Classifications

• • B—PERFORMING OPERATIONS; TRANSPORTING
  • B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
  • B62M—RIDER PROPULSION OF WHEELED VEHICLES OR SLEDGES; POWERED PROPULSION OF SLEDGES OR SINGLE-TRACK CYCLES; TRANSMISSIONS SPECIALLY ADAPTED FOR SUCH VEHICLES
  • B62M25/00—Actuators for gearing speed-change mechanisms specially adapted for cycles
  • B62M25/08—Actuators for gearing speed-change mechanisms specially adapted for cycles with electrical or fluid transmitting systems

Definitions

• the present invention addresses the problem of the control of variable-ratio transmissions .
• a typical example of a transmission of this type which will be referred to for simplicity in the following description, is that of a bicycle transmissions.
• a method and a device for automatically controlling the transmission ratio of a bicycle so as automatically to identify the optimal ratio in dependence on the pedalling effort or force are known from EP-A-0 831 021 which is taken as the model for the preambles to Claims 1 and 10.
• the object of the present invention is to improve systems for controlling the ratio in variable-ratio transmissions, particularly with regard to the optimization of the interaction between an apparatus having such a transmission and an operator using the apparatus.
• the invention also relates to the respective method of operation.
• the application of the invention is particularly advantageous in the cycling field and, in particular, in the field of competitive cycling.
• Figure 1 shows schematically the application of a system according to the invention to a cycle such as a bicycle
• Figure 2 shows the structure of a system according to the invention schematically in the form of a block diagram
• Figures 3 and 4 show the general criteria of the operation of a system according to the invention in two successive levels of detail
• Figure 5 shows the operation of the system according to the invention in the form of a flow chart
• Figures 6 and 7 are further graphs indicative of the criteria (affinity functions) for the operation of the system.
• apparatus having a variable-ratio transmission is generally indicated 1.
• the apparatus in question is constituted by a bicycle such as, for example, a sports cycle.
• the bicycle 1 is equipped with the following devices : - an electrically-operated gearbox 2, shown here associated with the rear derailer of the bicycle 1 (in possible variants of the invention such a gearbox could alternatively or additionally be provided on the front derailer of the bicycle, if it has one) ; the gearbox concerned can thus change the position in which the bicycle chain (driven by the pedal crank 3 by means of which the cyclist applies the input driving force to the transmission) cooperates in a meshed arrangement with the sprockets associated with the rear wheel hub of the bicycle, in dependence on a control signal,
• control device 4 the heart of which is preferably constituted by a microprocessor, for generating the control signal
• a set of sensors 5 to 7 which are sensitive to respective parameters of use of the bicycle and can generate respective signals to be received and processed by the device 4 ; these are, basically, a sensor 5 which can detect the pedalling force or effort exerted by the cyclist on the pedals (that is the driving force) , a sensor 6 which is sensitive to the forward speed of the bicycle 1 (in general terms, the frequency or speed of operation of the apparatus represented herein by the bicycle 1) , as well as a sensor 7 which is sensitive to the pedalling frequency, that is, the frequency of the periodic action by which the cyclist applies the input driving force by means of the pedal crank 3.
• a sensor 5 which can detect the pedalling force or effort exerted by the cyclist on the pedals (that is the driving force)
• a sensor 6 which is sensitive to the forward speed of the bicycle 1 (in general terms, the frequency or speed of operation of the apparatus represented herein by the bicycle 1)
• a sensor 7 which is sensitive to the pedalling frequency, that is, the frequency of the periodic action by which the cyclist applies the input driving force
• the various sensors in question may in fact be combined with one another and/or with other components of the system.
• the sensor 7 which detects the pedalling frequency can advantageously be combined with the sensor 5 which detects the pedalling effort .
• gearbox 2 which can advantageously be used in the context of the invention is constituted by the gearbox sold under the reference Z S 800 by the company MAVIC.
• FIG. 2 shows, in schematic terms, at the level of the general system architecture, the arrangement of connections between the various elements described above, and the control device 4 which receives the signals generated by the sensors 5 to 7 (also defined below as the first, second and third signals) and which acts on the gearbox 2 in order to change the transmission ratio of the gearbox 2 in dependence on the processing operation described further below.
• the control device 4 which receives the signals generated by the sensors 5 to 7 (also defined below as the first, second and third signals) and which acts on the gearbox 2 in order to change the transmission ratio of the gearbox 2 in dependence on the processing operation described further below.
• a set of sensors of this type (which can provide the three signals mentioned in digital form) is available under the trade name SRM TRAINING SYSTEM from the company Ingenieurb ⁇ ro Schoberer .
• the further functional blocks indicated 8 and 10 indicate that the method according to the invention allows the user to intervene in the operation of the system, in particular with regard to two basic factors, that is: - the way in which the device 4 interprets or classifies (as will be described further below) the values of the signals received from the sensors 5 to 7 , and
• the system according to the invention allows the following factors to be taken into account, by respective selective control interventions :
• the bicycle For a given resisting load, the bicycle enables a condition of impedance match, and hence an optimal frequency to be selected by changing the transmission ratios.
• the cyclist In practice, the cyclist is comparable to a high-efficiency motor which can produce its best output in terms of power produced within a fairly narrow band of pedalling frequencies (or cadences). By altering the ratios, the cyclist can keep his pedalling action within this band of greatest efficiency.
• the frequency is too low (below 60/75 rpm) , the risk of muscle damage increases, whereas if it is too high (90/120 rpm) the cyclist starts to go into oxygen deficit .
• block 1 indicates the bicycle system as a whole.
• the operating conditions of the system are determined (with regard to the transmission ratio) by what may be defined, by the conventional terminology of automatic control theory, as an "actuator” (constituted, in the specific example, by the gearbox 2) .
• the control system shown schematically in the form of the device 4 , can therefore act on the actuator 2 (in dependence on the signals of the sensors, as shown best in the diagram of Figure 4 which will be referred to below) so as to implement a feedback operation directed towards minimizing the deviation or error "e" which may be found between a theoretical reference frequency CR and the actual frequency determined from the corresponding signal generated by the sensor 7.
• the actuator 2 In practice, when the error signal (e) is above a predetermined, possibly variable, threshold, this causes the actuator 2 to be driven in a manner such as to change the transmission ratio so as to bring the signal (e) back below the predetermined threshold.
• the transmission ratio (understood as the ratio between output speed and input speed) is reduced when the pedalling frequency tends to fall (for example, because the cyclist is pedalling uphill or against the wind) and is increased when the frequency tends to rise (for example, because the cyclist is pedalling downhill or with a favourable wind) .
• the method according to the invention may be implemented either with the use of the specific criteria described further below with reference to Figure 5 and the following Figures (basically with the use of a so-called expert system, preferably of the "fuzzy" type) or, in less preferred embodiments, by adopting systems which perform the control operation (action on the gearbox/actuator 2 so as to minimize the deviation between the reference frequency CR and the actual pedalling frequency) with the use of mechanisms of different types for processing the signals, for example, of the type described in the various documents cited in the introductory part of the present description.
• the important element of the invention is that , instead of providing for operation on the basis of a reference frequency value CR which is fixed or predetermined (possibly selectively) the system according to the invention determines the reference frequency value CR (in accordance with a substantially adaptive criterion, preferably implemented in real time or substantially in real time) by deriving the reference frequency from the very parameters (pedalling effort, forward speed, actual pedalling frequency, etc.) which characterize the interaction between the cyclist and the bicycle at the time in question. All of this takes place in accordance with intervention criteria which can be determined and controlled selectively by the user.
• a substantially adaptive criterion preferably implemented in real time or substantially in real time
• the reference frequency CR is not static and determinable a priori , even selectively. In fact it depends, on the one hand, upon the resisting load (which in turn depends on various factors) and, on the other hand, on further external factors .
• the dependence of the reference frequency CR on the resisting load may be expressed as a dependence on factors such as :
• the cyclist may decide to pedal for a race, for a sprint, or simply for a tiring translocation, in accordance with criteria which express his will, and hence a basically predictive behaviour projected into the future and not based on parameters detected and/or detectable in the past or in the present , and
• the control device 4 preferably comprises a so-called expert system operating in accordance with a fuzzy logic.
• the fuzzy logic and the respective operating mechanisms are known per se, as are the advantages which this type of logic brings to complex problems the solution of which is based more on empirical considerations resulting from experiment and simulation than on mathematical modelling of the problem.
• the flow chart of Figure 5 shows a sequence of steps, between an initial step 100 and a final step 101, which is intended to be repeated by the control device 4 in order to perform an automatic control of the actuator constituted by the gearbox 2 and hence of the transmission ratio of the bicycle .
• the reference to automatic operation does not, however, mean that a bicycle 1 equipped with the system according to the invention should necessarily provide only for automatic operation.
• the cyclist can in fact exclude the operation of the system so as to be able to act on the transmission unit or transmission units manually in conventional manner (it is pointed out once more that the method according to the invention may be applied only to one or to both of the derailers normally provided on a sports cycle) or to provide for some form of semi-automatic operation.
• these methods of operation and of complete or partial deactivation of the system are such as not to require detailed description herein.
• the sequence of steps between steps 100 and 101 can be implemented with a periodic cadence and/or at a certain frequency which may be fixed or variable according to need, in dependence on specific requirements of use.
• the frequency of repetition of the steps described below does not need to be very high since, even in transitory racing stages, changes in the bicycle system 1 develop fairly slowly over time when compared with the processing speed of conventional electronic apparatus.
• the above-mentioned control sequence may be repeated, for example, at intervals of about 1 second.
• the action on the gearbox 2 in order to change the transmission ratio preferably provides for a certain low- pass filtering effect .
• This is to prevent instantaneous variations of one or more of the parameters used by the expert system 41 being translated into an undesired immediate change of the transmission ratio: for example, there might be a sudden change in the pedalling effort due to the fact that the cyclist has risen from the saddle upon starting to pedal, so to speak "standing" on the pedals; above all, the above-mentioned change may have a different sign in dependence on the instantaneous angular position of the pedal crank at the moment at which the cyclist starts to pedal standing up. Moreover, it seems advantageous, in any case, to prevent changes in transmission ratio, possibly with opposite signs, taking place in rapid succession.
• the expert system 41 may take account of the signal corresponding to the actual pedalling frequency in at least two different ways, that is:
• the expert system 41 operating in accordance with a fuzzy logic, converts the values of the input variables (for example, the signals read from the sensors 5, 6 and - possibly - 7) into a linguistic description in order then to work out a control strategy contained in a set of logic rules . The result is then converted back into a precise and unambiguous output datum.
• the input variables for example, the signals read from the sensors 5, 6 and - possibly - 7.
• the steps indicated 102 and 103 indicate the initial steps in which the system reads the signals coming from the sensor 5 (pedalling effort) and the sensor 6 (speed) , respectively. These are preferably signals already converted into digital form beforehand at the outputs of the respective sensors, as is the case in the SRM sensors already mentioned above. If this is not the case, the conversion is performed, in known manner, in the device 4.
• the expert system converts each of the two variables read as inputs into fuzzy values, that is, linguistic values such as "high”, medium”, “low”, medium-high” , etc.
• fuzzy values that is, linguistic values such as "high”, medium”, “low”, medium-high” , etc.
• it makes use of predefined functions which reflect the degree of affinity of the fuzzy variables to the various fuzzy values (affinity functions) .
• the affinity function relating to torque preferably has a curve of the type shown in Figure 7 in which the abscissa scale corresponds to the value of the pedalling torque (pedalling effort or force - sensor 5) expressed in N-m.
• steps 104, 105 have been shown as theoretically performed by a parallel processing method for each input parameter since this representation is, above all, more readily understood. It will be clear to experts that the same result can be achieved by serial processes. It will also be noted that the flow chart of Figure 5 makes clear that it is possible to intervene in steps 104, 105 by means of commands applied by the user by means of interface modules 8, 9 and 10. These interfaces may in fact be incorporated in a user interface 11 such as, for example, a keypad or an analogue control module disposed, for example, on the handlebars ( Figure 1) in a position readily accessible to the cyclist.
• a user interface 11 such as, for example, a keypad or an analogue control module disposed, for example, on the handlebars ( Figure 1) in a position readily accessible to the cyclist.
• the logic of the attribution of linguistic values to the data corresponding to the signals coming from the sensors 5 and 6 may in fact vary in dependence on various parameters set selectively on the interface modules 8, 9, 10.
• the cyclist's level of athletic preparation may mean that a speed or a pedalling effort which is to be considered high for an amateur or recreational cyclist may be considered differently (for example, medium- high, medium or even low) for a professional cyclist.
• a speed and/or pedalling effort value considered high during a translocation stage, even within a cycling race, may be considered low or even very low with reference to a sprint or a timed race.
• the intervention may, for example, be that of intervening in the operation to define the above-described affinity functions by removing or adding affinity functions; an amateur or recreational cyclist will usually be less interested in a very sophisticated and differentiated definition of the above-mentioned functions than a professional for whom the need continuously to achieve a close adaptation of the bicycle system to his physical performance may be very pressing and decisive.
• the box indicated 106 represents schematically the set of functions of the expert system dedicated to the definition of the reference frequency as a result of the attribution of the affinity functions performed in steps 104 and 105.
• the criterion for the application of the above-mentioned rules may be regarded as a type of scanning of a matrix table, for example, a two- dimensional table of which the lines are identified by the affinity functions relating to the speed and the columns are identified by the affinity functions relating to the pedalling effort .
• the corresponding affinity functions identify the third dimension of the matrix structure (which in any case is implemented with fuzzy logic and hence with respective boxes identified by probability functions) .
• the output fuzzy values could be:
• the system may also conclude that, in certain conditions, the value "medium” should be attributed to the reference frequency CR at 25% and "low” at 90%.
• each fuzzy value is independent of the others so that it is wholly legitimate for the system to reach the conclusions set out above, that is, with a sum of the fuzzy values other than 100%.
• the set of functions indicated schematically by the block 106 may be rendered variable in dependence on the parameters set by the user by acting on the interface 11: this applies in particular with regard to the possibility of modifying the rules which determine the attribution of the reference frequency value in dependence on the affinity functions corresponding to the input parameters so as to be able to implement different sets of rules .
• the expert system 41 it is also possible to provide - in accordance with known criteria - for the expert system 41 to be able to store the above-mentioned set of rules or possibly several sets of rules to be used in different conditions, in dependence on learning cycles, that is, to provide for a stage for the training of the system in which the cyclist acts on the transmission, changing the ratios by means of a positive action by the cyclist (an expression of his will) in dependence on various riding conditions whilst the system learns the respective rules, subsequently applying them automatically when this operating criterion is subsequently selected.
• control mechanism and, in particular, the learning mechanism may be implemented with the use of the configurations currently known as neural or neurone networks which, as is well known, can be applied well to the implementation (and to the learning) of operating data and conditions which are purely phenomenological and cannot be expressed directly in the form of a mathematical model, particularly of an algorithmic type.
• the expert system 41 implements a reverse conversion mechanism known as "defuzzyfication" in which the fuzzy values of the output variable are combined to produce a precise value of the reference frequency parameter CR.
• step 108 the reference frequency value thus obtained is compared (as also shown more specifically at the node 42 which is actually included in the expert system 41) with the pedalling frequency value derived by the sensor 7. If the respective modulus of the deviation (error signal "e") is below a given threshold, the system does not intervene, going on to the final step 101, thus deciding in practice not to act on the gearbox 2 and that the question of a possible change of the transmission ratio is to be reconsidered upon the next checking sequence .
• the system goes on to a further step 109 and then to yet another subsequent step 110, in which the action on the gearbox 2 is actually implemented so as to bring about the change in ratio (an increase or a decrease) in dependence on the adaptation requirements found, that is, in dependence on the sign of the deviation value "e" .
• the step indicated 109 corresponds to a time check (in practice a filtering mechanism) in which the device 4 ascertains that an adequate period of time has elapsed since the preceding change in ratio. If a period of time less than a predetermined time threshold has elapsed (negative result of the comparison step 109) the system goes directly to the final step 101 without acting on the gearbox 2, postponing any ratio changing operation to a subsequent control sequence .
• step 110 changing the transmission ratio.
• step 109 it has been explained that by acting on the interface 11 and, in particular, on the module 10 (relating to the changing of the operating rules of the system) the cyclist can selectively vary the value of the time threshold used to perform the filtering function.

Landscapes

• Engineering & Computer Science (AREA)
• Chemical & Material Sciences (AREA)
• Combustion & Propulsion (AREA)
• Transportation (AREA)
• Mechanical Engineering (AREA)
• Control Of Transmission Device (AREA)
• Transmitters (AREA)
• Radio Transmission System (AREA)
• Feedback Control In General (AREA)
• Communication Control (AREA)
• Traffic Control Systems (AREA)
• Channel Selection Circuits, Automatic Tuning Circuits (AREA)
• Gear-Shifting Mechanisms (AREA)

Priority Applications (11)

Applications Claiming Priority (2)

Publications (1)

Family

ID=11418033

Family Applications (1)

Country Status (15)

Cited By (4)

Families Citing this family (27)

Citations (7)

Family Cites Families (10)

• 1999
  • 1999-08-24 IT IT1999TO000722A patent/IT1310144B1/it active
• 2000
  • 2000-08-21 HK HK02108941.1A patent/HK1047267B/en not_active IP Right Cessation
  • 2000-08-21 AU AU69977/00A patent/AU6997700A/en not_active Abandoned
  • 2000-08-21 CN CNB008143943A patent/CN1165450C/zh not_active Expired - Fee Related
  • 2000-08-21 JP JP2001518312A patent/JP2003507261A/ja active Pending
  • 2000-08-21 US US10/069,046 patent/US6714849B1/en not_active Expired - Fee Related
  • 2000-08-21 EP EP00958477A patent/EP1212236B1/en not_active Expired - Lifetime
  • 2000-08-21 PT PT00958477T patent/PT1212236E/pt unknown
  • 2000-08-21 DK DK00958477T patent/DK1212236T3/da active
  • 2000-08-21 CA CA002382424A patent/CA2382424C/en not_active Expired - Fee Related
  • 2000-08-21 ES ES00958477T patent/ES2213039T3/es not_active Expired - Lifetime
  • 2000-08-21 WO PCT/EP2000/008125 patent/WO2001014203A1/en not_active Ceased
  • 2000-08-21 EA EA200200276A patent/EA004382B1/ru not_active IP Right Cessation
  • 2000-08-21 DE DE60008123T patent/DE60008123T2/de not_active Expired - Lifetime
  • 2000-08-21 AT AT00958477T patent/ATE258868T1/de active

Patent Citations (11)

Also Published As

Similar Documents

Legal Events