Unclamping tools and methods

Abstract

A method for operating a pre-opening and pre-positioning clamp movable between an open position and a closed position, the method including using a tool head (201) to release the clamp from the open position, taking measurements of movement of the tool head resulting from the release, recording data associated with the measurements, identifying frequency components of the data, and determining whether the clamp has moved to the closed position based on the identified frequency components.

Description

[Technical field]

[0001] The present invention relates to a method of operating a pre-opening clamp. More particularly, but not by way of limitation, the present invention relates to a method of operating a pre-opening, pre-positioning (POPP) clamp, which may be in the form of, for example, a clipless pre-opening spring band clamp. [Background technology]

[0002] Flexible pipes are used in many fields to transport fluids. Examples of products using such pipes include washing machines and automobiles. For example, in an automobile engine, the cooling medium is transported to the radiator through a rubber hose. The hose is attached to spigots such as the fluid inlet and outlet and fixed in place by a suitable clamp such as a Pre-Opened Pre-Positioned Clamp (often simply called a "POPP Clamp").

[0003] FIG. 1 illustrates an exemplary POPP clamp 100 of the prior art. The clamp 100 includes a substantially circular curved band 101. A first end of the band 101 includes a notch 103 shaped to accommodate a second end of the band 101. The band 101 can be pried open to have a larger diameter while maintaining its substantially circular shape. Opening the band 101 places the band 101 in tension such that upon release, the band 101 will contract back to its original diameter. The second end of the band 101 includes a protrusion 105 that functions to lock the band 101 in an open position when the band is fully opened. FIG. 1 illustrates the clamp 100 locked in such an open position. The band is in tension but is prevented from closing by the protrusion 105 pressing against a stop 106 at the first end of the band. To release the clamp 100 from the open position, an unclamping tool is inserted into the recess 107 in the band 101 and manipulated such that the unclamping tool is pressed down onto the prongs 109 of the clamp 100. This action forces the prongs 105 downward and away from the stops 106 until they can pass through the notches 103, allowing the band 101 to contract to a smaller diameter. Thus, the POPP clamp 100 is operable between an open position and a closed position.

[0004] The POPP clamp is installed by placing it in an open position, moving it into position surrounding a hose attached to a spigot, and releasing it from the open position. The POPP clamp then contracts until it contacts the hose, clamping it to the spigot. The POPP clamp can then be said to be in a closed position. Release of the POPP clamp, which results in the clamp moving to the closed position, can be said to be successful release of the clamp.

[0005] The POPP clamp does not always successfully release from the open position. For example, the projection 105 may move from its rest position but not move through the notch 103. In such a situation, the clamp 100 will not move to the closed position and therefore will not apply a clamping force to the hose.

[0006] In automotive production lines, POPP clamps are installed manually by an operator using a hand-held clamp release tool. It is not always immediately apparent to the operator whether the POPP clamp has been successfully released or not. Thus, the operator may not realize that the POPP clamp has not been successfully released and therefore is not applying clamping force to the hose. This can waste the operator's time by requiring an inspection of the clamp, and an unnoticed and not successfully released clamp may ultimately result in a leak at the pipe-spigot junction that is not properly clamped. In the case of an automotive engine coolant system, a leak can result in loss of coolant, which can lead to engine damage or failure.

[0007] Production lines, especially automotive production lines, also require traceability of the operations performed on the production line for quality control purposes. To meet this requirement, operators may need to record each clamp they release. This is usually a time-consuming manual process, further wasting the operator's time. Summary of the Invention [Problem to be solved by the invention]

[0008] The present invention seeks to alleviate one or more of the problems set forth above. Alternatively or additionally, the present invention seeks to provide an improved tool and / or an improved method for operating a pre-opened, pre-placed clamp. [Means for solving the problem]

[0009] According to a first aspect, the present invention provides a method of operating a pre-opened, pre-positioned clamp operable between an open position and a closed position. The method includes releasing the clamp from the open position using a tool head. Data from measurements of movement of the tool head resulting from the release is recorded. The method includes identifying a frequency component of the data and determining whether the clamp has moved to a closed position based on the identified frequency component.

[0010] Embodiments of the present invention using the above method can provide more reliable detection of whether a clamp has been successfully released. The inventors believe that, for example, embodiments of the present invention provide more reliable detection of whether a clamp has been successfully released compared to comparable methods in which only data directly related to absolute tool position and velocity is used to determine successful clamp release. Embodiments of the present invention can also provide a "smart" declamping tool that allows for automatic recording of declamping, for example to meet traceability requirements on automotive production lines.

[0011] Taking the motion measurements may include measuring relative motion in at least three different directions in three-dimensional space. Taking the motion measurements may include measuring relative rotation about at least three different axes in three-dimensional space. Taking the motion measurements may include measuring motion with respect to all three planes of motion. Taking the motion measurements may include measuring rotation with respect to pitch, roll, and yaw. Taking the motion measurements may include measuring forces, etc., rather than absolute position changes in three-dimensional space. The motion may include vibrations, for example at frequencies greater than 100 Hz.

[0012] When successfully released, the POPP clamp vibrates at a characteristic frequency, and therefore, by measuring the vibration of the tool head, embodiments of the present invention can determine whether the POPP clamp has been successfully released.

[0013] The frequency components may include a component related to one or more frequencies characteristic of the clamp being moved to a closed position.

[0014] An embodiment of the present invention that detects one or more frequency-related components characteristic of a clamp being moved to a closed position can determine whether the detected movement corresponds to successful POPP clamp unclamping.

[0015] The method may further include determining that a trigger event has occurred based on the measured motion. Determining the occurrence of such a trigger event may include detecting one or more significant vibrations. Significant vibrations may be detected within, for example, 100 ms. -2 It may be considered to include accelerations above a larger threshold. The threshold is 500ms -2 It can be larger than 900ms. -2 The threshold is about 1000ms. -2 or greater (i.e., at least ∼100 g). Significant vibration may alternatively or additionally be, for example, 500 ms -3 It can be considered to include a rate of change (absolute magnitude value) of acceleration that exceeds a larger threshold. The threshold is 900ms -3 Determining the occurrence of a trigger event may include detection of at least three (e.g., four, or optionally more) consecutive significant vibrations. Recording (of data derived from measurements of tool head movement) may be performed in response to the trigger event.

[0016] An embodiment of the present invention that operates based on a detected trigger event allows the declamping tool to reduce its power consumption, thereby extending the battery life of the tool by entering a low power mode until a trigger event is detected.

[0017] The recorded data may include data defining the movement preceding the triggering event and data defining the movement following the triggering event.

[0018] Embodiments of the present invention that evaluate data both immediately before and after a trigger event can allow for a more accurate determination of whether the measured movement corresponds to successful release of the POPP clamp.

[0019] As described above, the method includes identifying frequency components of the data and determining whether the clamp has moved to the closed position based on the identified frequency components. Identifying the frequency components may include performing a Fourier transform. Determining whether the clamp has moved to the closed position may include identifying a frequency of a maximum peak in the frequency components and / or a frequency band having a maximum amplitude (which may be referred to as a predominant frequency and / or predominant frequency band). Determining whether the clamp has moved to the closed position may include identifying an amplitude of the frequency component at the predominant frequency / frequency band. Determining whether the clamp has moved to the closed position may additionally or alternatively include identifying a frequency and / or amplitude of one or more peaks in the frequency components other than the maximum peak. Determining whether the clamp has moved to the closed position may include evaluating a rate of change of the frequency components (in the frequency domain). Determining whether the clamp has moved to the closed position may include evaluating a ratio of an amplitude of one frequency component (or frequency band) to another frequency component (or frequency band). Determining whether the clamp has moved to the closed position may include evaluating such ratios for a plurality of, e.g., 10 or more, pairs of different frequency bands. At least one frequency component (or frequency band) used to calculate the ratio may be the amplitude of the dominant frequency. At least one frequency component (or frequency band) used to calculate the ratio may be the second highest dominant frequency and / or the amplitude of the second highest dominant frequency. At least one frequency component (or frequency band) used to calculate the ratio may be the average amplitude (either the mode, median or mean) in the frequency domain. Kurtosis and / or skewness of the frequency components may be used in determining whether the clamp has moved to the closed position.

[0020] Determining whether the clamp has moved to a closed position may involve the use of other parameters in addition to those related to the frequency content.

[0021] Determining whether the clamp has moved to a closed position may include determining a change in orientation of the unclamping tool, which may be relative to a starting position and / or a position at a given moment, such as the position at the time of the trigger event.

[0022] Determining whether the clamp has moved to the closed position may include evaluating at least one of the correlations between (a) a first set of data representative of or derived from a rotation rate about a first axis over a given time period and a second set of data representative of or derived from a rotation rate about a second axis, e.g., different from the first axis, over the same time period, and (b) a first set of data representative of or derived from a translation acceleration in a first direction over a given time period and a second set of data representative of or derived from a translation acceleration in a second direction, e.g., different from the first direction, over the same time period. Determining whether the clamp has moved to the closed position may include evaluating a Pearson's correlation coefficient of the recorded data. Determining whether the clamp has moved to the closed position may include evaluating whether the data representative of or derived from a measurement of pitch rotation meets certain criteria. Determining whether the clamps have moved to the closed position may include evaluating a mathematical measure of diversity of the recorded data (e.g., the Shannon-Wiener diversity index of the data), and preferably a mathematical measure of diversity of data representative of or derived from a measurement of pitch rotation. Determining whether the clamps have moved to the closed position may include evaluating the kurtosis of the recorded data (or data derived from the recorded data). Determining whether the clamps have moved to the closed position may include evaluating the skewness of the recorded data. In each such case, the recorded data may include at least one of: (a) data representative of or derived from rotational velocity about at least one, preferably three different axes, and (b) data representative of or derived from translational acceleration in one direction, preferably three different directions.

[0023] Determining whether the clamp has moved to a closed position may include one or more of: calculating the number of significant vibrations; calculating the maximum or average slope of the slope of a curve representing the one or more significant vibrations in the time-position domain; calculating the rate of change of acceleration (in one or more directions); using the maximum acceleration (in one or more directions); using the minimum acceleration (in one or more directions); using the average (i.e., mean, mode or median) of the acceleration (in one or more directions); and using the first and third quartiles of the acceleration (in one or more directions).

[0024] Determining whether the clamp has moved to the closed position may include calculating a plurality of parameters from the recorded data. Determining whether the clamp has moved to the closed position may include using the parameters to assess whether the clamp has moved to the closed position. The plurality of parameters may include three or more different parameters. Each of the three or more different parameters may be associated with one of the methods described herein for determining whether the clamp has moved to the closed position. The plurality of parameters may include two or more (preferably three or more) different parameters from the group consisting of a measure of correlation, a Pearson correlation coefficient, a measure of diversity, a Shannon-Wiener diversity index, a total number of significant oscillations, a mean rate of change of significant oscillations, a rate of change of the data, a kurtosis of the data, a skewness of the data, a maximum value, a minimum value of the data, a mean value of the data, a median value of the data, a mode value of the data, a first quartile of the data, and a third quartile of the data. As can be seen, in each such case, the recorded data may include at least one of: (a) data representative of or derived from rotational velocity about at least one, and preferably three different axes; and (b) data representative of or derived from translational acceleration in one direction, and preferably three different directions.

[0025] Embodiments of the present invention that evaluate three or more of the various parameters discussed above may provide a more reliable detection of successful unclamping.

[0026] Determining whether the clamp has moved to the closed position may include applying one or more rules. The one or more rules may define characteristics of a movement that is deemed to correspond to the clamp moving to the closed position. The one or more rules may be based on an output of a classification tree. The one or more rules may be based on an output of multiple classification trees. The one or more rules may be based on a majority vote output of multiple classification trees. The classification tree may operate based on multiple weights determined by a machine learning algorithm.

[0027] The problem of reliably determining whether a detected movement corresponds to a successful release of a clamp can be intractable, with many uncontrollable variables. An embodiment of the present invention in which a machine learning algorithm is used to determine the weights can provide a more reliable method for determining whether a detected movement corresponds to a successful release of a clamp. Such an embodiment can also allow the unclamping tool to continue to evolve through ongoing training of the machine learning algorithm using data obtained by the deployed unclamping tool. An embodiment of the present invention in which one or more rules are based on the output of multiple classification trees can reduce the likelihood that the machine learning algorithm will generate an overfitted solution.

[0028] Determining whether the clamp has moved to the closed position can include evaluating an output of a sliding window algorithm. The sliding window algorithm can include calculating a ratio of an amplitude of the first frequency band to an amplitude of the second frequency band. The sliding window algorithm can include calculating a ratio of at least 10. The one or more rules can operate based on the output of the sliding window algorithm.

[0029] The step of determining whether the clamp has moved to the closed position may be considered equivalent to determining whether the clamp has been successfully released. Determining whether the clamp has moved to the closed position may include active detection of motion corresponding to the clamp moving to the closed position. In embodiments of the invention, this means actively requesting detection of a characteristic corresponding to successful operation of the POPP clamp, rather than merely detecting a characteristic corresponding to incomplete operation and assuming that operation of the POPP clamp was successful in the absence of such detection. The method may, for example, include the step of determining that the clamp has moved to the closed position, which may also include active determination that the clamp has been successfully released. Embodiments of the invention that perform such active detection of motion corresponding to the clamp moving to the closed position may provide a reduced false positive rate.

[0030] Taking the measurements may include measuring movement in each direction and about each axis at a sample rate of greater than 3 kHz, the sample rate may be greater than 4.5 kHz, the sample rate may be greater than 6 kHz.

[0031] Embodiments of the present invention having sample rates greater than 3 kHz may enable the unclamping tool to capture the full range of frequencies characteristic of successful unclamping. Embodiments of the present invention having higher sample rates may capture a larger range of frequencies, allowing a more reliable determination of whether the measured motion corresponds to a successful unclamping.

[0032] According to a second aspect of the invention, there is also provided a clamp installation tool for manipulating a pre-opened, pre-placed clamp operable between an open position and a closed position, the tool including a tool head for releasing the clamp from the open position, inertial measurement electronics configured to perform measurements of motion of the tool head, and signal processing electronics configured to identify frequency components of the detected motion and determine whether the clamp has moved to the closed position based on the identified frequency components.

[0033] The clamp may be suitable for clamping a hose, preferably an automotive hose, to a spigot. The spigot may include a fluid inlet or outlet.

[0034] The clamp in the open position may be held at an expanded diameter greater than the outer diameter of the hose. The clamp in the closed position may be disposed about the hose and the diameter of the clamp may be substantially the same as the outer diameter of the hose. The clamp may be resiliently deformable such that the clamp in the closed position applies a compressive force to the hose, clamping the hose to the spigot.

[0035] The clamp installation tool may be suitable for clamps having an inner diameter of 14mm to 43mm when in the closed position.

[0036] The clamp installation tool may be portable. The clamp installation tool may be hand-held.

[0037] The tool head may be configured to be removable from the clamp installation tool. The tool head is subject to wear and damage during use and must be replaced after a period of time. A clamp installation tool with a removable tool head allows for a worn tool head to be replaced without requiring replacement of the entire clamp installation tool. Thus, the tool head may be configured to be replaced separately with respect to the remainder of the clamp installation tool. The inertial measurement electronics (and optionally also the signal processing electronics) may be located within the body of the clamp installation tool such that the tool head may be removed and replaced without requiring replacement of the inertial measurement electronics and the signal processing electronics. Thus, the inertial measurement electronics and the signal processing electronics may be located remotely from the tool head. The inertial measurement electronics and the signal processing electronics may not be located within or on the tool head. Furthermore, the inertial measurement electronics and the signal processing electronics may not be permanently attached to the tool head. The body of the clamp installation tool and the means for attaching the tool head to the body of the clamp installation tool may be configured to transfer motion of the tool head (e.g. vibrations of the tool head) to the inertial measurement electronics. Thus, the body of the clamp installation tool and the means for attaching the tool head to the body of the clamp installation tool may be configured such that movement of the tool head causes corresponding movement in the inertial measurement electronics.

[0038] According to a third aspect of the invention, there is also provided a method of manufacturing a clamp installation tool. The tool may be according to the second aspect of the invention and / or suitable for carrying out the method of the first aspect of the invention. For example, the tool may be suitable for manipulating a pre-opened and pre-placed clamp operable between an open position and a closed position, the tool comprising a tool head configured to engage the clamp and inertial measurement electronics configured to make measurements of movement of the tool head. The method comprises programming signal processing electronics configured to, in use, receive signals corresponding to measurements made by the tool, record data relating to the measurements, identify frequency components of the data, and determine whether the clamp has moved to the closed position based on the identified frequency components.

[0039] The method may include determining one or more rules for classifying the detected movement as corresponding to the clamp moving to the closed position or not. The one or more rules may define characteristics of the movement that are deemed to correspond to the clamp moving to the closed position. Determining whether the clamp has moved to the closed position may include applying the one or more rules.

[0040] The method may include operating a machine learning algorithm to determine the one or more rules.

[0041] Of course, it will be appreciated that features described in relation to one aspect of the invention may be incorporated in other aspects of the invention, for example a method of the invention may incorporate any feature described with reference to an apparatus of the invention, and vice versa. [Brief description of the drawings]

[0042] [Figure 1] FIG. 1 is a perspective view of a typical pre-released, pre-placed clamp of the prior art. [Diagram 2] FIG. 1 is a perspective view of an unclamping tool according to a first embodiment of the present invention; [Diagram 3]FIG. 3 is a block diagram of the electronics package of the declamping tool of FIG. [Figure 4] 11 is a flow chart illustrating steps of a method of operating a pre-opened, pre-placed clamp in accordance with a second embodiment of the present invention. [Diagram 5] 10 is a flow chart illustrating steps of a method for manufacturing a declamping tool according to a third embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0043] Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying schematic drawings, in which:

[0044] FIG. 2 illustrates a perspective view of a pre-open, pre-place (POPP) declamping tool 200 according to a first embodiment of the present invention.

[0045] The declamping tool 200 includes a tool head 201 and a handle 203. The tool head 201 includes three prongs, each of which is configured to be received in a recess 107 of the POPP clamp 100 such that the tool head 201 is suitable for releasing the POPP clamp 100 from an open position. Alternative embodiments of the present invention may include a different number of prongs, such as one or two prongs, or any number of possible prongs. Incorporating three prongs allows the declamping tool 200 to engage the clamp 100 from three different angles, providing an operator of the declamping tool 200 with many possible angles for using the tool. This is particularly beneficial when the clamp is used in tight spaces, as is often the case in automotive applications. The declamping tool 200 further includes an electronics package 300 housed within the handle 203.

[0046] 3 shows a block diagram of electronics package 300. Electronics package 300 includes an inertial measurement unit 301 that is attached to tool head 201 such that movement of tool head 201 results in corresponding movement of electronics package 300 and thus inertial measurement unit 301.

[0047] The inertial measurement unit 301 includes three accelerometers and three gyroscopes configured to be able to detect motion, particularly vibration, in all three planes of motion, as well as pitch, roll, and yaw. Thus, the inertial measurement unit 301 may be said to include a six-axis gyroscope. Alternative embodiments of the invention may include fewer (i.e., one or two) accelerometers and / or fewer (i.e., one or two) gyroscopes. The inertial measurement unit 301 is configured to transmit signals corresponding to the detected motion to the signal processing unit 303 of the electronics package 300. The signals from the inertial measurement unit 301 may be said to include the inertial data of the inertial measurement unit 301. In this exemplary embodiment, the inertial measurement unit 301 detects motion and generates a full set of inertial data including data from each of the accelerometers and gyroscopes at a rate of 6.667 kHz. The inertial data therefore includes data defining the acceleration of each accelerometer in a respective sensing direction and data defining the angular velocity of each gyroscope about its respective sensing axis (rate of rotation about each such axis, i.e., angular velocity).

[0048] The signal processing unit 303 is configured to process the inertial data to determine whether the detected motion corresponds to a successful release of the POPP clamp. To that end, the signal processing unit 303 is configured to monitor the inertial data for a trigger event. A trigger event consists of motion including at least four significant vibrations, where significant vibrations are greater than or equal to 100 gs. -1 (i.e. 980ms -3) is a motion that corresponds to a rate of change of acceleration exceeding 0.01 V. The monitoring therefore involves checking the inertial data for four consecutive oscillations that meet the following criteria:

number

[0049] The signal processing unit 303 is further configured to record inertial data corresponding to a time period of 0.45 seconds in response to detecting a trigger event. Specifically, the signal processing unit 303 is configured to record inertial data for 0.27 seconds immediately prior to the trigger event (pre-trigger inertial data) and for 0.18 seconds immediately following the trigger event (post-trigger inertial data). In this exemplary embodiment, the inertial data is acquired at a rate of 6.667 kHz, so that the preceding 0.27 seconds corresponds to 1800 logged data points and the following 0.18 seconds corresponds to 1200 logged data points. Thus, the signal processing unit records six arrays, each consisting of 3000 data points, one array for each of the three accelerometers and three gyroscopes.

[0050] The signal processing unit 303 is further configured to calculate the change in orientation of the declamping tool 200 based on the recorded inertial data. This is calculated using the following formula:

number

[0051] Since the results of the GYX and GYY integration are prone to drift over time, a data fusion algorithm is used to provide estimates of roll and pitch. Thus, the final estimates of roll and pitch are calculated using the following formulas:

number

[0052] Changes in orientation of the declamping tool 200 include factors that may be evaluated in determining whether a detected movement corresponds to a successful release of a POPP clamp.

[0053] The signal processing unit 303 is further configured to determine frequency components of the recorded inertial data. Determining the frequency components includes performing a Fourier transform, such as a Discrete Fourier Transform (DFT) or a Fast Fourier Transform (FFT), on the recorded inertial data. The Fourier transform is performed on 128 data points following a trigger event for each of the accelerometer and gyroscope.

[0054] When a POPP clamp is released, a vibration is induced in the unclamping tool. A successfully released POPP clamp will induce a vibration with a particular characteristic frequency distribution that includes a single dominant frequency and a number of harmonic frequencies. This particular frequency distribution is therefore unique to a clamp that has been moved to a closed position and is indicative of a successfully released clamp. The frequency distribution may include a frequency component associated with the movement of the prongs 105 of the clamp 100. The frequency distribution may include a frequency component associated with the vibration of the prongs 109 of the clamp 100. POPP clamps come in a variety of sizes, each of which vibrates with a different characteristic frequency distribution. Thus, there are many different characteristic frequency distributions that may be indicative of a successful release of a POPP clamp.

[0055] The frequency components are processed using a sliding window algorithm. The data is first grouped into a finite number of bins, for example 64. In this exemplary embodiment, each bin corresponds to a frequency range of 52 Hz. Two windows, each 4 bins wide in this exemplary embodiment, are then applied to the data, with the first window including the first four bins and the second window including the second four bins. The average of the first window is divided by the average of the second window and the result is stored. The second window is then advanced by one bin to include bins 5-9, and the process is repeated. When the second window reaches the last four bins, the first window is advanced by one bin to include bins 2-5, and the second window is moved to include the four bins following the first window. This process is repeated until both windows reach the end of the data set. The output of this algorithm (DFT sliding window algorithm data) contains factors that can be evaluated in determining whether the detected movement corresponds to successful release of the POPP clamp.

[0056] The particular frequency bin corresponding to the maximum peak of the frequency component also comprises a factor that may be evaluated in determining whether the detected movement corresponds to successful release of the POPP clamp.

[0057] The signal processing unit 303 is further configured to determine one or more quantities related to the array of recorded inertial data.

[0058] A first exemplary quantity includes the Pearson correlation coefficient (PCC). The signal processing unit 303 is configured to calculate the PCC of pairs of arrays of pre-trigger inertial data. The PCC provides a measure of correlation between two sets of data, with a PCC of +1 indicating perfect positive linear correlation, 0 indicating no correlation, and -1 indicating perfect negative correlation. The PCC of arrays X and Y is calculated using the following formula:

number

[0059] A PCC is calculated for each possible pair of accelerometers (i.e., x and y, y and z, and x and z) and for each possible pair of gyroscopes (i.e., roll and pitch, pitch and yaw, and roll and yaw).

[0060] A second exemplary quantity includes the number of significant vibrations in the inertial data. In an embodiment of the invention, the quantity may include the number of significant vibrations in, for example only, the inertial data associated with the roll of the declamping tool 200. The signal processing unit 303 is configured to determine the number of significant vibrations by searching the array of inertial data for elements whose difference from a preceding element exceeds a predefined threshold. The signal processing unit 303 is further configured to determine the average magnitude of those differences that exceed the threshold.

[0061] A third exemplary quantity includes the skewness of the array. The signal processing unit 303 is configured to calculate the skewness of each array of inertial data using the following formula:

number

[0062] A fourth exemplary quantity includes the kurtosis of the array. The signal processing unit 303 is configured to calculate the kurtosis of each array of inertial data using the following formula:

number

[0063] A fifth exemplary quantity includes the rate of change of each array of inertial data. The signal processing unit 303 is configured to calculate the first and second derivatives of each array of inertial data using the following equations:

number

[0064] A sixth exemplary quantity includes a Shannon-Wiener diversity index. The signal processing unit 303 is configured to calculate the Shannon-Wiener diversity index for each array of inertial data using the following formula:

number

[0065] In an embodiment of the present invention, a Shannon-Wiener diversity index may be calculated based on, for example, solely on, inertial data related to the pitch of the declamping tool 200.

[0066] Further exemplary quantities include the first, second and third quartiles, maximum, minimum, mean and mode of a sorted array of inertial data. The array of inertial data for a sensor can be considered as two subarrays, with the pre-trigger inertial data comprising the first subarray and the post-trigger inertial data comprising the second subarray. Thus, the recorded inertial data comprises 12 subarrays. The signal processing unit 303 is configured to sort the inertial data within each subarray by magnitude and then determine the first, second and third quartiles (where the second quartile is the median), maximum, minimum, mean and mode of each of the sorted subarrays. In this exemplary embodiment, the inertial measurement unit 301 outputs 12 bits of data, so it is unlikely to have repeated values ​​within a subarray. Thus, when calculating the mode, the inertial data is grouped into 100, preferably equally sized bins, and the modal bins are determined.

[0067] Each of the above quantities includes factors that may be evaluated in determining whether a detected movement corresponds to a successful release of the POPP clamp.

[0068] To determine whether the detected movement corresponds to a successful release of the POPP clamp, the signal processing unit 303 is configured to evaluate one or more rules. The one or more rules define the characteristics of the movement that are considered to correspond to the clamp moving to the closed position. The one or more rules operate based on a number of parameters that may include one or more of the change in orientation of the declamping tool, the frequency components of the recorded inertial data, and one or more of the calculated quantities described above. Thus, in an embodiment of the invention, the signal processing unit 303 is configured to determine whether the clamp has moved to the closed position based on the identified frequency components. In an alternative embodiment, the signal processing unit 303 is configured to determine a percentage probability that the detected movement corresponds to the clamp moving to the closed position based on a count of the number of the one or more rules that are fulfilled.

[0069] The one or more rules are determined using a machine learning algorithm that includes a classification tree configured to classify a detected movement as either a successful release of the POPP clamp (a "clip") or not a successful release of the POPP clamp (a "hit") based on a number of parameters.

[0070] Examples of motions that should be classified as "hits" include a clamp that is not successfully released, an operator dropping the unclamping tool, and an operator hitting the unclamping tool against another object.

[0071] The classification tree is configured to operate based on a number of weights that determine the significance of each of a number of parameters in classifying the detected motion.

[0072] The machine learning algorithm is configured to train a classification tree based on a corpus of training data. The corpus of training data includes a plurality of entries, each of which corresponds to a past move that corresponds to either a "clip" (success) or a "hit" (failure). Training the classification tree includes evaluating the performance of a given plurality of weights in classifying entries from the corpus of training data, and generating one or more new candidate plurality of weights based on the evaluation. By repeating the training process, the machine learning algorithm iterates towards a plurality of weights optimized for classifying moves as "clips" or "hits."

[0073] A single classification tree is very sensitive to the content of the corpus of training data and is prone to generating overfitted solutions. Thus, even small changes to the corpus of training data can result in very different classification trees. Therefore, the machine learning algorithm is configured to generate a forest of classification trees, where each classification tree in the forest is trained using a randomly selected subset of the training data. In this exemplary embodiment, the machine learning algorithm is configured to generate a forest that includes 500 classification trees. In this case, the classification decision is based on a simple majority vote of the individual classifications of the trees in the forest.

[0074] The machine learning algorithm is further configured to generate one or more rules based on the forest of trees. The one or more rules are generated by a rule-based learner, for example, using a simplified tree ensemble learner that generates one or more rules that approximate the output of the forest of trees. The resulting rules often do not perfectly reproduce the output of the forest of trees and often result in more classification errors, but are significantly simpler and easier to implement.

[0075] The most important parameter input to the machine learning algorithm (in terms of yielding the greatest improvement in classification accuracy) was found to be the DFT sliding window data, the second most important was the maximum peak in frequency content, the third was the Shannon-Wiener diversity index, and the fourth was the number and mean change of significant oscillations.

[0076] Thus, the signal processing unit 303 is configured to receive signals corresponding to the detected movement of the tool head 201; record inertial data defining the detected movement of the tool head; and determine whether the clamp has moved to the closed position based on the inertial data. The signal processing unit 303 primarily operates by detecting characteristic vibrations resulting from successful unclamping, the presence of which is indicative of successful unclamping.

[0077] If the signal processing unit 303 determines that the detected movement corresponds to a successful release of the POPP clamp, the signal processing unit 303 is configured to illuminate the LED 305 to inform an operator of the declamping tool that a successful release has occurred. The signal processing unit 303 is further configured to transmit a signal to the wireless communication module 307 indicating that a "clip" has been detected.

[0078] The wireless communication module 307 is configured to transmit a signal to a receiver indicating that a "clip" has been detected. In an automotive production line, this would allow the number of clips detected by a given tool at a given location in the production line to be recorded, thus providing traceability data (e.g., the number of successfully operated clamp closures per vehicle performed at a given time by a given operator with a particular tool).

[0079] In this exemplary embodiment, the wireless communication module 307 is configured to transmit messages wirelessly. Alternative embodiments of the invention may instead incorporate a wired communication link. The wireless communication module 307 is further configured to transmit a periodic "heartbeat" signal to the receiver. The "heartbeat" signal includes a "low battery" flag that indicates the battery status of the declamping tool 200. The channel selection 309 includes one or more switches, the position of which indicates a selected one of multiple frequency bands (or channels) over which the wireless communication module 307 may communicate with the receiver.

[0080] The receiver is configured to receive a signal from the wireless communication module 307 indicating that a "clip" has been detected. The receiver includes a "closed clamp signal received" LED and is configured to illuminate the "closed clamp signal received" LED in response to receiving a signal from the wireless communication module 307 indicating that a "clip" has been detected. The receiver is configured to communicate with a quality control system such that the quality control system is updated when the clamp is successfully released. The receiver is further configured to receive a "heartbeat" signal from the wireless communication module 307. The receiver includes a "low battery" LED and is configured to illuminate the "low battery" LED in response to receiving a "heartbeat" signal with a "low battery" flag set. The receiver includes a channel selection switch, the position of which indicates a selected one of a number of frequency bands (or channels) over which the receiver may communicate with the wireless communication module 307. The receiver further includes a "tool present" LED, a "data received" LED, a "receiver OK" LED, and a "power on" LED configured to illuminate as long as a "heartbeat" signal is periodically received.

[0081] During operation, the inertial measurement unit 301 detects movement of the tool head 201. When the tool head 201 is used to release the POPP clamp, vibrations are induced in the tool head 201 that are detected by the inertial measurement unit 301. The vibrations induced by the release of the POPP clamp are large enough to satisfy a trigger event condition. Thus, the signal processing unit 303 records inertial data defining the movement of the tool head 201. The signal processing unit 303 then calculates a number of parameters for use in determining whether the inertial data corresponds to a successful unclamping, the number of parameters including a change in orientation of the unclamping tool 201, frequency components of the inertial data, and a number of statistics. The number of parameters include inputs to one or more rules applied by the signal processing unit 303, the output of which determines whether the detected movement corresponds to the clamp being moved to a closed position.

[0082] FIG. 4 shows a flow chart illustrating steps of a method 400 for operating a pre-opened, pre-placed clamp operable between an open position and a closed position in accordance with a second embodiment of the present invention.

[0083] The first step of the method, represented by box 401, involves releasing the clamp from an open position using a tool head, which is included in a declamping tool, and the release is performed by an operator of the declamping tool, which is typically a human operator, but which may be a machine in embodiments of the invention.

[0084] A second step of the method, represented by box 403, includes detecting tool head motion resulting from the release. Using the tool head to release the clamp from the open position causes the tool head to vibrate. Thus, releasing the clamp from the open position is associated with tool head motion, which includes both the operator motion of the tool head to release the clamp and the vibration of the tool head induced by the release of the clamp. The detecting includes detecting the tool head motion about all three planes of motion, as well as the rotation of the tool head about its pitch, roll and yaw.

[0085] An optional third step of the method, represented by box 405, involves determining that a trigger event has occurred based on the detected motion. A trigger event includes the detection of four consecutive significant vibrations, where a significant vibration is detected within 980 ms. -2 Includes accelerations exceeding

[0086] A fourth step of the method, represented by box 407, includes recording inertial data defining the detected movement of the tool head. The recording is performed in response to detecting a trigger event and includes recording the inertial data immediately before and after the trigger event. Thus, the recorded inertial data includes data defining the movement preceding the trigger event and data defining the movement following the trigger event.

[0087] A fifth step of the method, represented by box 409, includes identifying frequency components of the inertial data. Identifying frequency components includes performing a Fourier transform on the recorded inertial data.

[0088] A sixth step of the method represented by box 411 includes determining whether the clamp has moved to the closed position based on the identified frequency components. If the clamp is successfully released and thus clamping the hose as expected, the frequency components will include one or more frequencies characteristic of a clamp successfully moving to the closed position. Determining whether the clamp has moved to the closed position includes evaluating one or more parameters. The one or more parameters may include a change in orientation of the declamping tool. Determining whether the clamp has moved to the closed position may therefore include determining a change in orientation of the declamping tool.

[0089] The one or more parameters may include a Pearson correlation coefficient, a Shannon-Wiener diversity index, a kurtosis or a skewness of the recorded inertial data. Determining whether the clamp has moved to the closed position includes applying one or more rules. The one or more rules define characteristics of a movement that is deemed to correspond to the clamp moving to the closed position. The one or more rules are based on an output of a classification tree. More specifically, the one or more rules are based on an output of a majority vote by a plurality of classification trees. The classification tree operates to determine whether the inertial data defines a detected movement that corresponds to a successful release of the clamp. The classification tree performs the classification based on a plurality of weights predetermined by a machine learning algorithm.

[0090] FIG. 5 shows a flow chart illustrating steps of a method 500 for manufacturing signal processing electronics for a clamp installation tool according to a third embodiment of the present invention.

[0091] The tool includes a tool head configured to engage the clamp and inertial measurement electronics configured to provide measurements of motion of the tool head, The clamp is operable between an open position and a closed position.

[0092] The first step of method 500, represented by box 501, involves providing signal processing electronics configured to receive a signal corresponding to a measurement.

[0093] An optional second step of method 500, represented by box 503, involves determining one or more rules for classifying detected motion as either corresponding to the clamp being moved to a closed position or not. The one or more rules define the characteristics of motion that are deemed to correspond to the clamp being moved to a closed position.

[0094] Determining the one or more rules includes running a machine learning algorithm. Running the machine learning algorithm includes training a classification tree based on a corpus of training data. The corpus of training data includes a plurality of entries, each entry corresponding to a past movement that corresponds to either a "clip" or a "hit". Training the classification tree includes evaluating the performance of a given plurality of weights in classifying entries from the corpus of training data and generating one or more new plurality of candidate weights based on the evaluation. By repeating the training process, the machine learning algorithm iterates towards a plurality of weights optimized for classifying movements as a "clip" or a "hit".

[0095] A single classification tree is very sensitive to the content of the corpus of training data and is prone to generating overfitted solutions. Thus, even small changes to the corpus of training data can result in very different classification trees. Thus, running the machine learning algorithm includes generating a forest of classification trees, where each classification tree in the forest is trained using a randomly selected subset of the training data. In this exemplary embodiment, running the machine learning algorithm includes generating a forest that includes 500 classification trees. In this case, the classification decision is based on a simple majority vote of the individual classifications of the trees in the forest.

[0096] Running the machine learning algorithm further includes generating one or more rules based on the forest of trees. The one or more rules are generated by a rule-based learner, for example, using a simplified tree ensemble learner that generates one or more rules that approximate the output of the forest of trees. The resulting rules often do not perfectly reproduce the output of the forest of trees and often result in more classification errors, but are significantly simpler and easier to implement.

[0097] A third step of method 500, represented by box 505, includes programming the signal processing electronics to record data related to the measurement, identifying a frequency component of the data, and determining whether the clamp has moved to a closed position based on the identified frequency component. Determining whether the clamp has moved to a closed position includes applying one or more rules.

[0098] Although the present invention has been described and illustrated with reference to specific embodiments, those skilled in the art will appreciate that the present invention is susceptible to many different variations not specifically illustrated herein. Merely by way of example, some possible variations are described herein.

[0099] In alternative embodiments of the invention, some or all of the functionality provided by the signal processing unit 303 in the first embodiment is instead provided by a receiver. In such embodiments, the wireless communication module 307 is configured to transmit inertial data from the inertial measurement unit 301 to the receiver, which then determines whether the detected movement constitutes a "clip" or a "hit."

[0100] In alternative embodiments of the invention, the signal processing electronics 303 is implemented with one or more of a processor, a field programmable gate array, a programmable logic device, and discrete electronic components. In alternative embodiments of the invention, the signal processing electronics 303 is implemented in whole or in part as software modules.

[0101] Where the foregoing description refers to integers or elements having known, obvious or foreseeable equivalents, such equivalents are incorporated herein as if set forth separately. Reference should be made to the claims to determine the true scope of the invention, which should be interpreted to encompass any such equivalents. It will also be understood by the reader that any integers or features of the invention described as preferred, advantageous, convenient, etc. are optional and do not limit the scope of the independent claims. Furthermore, it should be understood that any such integers or features may be a possible advantage in some embodiments of the invention, but may be undesirable and therefore may not be present in other embodiments.

Claims

1. 1. A method of operating a pre-open, pre-positioned clamp movable between an open position and a closed position, the method comprising: Releasing the clamp from the open position using a tool head; using an accelerometer and a gyroscope to measure movement of the tool head resulting from the release; recording data associated with said measurements; identifying a frequency content of said data; and determining whether the clamp has moved to the closed position based on the identified frequency components; Determining whether the clamp has moved to a closed position includes calculating a number of parameters derived from the data or from frequency components of the data and using the parameters to assess whether the clamp has moved to a closed position, the number of parameters comprising: - correlation coefficient, - diversity index, - A measure of kurtosis, - A measure of skewness, - the amplitude of the maximum peak in the frequency domain, the frequency of the maximum peak in the frequency range, and the amplitude and / or frequency of at least one other frequency peak The method includes at least one of the following:

2. The method of claim 1 , wherein taking measurements of motion includes measuring motion about all three planes of motion and rotation about pitch, roll, and yaw.

3. The method of claim 1 or 2, wherein the movement comprises vibration.

4. The method of any one of claims 1 to 3, wherein the frequency components include a component related to one or more frequencies characteristic of the clamp being moved to the closed position.

5. The method of any one of claims 1 to 4, further comprising determining the occurrence of a trigger event based on the measured movement.

6. Determining the occurrence of a trigger event includes detecting a significant vibration, the significant vibration being within 500 ms -2 Acceleration above a larger threshold and -3 The method of claim 5 , further comprising at least one of: a rate of change of acceleration exceeding a greater threshold.

7. The method of claim 6 , wherein determining the occurrence of the triggering event comprises detecting at least four consecutive significant vibrations.

8. A method according to any one of claims 5 to 7, wherein recording is performed in response to said triggering event.

9. A method according to any one of claims 5 to 8, wherein the recorded data includes data defining movement preceding the trigger event and data defining movement following the trigger event.

10. The method of any one of claims 1 to 9, wherein identifying the frequency components comprises performing a Fourier transform.

11. The method of any one of claims 1 to 10, wherein determining whether the clamp has moved to a closed position comprises identifying a change in orientation of the declamping tool.

12. The method of any one of claims 1 to 11, wherein determining whether the clamp has moved to a closed position comprises calculating a plurality of parameters of or derived from the recorded data and using those parameters to assess whether the clamp has moved to a closed position, the plurality of parameters comprising two or more different parameters from the group consisting of: - correlation coefficient, - diversity index, - Counting significant vibrations, the average rate of change of the significant vibration; - rate of change of acceleration, - rate of change of angular velocity, - A measure of kurtosis, - A measure of skewness, - Maximum value of one or more of acceleration, angular velocity, rate of change of acceleration, and rate of change of angular velocity, The minimum value of one or more of acceleration, angular velocity, rate of change of acceleration, and rate of change of angular velocity, - Average value of one or more of acceleration, angular velocity, rate of change of acceleration, and rate of change of angular velocity; and - First and / or third quartile of one or more of acceleration, angular velocity, rate of change of acceleration, rate of change of angular velocity.

13. 13. The method of any one of claims 1 to 12, wherein determining whether the clamp has moved to a closed position comprises applying one or more rules defining characteristics of movement that are deemed to correspond to the clamp moving to the closed position.

14. 14. The method of claim 1, wherein determining whether the clamp has moved to the closed position comprises evaluating an output of a sliding window algorithm, the sliding window algorithm comprising calculating a ratio of an amplitude of a first frequency band to an amplitude of a second frequency band.

15. The method of claim 14 , wherein the sliding window algorithm includes calculating at least 10 ratios.

16. 16. A method according to claim 14 or 15, when dependent on claim 13, wherein the one or more rules act based on the output of the sliding window algorithm.

17. The method of any one of claims 1 to 16, wherein determining whether the clamp has moved to the closed position comprises active detection of movement corresponding to the clamp moving to the closed position.

18. A method according to any preceding claim, wherein taking the measurements comprises measuring movement in each direction and about each axis at a sample rate of greater than 3 kHz.

19. 1. A clamp installation tool for manipulating a pre-opened, pre-placed clamp movable between an open position and a closed position, the tool comprising: a tool head for releasing the clamp from said open position; inertial measurement electronics including accelerometers and gyroscopes configured to provide measurements of motion of the tool head; and signal processing electronics configured to record data associated with a measurement, identify a frequency component of the data, and determine whether the clamp has moved to the closed position based on the identified frequency component; Determining whether the clamp has moved to the closed position includes calculating a number of parameters derived from the data or from frequency components of the data and using the parameters to assess whether the clamp has moved to the closed position, the number of parameters comprising: - correlation coefficient, - diversity index, - A measure of kurtosis, - A measure of skewness, - the amplitude of the maximum peak in the frequency domain, the frequency of the maximum peak in the frequency range, and the amplitude and / or frequency of at least one other frequency peak 13. A clamp installation tool comprising:

20. 20. The clamp installation tool of claim 19, wherein the clamp is suitable for clamping a hose to a spigot.

21. the clamp in the open position is held at an expanded diameter greater than an outer diameter of the hose; the clamp in the closed position is disposed around the hose at a diameter substantially the same as the outer diameter of the hose; 21. The clamp installation tool of claim 20, wherein the clamp is resiliently deformable such that in the closed position the clamp applies a compressive force to the hose to clamp the hose to the spigot.

22. A clamp application tool according to any one of claims 19 to 21, wherein the clamp has a diameter of between 14mm and 43mm when in the closed position.

23. 1. A method of manufacturing signal processing electronics for a clamp application tool adapted to operate a pre-opened, pre-placed clamp operable between an open position and a closed position, the tool including a tool head configured to engage the clamp and inertial measurement electronics including an accelerometer and a gyroscope configured to provide measurements of motion of the tool head, the method comprising: providing signal processing electronics configured to receive a signal corresponding to said measurement; programming the signal processing electronics to record data associated with the measurements, identify frequency components of the data, and determine whether the clamp has moved to the closed position based on the identified frequency components; Determining whether the clamp has moved to a closed position includes calculating a number of parameters derived from the data or from frequency components of the data and using the parameters to assess whether the clamp has moved to a closed position, the number of parameters comprising: - correlation coefficient, - diversity index, - A measure of kurtosis, - A measure of skewness, - the amplitude of the maximum peak in the frequency domain, the frequency of the maximum peak in the frequency range, and the amplitude and / or frequency of at least one other frequency peak The method includes at least one of the following:

24. 24. The method of claim 23, further comprising determining one or more rules for classifying detected movement as either corresponding to the clamp being moved to the closed position or not, the one or more rules defining characteristics of movement that is deemed to correspond to the clamp being moved to the closed position, and determining whether the clamp has been moved to the closed position comprises applying the one or more rules.

25. 25. The method of claim 24, comprising operating a machine learning algorithm to determine the one or more rules.

Images