability, and to disclose an apparatus that provides an inexpensive component that measures the power dissipated by the rider during exercise.

In some embodiments, one can implement closed loop control in stationary exercise equipments by measuring the torque and the angular velocity of the flywheel (or at the pedals). A common way to measure the torque is to use strain gauges and angular velocity via markers on the flywheel or the pedal wheel, and then use this information in an electronic (or software) feedback loop to correct the variable that sets the resistance and to make the measured resistance equal to the set value. In one embodiment, a strain gauge and a tachometer are placed on the pedal gear. A tachometer is a device for indicating speed of rotation. In another embodiment, the deflection of a mechanical spring caused by the torque exerted on the flywheel is used to measure the torque exerted on the electromagnets and a counter is used to measure the angular velocity of the flywheel.

In one embodiment, the information gathered can then be used to determine the power dissipation at any time the rider is dissipating power. In another embodiment, the same information is used to correct the original electrical current setting until the power dissipation becomes arbitrarily close to the power setting. The power dissipation setting may be either independently set by the rider or by a computer program running a virtual reality program and the power setting may be dependent on the rider's virtual position in the predetermined landscape and the rider's virtual ground velocity.

By using closed-loop control, the tolerance variations as well as systematic uncertainties present by design, the temperature changes, and gap variations are all eliminated from the relationship between the dissipation setting and the actual power dissipated by the rider of the cardio-fitness station.

Introduction

FIG. 1 shows a photograph of an embodiment of an exemplary cardio-fitness station with virtual reality capability and employs a magnetic resistance component. The illustrated cardio-fitness station is modeled after a real outdoor bicycle, but has elements of stationary exercise equipment. The cardio-fitness station 100 includes handlebars 141, a gear-shifting lever 142, a pedal assembly 130, a seat 121A, a computer 160, and a video monitor 150, all mechanically connected or attached to a frame assembly 110. The rider, desiring to exercise, sits on the seat 121 as one would on a real bicycle and turns pedals 131 while holding the handlebars 141. The video monitor 150 is positioned in the plain view of the rider while the rider is seated on the seat 121. The rider may watch the images on the video monitor 150, listen to sounds coming from the headphones (not shown), and optionally speak into a microphone (not shown).

In some embodiments, the computer 160 runs a virtual reality program and accepts input from the rider exercising via the position of the handlebars 141, the momentary gear number via the motion of the gear-shifting lever 142, and the rotation of the pedals 131. The listed input parameters are used to determine the motion of the rider's own virtual bicycle in the virtual landscape. Exercise parameters used to follow the rider's actions include (a) angular velocity of pedal rotation Ω, also referred to as cadence, (b) angular position of the handlebars, (c) gear number, and (d) the history of all of those parameters. Turning the stationary bicycle pedals 131 by the rider results in forward motion of the rider's own virtual bicycle in the virtual environment tracked by the computer running a virtual reality program. Steering the handlebars 141 on the cardio-fitness station results in rider's own virtual bicycle turning left or right in the predetermined virtual landscape. (Angular velocity is the rate of rotation around an axis usually expressed in radians per second (rad/sec) or revolutions per minute (RPM)). The predetermined landscape displayed on the video monitor is computer-generated or is a real reconstructed landscape.

In some embodiments, the rider steers through a path y(x) with predetermined length and elevation profile, z(x, y), where x and y are horizontal coordinates used to define the location of the rider's own virtual bicycle in the predetermined landscape tracked by the virtual-reality program running on the computer. Such a path with predetermined length is referred to as virtual exercise route (VER). The VER exhibits upward or downward slopes. The slope at a position x, y is determined by taking gradient of the elevation profile ∇z(x, y) and is expressed as s≡sin θ, where tan θ=|∇z(x, y)|. If the elevation of the path in the virtual environment increases as the virtual bicycle is moving forward, the slope is said to be positive or upward (s>0) and the torque resisting pedal rotation is increased proportionally to the slope. If the elevation of the path in the virtual environment decreases as the virtual bicycle moves forward, the slope is said to be negative or downward (s<0) and the slope-related contribution to the resistance to pedal rotation is set to zero.

As the rider's own virtual bicycle rides along this VER, the virtual reality program communicates to the pedal assembly to set a specific level of pedaling resistance. Accelerating a real bicycle requires additional power from the rider to exert on the pedal to add kinetic energy to the bicycle and rotational energy to the wheels. This power is proportional to the acceleration and is appropriately modeled by the virtual-reality computer program and suitable increased torque exerted on the pedals. Additionally, when the rider rides very fast (tens of miles per hour) most of the resistance comes from the aerodynamic drag, and the pedal resistance must reflect that power loss. The listed variables: terrain slope, aerodynamic drag, and acceleration are variables that influence the pedal resistance. A realistic implementation of these variables on a cardiofitness station involves sophisticated control. The request (or command) for a specific pedaling resistance (or torque) results in an approximate value of the resistance in the controlled device (bicycle). Measurement of torque performed by a magnetic resistance device can enable the computer program to correct the setting and establish correct value of the resistance.

Hardware Description

The hardware concept of some embodiments is intended for use in conjunction with a cardio-fitness station with virtual reality capability, but may be applied to any regular stationary exercise equipment. The functional schematic of a cardio-fitness station with virtual-reality capability is illustrated in FIG. 2. It includes at least the following components and assemblies: a frame assembly 210, a seat 221, a pedal assembly 230, a steering assembly 240, a video monitor 250, and a computer 260. The components and assemblies 221, 230, 240, 250, and 260 are mechanically connected to the frame assembly 210. The purpose of the frame assembly 210 is to support the rider and all of the associated components and assemblies of the cardio-fitness system.

For the purpose of some embodiments, the handlebars 241 provide the rider a facility to steer the direction of the virtual rider bicycle, the gear-shifting lever 242 allows the rider to optimize between pedaling speed (cadence) and pedaling resistance in according to his or her exercise level and ability. Its purpose is identical to the purpose of gear shifting on real bicycles with multiple speeds, for a non-limiting example. The gear-shifting lever includes a movable lever (or handle) that is internally coupled to an electrical switch. The electrical switch is, in turn, sensed by the computer 260 and interpreted as a directive to increment or decrement the gear number to next value up or down, depending whether the lever was moved up or down.

In some embodiments, the computer 260 communicates with the steering assembly 240 via a link 246 and with the video monitor 250 via a link 256. The computer 260 runs a virtual reality program, which sends sensory stimuli to the rider by one or more of: (a) sending images and information to the video monitor 250 via link 256, (b) sending sound to the rider's headphones (not shown) that are plugged into the steering assembly 240 via link 246, and (c) controlling the resistance of the pedal rotation in the pedal assembly 230 via links 243 and 246. Furthermore, the computer 260 acquires exercise parameters by receiving information about the pedal 231 rotation via links 243 and 246, position of the handlebars 241, gear number, and rider program selection from the steering assembly 240 via link 246. (The listed exercise parameters need not be all the exercise parameters that the computer may acquire, the listed parameters are relevant to this description.)

The purpose of the pedal assembly 230 is to provide the rider of the cardio-fitness system a device to exercise leg muscles and dissipate energy while exercising. The pedals 231 are rotated in the same manner as one would when riding a road bicycle. The pedal assembly 230 may include the magnetic resistance device, and will be described in more detail in the next section.

Pedal Assembly

A pedal is a foot lever or treadle by which a part is activated in a mechanism. In case of a road bicycle there are two pedals, the rotation of the pedals sets the bicycle in motion. On a stationary exercise equipment (bicycle), there also two pedals and their rotation is used to provide exercise to the rider of the stationary bicycle in the same sense as rotation of the pedals, i.e., pedaling, the pedals on a real bicycle. The pedals are rotatable, i.e. they can be rotated by the action of feet as on a typical road bicycle or a typical stationary bicycle.

The pedal assembly 300 is explained using FIG. 3. The rider rotates the pedals 301 while exercising. The resistance to rotation of the pedals 301, also referred to as pedaling difficulty, is varied in a controlled manner, thereby delivering to the rider a varying degree of exercise difficulty.

In some embodiments, the pedals 301 are mechanically coupled to pedal pulley 302 and are able to rotate as indicated with arrow 316. The arrow points only in one direction, but the pedals can rotate in either direction. The pedals 301 are mechanically coupled to a magnetic resistance device 303 via a pedal pulley 302 and a belt 304. The cadence sensor 313 is mechanically attached (not shown) to the bicycle frame 314. The frame 314 is a part of the frame assembly 210 shown in FIG. 2.

In some embodiments, the cadence sensor is a tachometer and may be implemented in a number of ways. In one embodiment, the cadence sensor is a counter that counts the number of impulses produced by the passing cadence pulley. In another embodiment, the cadence may be measured on another pulley in the belt system.

An embodiment of the magnetic resistance device 303 includes a flywheel 305, a flywheel pulley 306, a flywheel shaft with clutch 307, at least one electromagnet 308, a spring 310, a shock absorber 311, a deflection measuring assembly 309, a magnet counterweight 320, a magnet support panel 321, and a flywheel rotation sensor 312. FIG. 4 shows a mechanical drawing of an implementation of the embodiment of an exemplary magnetic resistance device and FIG. 5 shows a photograph of the same embodiment of the exemplary magnetic resistance device.

In some embodiments, the flywheel 305, the flywheel pulley 306, and the magnet support panel 321 all are able to rotate around the same rotational axis as the flywheel shaft 307. The magnet support panel 321 rotates independently from the flywheel 305 and the flywheel pulley 306. The flywheel shaft 307 contains a clutch which allows relative rotation between the flywheel 305 and the flywheel pulley 306 only in one direction. When the pedals 301 rotate in the direction indicated by arrows 316, the clutch in the flywheel shaft 307 is engaged and the flywheel pulley 307 and the flywheel 305 rotate accordingly as indicated with arrow 318. When the pedal rotation direction is opposite from the one indicated by arrow 316 or the flywheel rotates faster than the flywheel pulley 306 in the same direction, the clutch in the flywheel shaft 307 is disengaged and the pulley 306 rotates accordingly with the pedals, but the flywheel 305 rotates independently. This ensures that when the pedal rotation 316 suddenly ceases or reduces, the flywheel can continue to rotate due to its inertia.

In some embodiments, both the pedal pulley 302 and the flywheel shaft 307 (together with the flywheel 305 and flywheel pulley 306) rotate around axes that are mechanically attached (not shown) and fixed relative to the bicycle frame 314. The electromagnet 308 and the electromagnet counterweight 320 are mechanically attached to the magnet support panel 321. The weight of the magnet counterweight 320 is approximately equal to the weight of the electromagnet 308. The electromagnet 308 is mounted in the proximity of the flywheel 305 so when the electromagnet 308 is energized (electric current flows through it) the magnetic field from the electromagnet 308 penetrates the flywheel 305 as illustrated with magnetic field lines 319.

In some embodiments, the flywheel 305 is made out of a material that is electrically conductive, most commonly metal. When the flywheel 305 rotates and the magnetic field 319 is present, electrical currents are induced inside the metal flywheel (so called eddy currents). These currents in turn produce a magnetic field that opposes the rotation of the flywheel according to well known laws of physics. The resistance of the metal flywheel 305 is finite. Electrical power is lost in the flywheel (converted into heat) and this manifests itself as resistance to the rotation of the flywheel 305, i.e., slowing down the flywheel rotation. The torque slowing down the flywheel 305 is exerted by the electromagnet 308, which is mounted on the magnet support panel 321 and can rotate around the same axis as the flywheel 305.

In some embodiments, the magnet support panel can partially rotate around the same axis as the flywheel, e.g., it can rotate around the same axis over a limited range. In one embodiment this limited range is 18°. This limited freedom in rotation of the magnet support panel is used to quantify the torque resisting the flywheel rotation. The deflection of the magnet support panel 321 is constrained with the spring 310, which is attached to the bicycle frame 314. Any amount of torque resisting the rotation of the flywheel 305 will stretch or compress the spring 310 proportionally to the magnitude of torque. A deflection sensor 309 is used to quantify the stretching (or compression) 315 of the spring 310. In one embodiment, the deflection sensor is realized as an optical sensor that senses the passing of a perforated screen attached to the electromagnet support panel by counting pulses of light generated by a light-emitting diode on the opposite side of the perforated screen. In another embodiment, there is one deflection measuring spring. In yet another embodiment, more than one spring is used. Both compression and tension springs may be used to accomplish the same function.

In some embodiments, the magnet counterweight 320 ensures that the deflection of the spring 310 does not depend on the angular position of the electromagnet in respect to the axis of rotation. Namely, the torque resulting from the weight of the magnet changes with the angle at which the magnet is positioned—this dependency can be removed by adding a counter weight. The counterweight may be replaced by another (one or more) electromagnets of the same weight to accomplish the same function. Additionally, a single magnet may be used in the device—additional magnets allow for greater effects, but are not necessary to simply achieve the desired magnetic resistance. Moreover, just as the counterweight may be replaced by one or more magnets, the overall system of magnets and the component on which the magnets are mounted may be sized as desired for other design constraints, as long as the center of mass of the system of magnets and mounting component coincides with the axis of rotation of the flywheel.

In some embodiments, the torque experienced by the electromagnet (and the support panel) varies in time. The shock absorber 311 is used to dampen any mechanical oscillations of the magnet support panel (with the electromagnet and the magnet counterweight). A shock absorber is any of several devices for absorbing the energy of sudden impulses or shocks in machinery or structures, and dampens oscillations. Another common word used for shock absorber is damper.

In some embodiments, the described deflection sensor 309 and the spring 310 provide a torque-measuring device. There are numerous ways to realize a torque-measuring devices. In one embodiment, the electromagnet support panel 321 is stationary in respect to the frame 314 and the torque exerted on the electromagnets 308 is quantified by a torque-measuring device including a semiconductor strain gauge disposed between the electromagnet 308 and the frame 314 or the electromagnet support panel 321 and the frame 314. In another embodiment, the torque-measuring device is disposed between at least one pedal and the pedal pulley, thereby directly measuring the torque on the pedals.

In some embodiments, the number of pulleys and belts in FIG. 3 may vary, and the arrangement shown in FIG. 3 involving three pulleys—302, 306, and the tightening pulley. For a non-limiting example, two belts each with two pulleys and one tightening pulley may be used to increase the torque capability of the cardio-fitness station. The ratio of the flywheel angular velocity 318 to the cadence 316 is fixed by the ratio of the perimeters of the pulleys. The typical value of this ratio ranges from 25:1 to 35:1 with the flywheel 305 rotating faster than the pedal pulley 302 (and thus the pedals 301) in some embodiments.

In some embodiments, the power delivered by the person exercising is quantified by measuring the torque exerted onto the electromagnet 308 (i.e., the magnet support panel 321 on the same axis as the flywheel 305) and the angular velocity of the flywheel measured by sensor 312. The flywheel angular velocity sensor 312 measures electrical, optical, or magnetic impulses generated by the passing flywheel and converts the rate of the pulses into angular velocity. There is more than one way that the flywheel angular velocity can be measured. In one embodiment the sensor uses a hall effect sensor and senses the rotation of at least one magnetic disk attached to the flywheel (shown in FIG. 4), where the number of pulses per flywheel revolution is 12. The torque resisting rotation depends on the strength and the distribution of the magnetic field in the flywheel 305, the electrical resistance of the metal flywheel 305, and the diameter and cross-sectional shape of the flywheel.

Exact calculations of the relationship between the torque, the applied magnetic field, and the electrical current that generates the magnetic field are complex and difficult, and are also not necessary to operate the magnetic resistance device effectively. Empirical data, obtained from measurements, even data that is not very accurate can be used efficiently because closed-loop control is available. Generally, the magnetic field density is proportional to the electromagnet current I and inversely proportional to the gap h between the magnet pole and the flywheel. The torque NMRD is proportional to the angular velocity ω of the flywheel and the magnetic field squared This means that increasing the electromagnet current I will increase the torque NMRD. Adjusting the amount of current flowing into the electromagnet results in the adjustment of the resistance to rotation of the flywheel 305, and consequently the resistance to rotation of the pedals 301. If one applies a weak electric current to the electromagnet, the pedals 301 rotate easily. If one applies a large current to the electromagnet, high resistance to rotation of the pedals 301 will be experienced by the rider.

If the gap h between the electromagnet pole and the flywheel changes due to manufacturing tolerance or temperature, the torque NMRD will not be precisely known. Because of this inherent uncertainly, all magneto-resistive devices used in stationary bicycles exhibit imprecise magnitude of the torque NMRD for any given electromagnet current.

This device provides a system for measurement of the torque exerted on the electromagnet and the angular velocity of the flywheel, and thereby determining the power delivered by the person exercising. This measured power enables calibration of the cardio-fitness machines independent of the manufacturing tolerances and temperature of the flywheel, the electromagnets, and independent of any other variables that can fluctuate from machine to machine or from one magnetic resistance device to another.

The described system for monitoring the torque exerted on the flywheel and the flywheel angular frequency may be used to determine the power dissipated by the rider exercising. This system may be used to either monitor the power or as a signal to correct the value of the electromagnet current to adjust the torque on the flywheel to match the value requested or commanded by the pedal-resistance setting set by the rider or set by a computer program.

Also described in this document is a method for using this feature of the cardio-fitness station in conjunction with the virtual-reality capability.

Analysis of Power Dissipation in a Real Bicycle

In a read road bicycle, the gear is characterized by a predetermined transmission ratio GB between the cadence ΩB and the rear wheel rotation ωB: GB=ωB/ΩB. A real road bicycle will have an integer number of gear values—typically between 1 and 15 gears, i.e., GB is an array of discrete rational values: GB (nB), with nB being the gear number (an integer varying between 1 and the maximum number of gears nmaxB) Together with the radius of the rear wheel rB, the velocity of a real bicycle vB is given by vB=2πrBGB(nB)ΩB. The bicycle rider controls ΩB and nB, while the bicycle manufacturer defines rB and the discrete vales of the GB(NB) array for every nB. The instantaneous value of torque NB(t) applied to the pedals multiplied by the instantaneous value of cadence ΩB(t) gives the instantaneous power PB(t) delivered by the biker to move forward with ground velocity vB(t).

In some embodiments, the relationship between the power PB(t), cadence ΩB(t), and gear-number changes in time nB(t) depends on the bicycle design, the weather (wind speed), the road conditions, the terrain (hills and valleys), and the style of riding (constant or accelerating). The terrain profile is expressed as elevation profile z(x, y) defined for every location with coordinate in a horizontal plane x, y measured against a reference. The force resisting the movement of the bicycle forward and the power needed to move the bicycle with ground velocity vB(t) is approximately expressed as:

F(t)=KA(vB+vw)2+mg(s+CR)+mrffî¢žï ŒvBï Œr(1)PB(t)=FB(t)vB(t)(2)

This relationship is referred to as the physical model of the bicycle motion. Its interpretation and development is well known, and can be found in widely available literatures, for a non-limiting example, Bicycle Science, by David Gordon Wilson (3rd edition, The MIT Press, Cambridge, Mass., 2004). The time-dependency of the velocity, force, and power is explicitly shown in equation (1) and (2). The first term in equation (1) models the aerodynamic drag, where vB and vW are ground velocities of the bicycle and headwind (SI units m/s), respectively, and KA is the coefficient of aerodynamic drag (SI units in Ns2/m2). The second term models the change in potential energy due to a slope in the terrain and the road resistance/tire friction, where m is the mass of the bicycle and biker together (SI units kg), g is the gravitational acceleration (9.81 m/s2), CR is the rolling resistance coefficient (dimensionless), and s is the slope in the terrain with level z(x, y) at location x, y given with s≡sin θ, where tan θ=|∇z(x, y)| (s is dimensionless). The third term accounts for the force required to accelerate the bicycle and the third term includes both the increase in kinetic energy of the biker-bicycle body as well as the increase in energy stored in the rotation of the wheels and gears on the bicycle. Since a majority of the rotational energy is contained in the rotation of the wheels and the angular velocity of the wheels directly related to the ground velocity of the bicycle, these linear and rotational energies are jointly expressed in terms of an effective mass meff. Naturally, meff>m, and this last term makes the temporal variation in the ground velocity of the bicycle to exhibit significantly smaller variation in amplitude than the torque exerted on the pedals.

In some embodiments, the instantaneous power required to move a real bicycle is given by equation (2), where the time dependency is explicitly shown. The torque on the pedals is given by NB(t)=FB(t)vB(t)/ΩB(t).

Analysis of a Power Dissipation in a Stationary Exercise Bicycle

In some embodiments, the rider (person exercising), has control over the cadence Ω(t), gear number nG, and the torque N(t) applied to the pedals in a cardio-fitness station. The transfer ratio between the stationary-bicycle pedal angular-velocity Ω and the angular velocity of the flywheel ω, Gs=Ω/ω, is typically fixed by the size of the pulleys (or gears), but may be implemented as variable, such as an automatic gear shifting mechanism. In one embodiment, this ratio equals 12, but may be higher or lower in other embodiments. Rotating pedals delivers rotational energy to a flywheel (and any other wheels or gears in the assembly) with a moment of inertia equal to IFW (all other inertia included). The torque N(t) that needs to be provided by the stationary bicycle is thus given by,

N(t)=IFWî¢žï ŒÏ‰ï Œt+CF(ω)+NMRD(ω,I)(3)

Here CF(ω) represents the friction coming from gears, belts, and bearings involved in the mechanical assembly. The effect of friction depends on the angular velocities of the gears and shafts, and has here been normalized to the angular velocity of the flywheel. This factor contributes to the natural mechanical power loss. The last term in equation (3) is the torque NMRD (t) exerted by a device that produces controlled resistance to the pedal rotation. This device may be any device used for this purpose (for a non-limiting example, mechanical friction or alternator powering an electrical load). In one embodiment, the resistance producing device is an MRD.

In some embodiments, the torque of an MRD depends primarily on the strength of the magnetic field overlapping the flywheel and the angular velocity of the flywheel. The magnetic field is controlled by the value of direct electric current flowing through the at least one electromagnet used to produce the magnetic field. The exact relationship between the torque, the angular frequency of the flywheel, and the electric current flowing through the magnets is determined by size and shape of the flywheel, its electrical resistance, and the specific properties of the magnetic circuit that is formed by the electromagnets, the flywheel, and the air gap present between the flywheel and the electromagnet core. Energizing at least one electromagnet provides a magnetic field in the flywheel. Due to the motion of the flywheel, this magnetic field induces the eddy currents in the flywheel and these currents give rise to a magnetic field that opposes, i.e., resists the flywheel motion. This phenomenon is well known in the electrical engineering field. The torque resulting from current I at angular velocity w of the flywheel is denoted with NMRD(ω, I).

The power delivered to the stationary bicycle by the rider is given by


Ps(t)=N(t)Ω(t)=N(t)·G·ω  (4)

In some embodiments, a rider familiar with riding through a real countryside with elevation profile z(x,y) under known atmospheric and bicycle conditions may prefer to sit down on a stationary bicycle, look at the computer screen, follow a bike ride though a virtual landscape with the same z(x,y) elevation profile, and have the stationary bicycle deliver approximately the same cadence, gear number, and torque (Ω, nG, N) in a similar manner. This means that ideally on the same landscape profile z(x,y), a rider riding a real bicycle and a rider riding the cardio-fitness station with virtual reality capability along the same path y(x) should exhibit equal cadence Ω(t)=ΩB(t) and change the gear number at the same time nG(t)=nB(t). The velocity vB(t) of the real bicycle and the velocity v(t) of the rider's own virtual bicycle in the predetermined virtual landscape will be equal, v(t)=vB(t), and hence the powers dissipated by the two riders will also be equal PB(t)=P(t).

Naturally, meeting equality in the above relations exactly is practically impossible, but it is not necessary to create a perception and entertainment value to the rider riding the cardio-fitness station with virtual reality capability. What is more important is that the approximate relationship between the above relations remains consistent for repeated use and between cardio-fitness equipments. An approximation is established by assuming values of the following variables: the weight of the rider and the rider's own virtual bicycle, the landscape elevation profile corresponding to an actual geographic location, weather and road conditions corresponding to an actual time and place, and bicycle of a specific design. In one embodiment, the virtual bicycle weight is entered by the rider exercising and the virtual reality accounts for the fact that bicycle riders with different weights may experience different levels of power loss when going uphill.

In order to accomplish an approximate modeling of real biking experience, the current I controlling the resistance on the MRD in equation (3) is programmed so that it compensates for the physical phenomena modeled with equation (1). The target current that accomplishes this is given implicitly by combining equations (1) through (4). The velocity of the virtual bicycle is given by v=2πrG(nG)Ω, r is the assumed wheel size of the virtual bicycle, and G(nG) is assumed gear ratios of the virtual bicycle. Besides providing the pedaling resistance when cadence is constant, the MRD also provides the increased resistance due to inertia when the cadence increases just as it would on a real bicycle. This is described by equation (5):

NMRD(ω,I)=KA(v+vw)2v+mg(s+CR)v+meffvî¢žï Œvï ŒtGSΩ-IFWGSî¢žï ŒÎ©ï Œt-CF(GSΩ)(5)

The effective mass of a real bicycle determines how much more pedal torque is necessary to increase the speed of the real bicycle vB, while the moment of inertia IFW of the flywheel determines the amount of extra torque needed to increase the angular velocity of the flywheel on the stationary bicycle. Generally, these two effects are not equal because IFW is fixed by the stationary bicycle design, while meff depends on the rider's mass and the type of real bicycle design. The intent of the cardio-fitness station is to approximate the behavior of a real bicycle and hence the inertial behavior of the real bicycle, e.g., the response to temporal changes in cadence, dΩ(t)/dt, is approximated by the resistance provided jointly by the MRD and the inertia of the stationary bicycle IFW.

In some embodiments, the implementation of this model on a stationary bicycle involves a number of assumptions and simplifications while still maintaining the principle of the stationary bicycle mimicking the actions of real bicycle. The dominant effect difficult to predict is that of the efficiency of the magnetic circuit: The strength of the magnetic field in the flywheel is dependent on the size of the air gap between the flywheel and the electromagnet. A typical variation in the size of this gap due to manufacturing variations and thermal expansion of the flywheel and the magnets may vary from 0.1 mm to 1 mm, which produces greater than 25% change in the pedaling resistance. This level of variance is noticeable by the rider and is unacceptable for a state-of-the-art cardio-fitness station. FIG. 6 illustrates power dissipated by the flywheel under various air gap sizes. FIG. 7 shows measured flywheel torque for flywheel angular velocity 600 RPM illustrating the variation of torque with electromagnet current and gap between the electromagnet and the flywheel.

In some embodiments, the size of the gap is estimated from the known physical properties of the flywheel (size and thermal expansion coefficient) and the flywheel temperature. The temperature of the flywheel is measured using a thermostat or thermocouple located on or in the close proximity of the flywheel. Based on the temperature of the flywheel as additional information about the MRD, the relationship between the electromagnet current and the torque is characterized for a range of angular frequencies, torques exerted on the fly wheel, and temperatures of the flywheel. The obtained data is used to create a formula or a lookup table, which is then used by the computer to set the current through the electromagnets depending on the temporal variation of the cadence. The input variables to the formula are the flywheel angular frequency, gap between the electromagnet and the flywheel, and the required torque (resistance). Accounting for the temperature variation of the gap improves the MRD performance, but it does not eliminate the manufacturing variation in dimensions.

In order to improve this pedaling force uncertainty further, one may attempt to measure the gap size in real time and/or specify a tighter manufacturing tolerances. However, these approaches are potentially impractical as they increase the complexity and the price of the cardio-fitness station to an unacceptable level. Another approach described herein is to use the measurement of torque as a feedback to make a correction to the assumed relationship between the current, angular velocity, and the torque. The method by which this is implemented in a cardio-fitness station with virtual reality is described with reference to FIG. 8.

Method of Use

In some embodiments, the rider sitting on the cardio-fitness station (such as the station shown in FIG. 1) watches the images on the video monitor 150 and listens to the sounds coming from the headphones (not shown) while rotating the pedals 131, steering the handlebars 141, and occasionally changing the gear using the gear-shifting lever 142. The computer 160 runs a virtual reality program and accepts inputs from the rider exercising via the position of the handlebars 141, the momentary gear number via the history of all gear changes applied to the gear-shifting lever 142, and the rotation of the pedals 131. The listed input parameters are used to determine the motion of the rider's own virtual bicycle in a predetermined virtual landscape.

FIG. 8 is used to describe an exemplary process for control of pedaling resistance in accordance with one embodiment of the present invention. The control system is implemented in discrete time steps because computers operate by calculating quantities at discrete time instances. The time increments of this physical model are denoted with ΔtPM and are typically 1/60 of a second, but can be smaller. Here subscript PM refers to Physical Model. The method of FIG. 8 and other methods of this document are composed of modules which may be rearranged into parallel or serial configurations, and may be subdivided or combined. The method may include additional or different modules, and the modules may be reorganized to achieve the same result, too.

Starting at the bottom left part of the chart, the instantaneous value of cadence Ω(t) (shown with 802) and the gear number nG(t) (shown with 801) are captured by the computer. The computer calculates in block 803 the velocity v(t) of the virtual bicycle from these two variables and the assumed virtual bicycle wheel size r and gear ratio G(nG), as illustrated with block 804. In block 805, the location x,y of the rider's own virtual bicycle in the virtual landscape (or position along a VER y(x)) with elevation profile z(x, y) is used to determine the slope virtual s(x, y). The computer has previously stored a set of constants needed to completely define the bicycle power dissipation model: the assumed mass of the rider (constant or entered by the rider externally prior to exercise), and aerodynamic and rolling resistance factors (block 806). The virtual bicycle velocity calculated in the previous step v(t−ΔtPM) and cadence Ω(t=ΔtPM) have also been stored. In block 807, all these parameters are used to calculate the power P(t) that a real road bicycle would be dissipating using equations (1) and (2), and the power is converted to torque Ns(t) that needs to be exerted by the MRD using equation (4). Here, subscript s refers to a set value (e.g. a predetermined value), and the torque is given by NS(t)/GSP(t)/Ω. The value of the torque on the flywheel required by the physical model 808 is now applied to the MRD torque control loop 809.

An efficient use of the MRD involves a computer predicting the MRD operation from its input variables. The implicit relationship between the relevant variables at steady state is expressed as


M(N,ω,I,h)=0.  (6)

This equation will in upcoming text be referred to as MRD master equation. The input to the MRD are the electric current I (into the electromagnets) and the flywheel angular velocity ω. The current, flywheel angular velocity and the resulting torque NMRD are fast-changing MRD parameters, i.e., they can change significantly between the turns of the pedals of the cardio-fitness station. The MRD performance is also affected by slow-varying parameters such as temperature, which affects the flywheel-electromagnet gap and the electrical resistance, and also by unknown starting values of the resistance and flywheel-electromagnet gap. The flywheel-electromagnet gap and its variation have the strongest affect on the accuracy of the MRD performance prediction and for this reason all of the slow-varying parameters (including the gap) are combined in one parameter h, referred to as the gap parameter.

In some embodiments, the time steps ΔtTC, through the MRD torque-control loop may be equal to the physical model times step ΔtPM, but may be faster if smoother variable changes are necessary. The discrete time steps are numbered with integer j. At the time when the torque Ns required by the MRD is set and delivered to the torque-control loop 809, the measurement of the angular velocity ωm(j) and the gap parameter h(j) are available and the value of electromagnet current I(j) that would give the requested torque: M(Ns,ωm(j),I(j),h(j))=0 is solved in block 810. Although the MRD master equation (6) is written implicitly, the current I(j) is obtained from an explicit formula or a look-up table based on the MRD master equation (6). The calculated current I(j) is applied to the electromagnets in the MRD 811. A measurement of torque Nm(j+1) 812 and flywheel angular velocity ωm(j+1) 813 are performed using the preferred embodiment of the MRD described above. The new torque value Nm(j+1) includes the contribution from both the MRD and the inertia of the flywheel, as shown in equation (7):

Nm(j+1)=IFWω(j+1)-ω(j)Δt+NMRD(j+1)(7)

This ignores the friction of the stationary bicycle CF(ω), but it can be added in a straight forward way. Block 814 solves the torque delivered by the MRD NMRD(j+1) using equation (7) with known flywheel moment of inertia IFW. Even without acceleration of the flywheel, the set Ns value and the NMRD value are seldom equal and a correction should be performed. This correction is necessary because the previous (or the first) guess (provided by a temperature measurement 816) for the gap parameter was not correct. The values NMRD(j+1), ωm(j+1) and I(j+1) are now used with the MRD master equation in block 815 to determine a corrected value of the gap parameter h(j+1) (Namely, solving M(Ns,ωm(j+1),I(j+1),h(j+1))=0 for the new, corrected value of the gap parameter to get a corrected value for the gap parameter h(j+1)). This is again done using a formula or a look-up table. The corrected value of the gap parameter h(j+1), the current flywheel angular velocity ωm(i+1), and MRD torque Ns required by the physical model are now again used to set the MRD current in block 810, i.e., the torque-control loop now repeats.

In some embodiments, the variation in the gap parameter is slow in comparison with the time variation in the angular velocity and torque, and in this arrangement the approach converges efficiently and provides satisfactorily small errors (rider un-noticeable) within very few loops of the torque-control algorithm.

It will be understood that other embodiments of the torque-loop algorithm are possible within the context of the systems and methods of this document. The time steps ΔtPM for the physical model may be equal to slower than the torque-control loop time steps ΔtTC. A typical range for ΔtPM may be between about 1/60 and about 1/200 of a second. Moreover, many embodiments have been specifically described as including components from one or more figures in combination. However, other components may be substituted. Similarly, components may be grouped or subdivided in various ways. Thus, embodiments may be formed using some of the components and offering some of the features described, and may include components not described or offer features not described in this document. Moreover, features of one embodiment may be incorporated into other embodiments, even where those features are not described together in a single embodiment within the present document.

Patent Prosecution report image

Empower your practice with Patexia Publication Prosecution IP Module.

Get access to our exclusive rankings and unlock powerful data.

Looking for a Publication Attorney?

Get in touch with our team or create your account to start exploring a network of over 120K attorneys.