Method for classifying motion, motion classification system, computing unit and computer program product
The system, which uses multiple peripheral classification elements and a central classification unit, generates accurate motion classification values using sensor data, solving the problem of insufficient motion classification accuracy in existing technologies and achieving high-precision motion recognition and performance evaluation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ROBERT BOSCH GMBH
- Filing Date
- 2025-11-07
- Publication Date
- 2026-05-08
AI Technical Summary
Existing motion classification methods lack sufficient accuracy, making it difficult to effectively utilize motion data from multiple body parts for comprehensive analysis.
The system employs multiple peripheral classification elements and a central classification unit to classify human movement using sensor data. The central classification unit receives and processes the data from the peripheral classification elements, generating and outputting classification values, including count values and speed values.
It improves the accuracy and reliability of motion classification, enabling more precise identification of motion types and counting of motion cycles, and provides performance evaluation and pause detection.
Smart Images
Figure CN121997198A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method for classifying motions and a motion classification system.
[0002] Furthermore, the present invention also relates to a computing unit and a computer program product. Background Technology
[0003] Methods for classifying motion are known from existing technologies. Summary of the Invention
[0004] The objective of this invention is to provide an improved method for classifying motion and a motion classification system.
[0005] This task is solved by the method and system for classifying motion according to the present invention. Advantageous embodiments are the subject of extended technical solutions.
[0006] According to one aspect, a method is provided for classifying motion using a motion classification system with multiple peripheral classification elements and a central classification unit, wherein the peripheral classification elements are arranged on different body parts of a person and classify the motion of the person based on sensor data, the method comprising: The central classification unit receives classification data from the plurality of peripheral classification elements, wherein each of the classification data includes at least one peripheral classification value; The central classification unit classifies the motion and generates classification values based on the peripheral classification values of the plurality of peripheral classification elements; and The central classification unit outputs the classification value to the display unit.
[0007] This provides the following technical advantages: an improved method for classifying motion can be provided. Here, the method is implemented through a motion classification system comprising multiple peripheral classification elements and a central classification unit.
[0008] Here, multiple peripheral classification elements are arranged on the body parts in motion and configured to perform classification of a person's movements based on sensor values. A central classification unit is technically connected to the multiple peripheral classification elements and receives classification data from them. The classification data includes at least peripheral classification values, where each peripheral classification value represents the classification of the movement performed by the corresponding peripheral classification element.
[0009] The central classification unit creates a motion classification value based on multiple peripheral classification values from multiple peripheral classification elements. Therefore, the motion classification value is based on multiple peripheral classification values from each peripheral classification element. This enables higher accuracy in classifying human motion.
[0010] By arranging different peripheral classification elements on different parts of a person's body, multiple movements of various body parts can be considered in the classification process. This can further improve the accuracy of motion classification. Subsequently, the corresponding classification values are output to the display unit. This allows the system to display the relevant motion classification to the user.
[0011] According to one implementation, if one of the peripheral classification values provided by a plurality of peripheral classification elements is provided by at least a predetermined number of peripheral classification elements, then the classification value obtained by the central classification unit is consistent with the corresponding peripheral classification value.
[0012] This provides the following technical advantage: it enables higher accuracy in motion classification. Here, if a predetermined number of peripheral classification elements provide the same peripheral classification value, then the classification value generated by the central classification unit will be consistent with the peripheral classification values provided by these peripheral classification elements.
[0013] According to one embodiment, the classification data further includes at least one motion value, wherein the method further includes: The central classification unit calculates the count value of the periodic portion of the motion based on the motion values of the plurality of peripheral classification elements; and The central classification unit outputs the count value to the display unit.
[0014] This provides the following technical advantages: in addition to classification values, it can also provide count values of the periodic partial movements of a person's motion. Here, in addition to the peripheral classification values, multiple peripheral classification elements also provide motion values. The count values created by the central classification unit are based on the motion values of the multiple peripheral classification elements, thereby improving the accuracy of the created count values.
[0015] According to one embodiment, motion values provided by peripheral classification elements describe incremental motion segments of periodic partial motions of corresponding body parts on which these peripheral classification elements are arranged, wherein obtaining the count values includes: Select peripheral classification elements that provide peripheral classification values that are consistent with the classification values of the central classification unit; The motion values provided by the peripheral classification elements at different time points are summed, and a cumulative motion value is generated for at least one selected peripheral classification element; If at least one cumulative motion value of at least one selected peripheral classification element reaches or exceeds the value of a complete cycle of the periodic partial motion, then the average motion value is calculated based on the cumulative motion values of the selected peripheral classification elements; and Once the average motion value reaches or exceeds a predetermined share of a complete cycle of the periodic motion, the count value is incremented by "1".
[0016] This yields the following technical advantages: it enables improved accuracy of the count values. To this end, peripheral classification elements are first selected whose peripheral classification values are consistent with those of the central classification unit.
[0017] Therefore, only the motion values of the peripheral classification elements that provide a correct classification of the motion are considered. Here, the motion values provided by multiple peripheral classification elements demonstrate the segmentation of the periodic portion of the person's motion.
[0018] Here, the motion values provided by the peripheral classification elements at different time points are cumulatively accumulated until the cumulative motion values of the peripheral classification elements describe a complete cycle of the periodic partial motion.
[0019] Based on this, the average motion value is obtained from all motion values of the selected peripheral classification elements.
[0020] Once the average motion value of all selected peripheral classification elements describes a complete cycle of periodic motion, the counter increments by 1. This allows for extremely precise determination of the periodic motion count.
[0021] On the one hand, only peripheral classification elements are considered in the calculation of count values. On the other hand, different part movements of different body parts are considered simultaneously in the average motion value.
[0022] According to one embodiment, the method further includes: If the peripheral classification value provided by a peripheral classification element matches the classification value obtained by the central classification unit, then the corresponding peripheral classification element has successfully obtained a value; and The central classification unit evaluates the performance of multiple peripheral classification elements based on the success value.
[0023] This provides the following technical advantages: the performance of classification by each peripheral classification element can be calculated, and the generated performance data can be displayed to users.
[0024] Alternatively or additionally, this performance may be taken into account in subsequent classification steps, in that poor-performing peripheral classification elements are no longer considered in subsequent steps.
[0025] According to one implementation, obtaining the count value further includes: The motion velocity of the periodic partial motion obtained by the peripheral classification element is obtained based on the motion value provided by the peripheral classification element and the timestamp provided by the peripheral classification element for its respective motion value, and the velocity value is generated by the central classification unit, wherein the timestamp defines the time points at which the motion value was obtained by the peripheral classification element.
[0026] This provides the following technical advantages: the motion speed of periodic partial motion can be determined based on the motion values of different peripheral classification elements. This motion speed can then be displayed to the user as additional information.
[0027] According to one embodiment, if the time interval between peripheral classification values and / or motion values provided sequentially by one of a plurality of peripheral classification elements exceeds a predetermined boundary value, then the peripheral classification values and / or motion values of that peripheral classification element are not considered when calculating the classification values and / or count values.
[0028] This results in the following technical advantages: it allows for higher accuracy in obtaining classification and / or count values.
[0029] When subsequently obtaining classification or count values, peripheral classification elements that have not transmitted new classification data to the central classification unit for a period longer than the predetermined time period are not considered, and the transmitted peripheral classification or motion values are not considered in subsequent steps.
[0030] Therefore, faulty peripheral classification elements or such peripheral classification elements can be excluded from motion classification: these peripheral classification elements are arranged on body parts that do not perform periodic movements during that time.
[0031] According to one embodiment, the method further includes: If the number of peripheral classification elements that provide peripheral classification values consistent with the classification values obtained by the central classification unit is less than a predetermined number, and the time interval between peripheral classification values provided sequentially by at least one peripheral classification element is equal to or less than a predetermined time interval, then the central classification unit determines a pause in motion.
[0032] This provides the following technical advantages: it allows for the precise detection of pauses in the user's movement. A pause is detected when fewer than a predetermined number of peripheral classification elements provide peripheral classification values consistent with those of the central classification unit, but simultaneously, these peripheral classification elements provide classification data to the central classification unit at intervals shorter than a predetermined time interval.
[0033] Here, the predefined time interval describes the maximum time interval allowed to elapse between two temporally successive classification data provided by the corresponding peripheral classification element before the peripheral classification element is evaluated as inactive.
[0034] This means that the peripheral classification elements are still providing classification data at time intervals shorter than the predefined time interval, which in turn indicates that the motion classification system is still operating.
[0035] By recognizing pauses in movement, the system retains the generated classification and count values, thus allowing the motion classification system to wait for the user to resume movement. Instead of a complete system reset, the method allows the system to continue based on the displayed values when movement resumes.
[0036] According to one embodiment, the method further includes: If the number of peripheral classification elements that provide peripheral classification values consistent with the classification values obtained by the central classification unit is less than a predetermined number, and the time interval between peripheral classification values provided sequentially by at least one peripheral classification element exceeds a predetermined time interval, then the motion classification system is in an idle state.
[0037] This provides the following technical advantages: it allows for the accurate determination of the idle state of a motion classification system.
[0038] This occurs when the number of peripheral classification elements providing classification values consistent with the classification unit falls below a predetermined number again, and a time interval longer than a predetermined time interval has elapsed since the last time classification data was provided by the peripheral classification elements. This means the system is in an idle state.
[0039] According to one embodiment, the method further includes: If the idle state of the classification system is obtained, the central classification unit resets the classification value and / or count value and / or speed value and / or success value.
[0040] This provides the following technical advantages: if the classification system is idle, by resetting the above values, the motion classification system can create an accurate classification of the motion when reused without being affected by the previous classification.
[0041] According to one implementation, the classified exercise comes from one of the following lists: walking, running, hiking, mountaineering, swimming, cross-country skiing, and cycling, wherein the count value involves the periodic portion of the movement from the following lists: stride, swimming arm stroke, swimming leg kick, and cycling pedal rotation.
[0042] This provides the following technical advantages: the method of this invention can classify different motions that have periodic partial motions.
[0043] According to one aspect, a motion classification system having a central classification unit and a plurality of peripheral classification elements is provided, wherein each of the peripheral classification elements includes at least one motion sensor, wherein the motion classification system is configured to implement a method for classifying motion according to any of the above embodiments.
[0044] This provides the following technical advantages: an improved motion classification system is provided, which is configured to implement a method for classifying motion with the aforementioned technical advantages.
[0045] According to one embodiment, the peripheral sorting elements and / or the central sorting unit are configured as wearable devices.
[0046] This provides the following technical advantages: the classification system can be easily carried by personnel, allowing the movement of that personnel to be classified by the motion classification system.
[0047] According to one aspect, a computing unit is provided, configured to implement a method for classifying motion according to any of the above embodiments.
[0048] According to one aspect, a computer program product including instructions is provided, which, when implemented by a data processing unit, cause the data processing unit to implement a method for classifying motion according to any of the above embodiments. Attached Figure Description
[0049] Embodiments of the invention are described below with reference to the accompanying drawings. The drawings show: Figure 1 A schematic diagram of a motion classification system according to one embodiment; Figure 2 A flowchart of a method for classifying motion according to one embodiment; Figure 3 Another flowchart of a method for classifying motion according to another embodiment; Figure 4 Another flowchart of a method for classifying motion according to another embodiment; Figure 5 Another flowchart of a method for classifying motion according to another embodiment; Figure 6 Another flowchart of a method for classifying motion according to another embodiment; and Figure 7 A schematic diagram of a computer program product. Detailed Implementation
[0050] Figure 1 A schematic diagram of a motion classification system 200 according to one embodiment is shown.
[0051] exist Figure 1 The image shows a user 300 with a motion classification system 200 according to the present invention. In the illustrated embodiment, the motion classification system 200 includes four peripheral classification elements 201, which are respectively positioned on the four limbs of the user 300.
[0052] In addition, the motion classification system 200 includes a central classification unit 203, which is located on the torso of the person 300 in the illustrated embodiment.
[0053] Each of the multiple peripheral classification elements 201 includes at least one motion sensor 223, through which the motion of the person 300 can be determined. The motion sensor 223 may be configured as, for example, an accelerometer, a gyroscope, or other motion sensors known from the prior art.
[0054] The peripheral classification element 201 is configured to independently classify the movement of the person 300 and generate peripheral classification value 207 based on the sensor value of the motion sensor 223.
[0055] Here, the peripheral classification value 207 describes the classification of the movement of the person 300, which has been independently generated by the corresponding peripheral classification element 201.
[0056] Different peripheral classification elements 201 are connected to the central classification unit 203 in terms of data technology, and send classification data 205, including at least peripheral classification values 207, to the central classification unit 203.
[0057] The central classification unit 203 generates a classification value 209 based on the peripheral classification value 207. Here, the classification value 209 represents the classification of the movement of the person 300 and is based on the peripheral classification value 207 of the peripheral classification element 201.
[0058] According to one embodiment, the central classification unit 203 generates the classification value 209 in such a way that the classification unit 203 adopts the peripheral classification value 207 transmitted by these peripheral classification elements 201 as long as the number of peripheral classification elements 201 that transmit the same peripheral classification value 207 reaches or exceeds a predetermined boundary value.
[0059] In the illustrated embodiment, a user 300 is shown in a running motion. Furthermore, the user 300 has a peripheral classification element 201 on each of its two arms and two legs.
[0060] Now, the central classification unit 203 can be configured such that if at least two of the peripheral classification elements 201 provide the same peripheral classification value 207, then the peripheral classification value 207 is adopted as the classification value 209.
[0061] For example, if at least one peripheral classification element 201 on one arm and one peripheral classification element 201 on one leg of person 300 classifies the movement of person 300 as running, then the central classification unit 203 will also classify the movement of person 300 as running.
[0062] According to one embodiment, in addition to the peripheral classification value 205, the peripheral classification element 201 also transmits the motion value 213 as part of the classification data 205 to the central classification unit 203.
[0063] The central classification unit 203 generates a count value 215 of the movement of personnel 300 based on the movement value 213 of the peripheral classification element 205.
[0064] Here, the motion value 213 of the peripheral classification element 201 describes the motion segments of the periodic partial motion of the person 300. In the illustrated embodiment, these periodic partial motions describe the back-and-forth swinging of the person 300's arms or the stepping motion of the legs.
[0065] Here, the count value 215 generated by the central classification unit 203 based on the motion value 213 quantifies the sequential execution of periodic partial movements. In the example shown, the count value 215 may, for example, describe the count of steps performed during a running motion.
[0066] According to one implementation, the peripheral classification element 201 also transmits the timestamp 221 as part of the classification data 205 to the central classification unit 203. Here, the timestamp 221 describes the time point at which the motion value 213 was collected.
[0067] The central classification unit 203 generates the velocity value 219 based on the timestamp 221 and the motion value 213.
[0068] Here, velocity value 219 describes the implementation velocity of the periodic partial motion demonstrated by motion value 213 of peripheral classification element 201.
[0069] According to one implementation, the central classification unit 203 also generates a success value 217. Here, for each peripheral classification element 201, a success value 217 is generated for that peripheral classification element 201 each time it provides a peripheral classification value 207 that is consistent with the classification value 209 generated by the central classification unit 203.
[0070] According to the present invention, each peripheral classification element 201 continuously classifies the movement of the user 300 during the movement and provides classification data 205, which includes at least peripheral classification values 207 and, if necessary, additionally includes movement values 213 and timestamps 221.
[0071] Here, the peripheral classification element 201 provides classification data 205 to the central classification unit 203 according to its own time rhythm.
[0072] Each peripheral classification element 201 provides classification data 205, especially when successful classification of motion is achieved.
[0073] Therefore, the central classification unit 203 receives classification data 205 from each peripheral classification element 201 at time-inconsistent intervals.
[0074] Each time the peripheral classification value 207 of the peripheral classification element 201 matches the previously generated classification value 209 of the central classification unit 203, a corresponding success value 217 is assigned to the peripheral classification element 201. The performance of different peripheral classification elements 201 can be evaluated through the corresponding success value 217.
[0075] In the illustrated embodiment, the motion classification system 200 also includes a display unit 211. After generating classification values 209 and / or count values 215 and / or speed values 219 and / or success values 217, these values are provided to the display unit 211 by the central classification unit 203 and can be displayed to the user 300 through the display unit.
[0076] In the illustrated embodiment, the peripheral classification element 201 and the central classification unit 203 are configured as wearable devices and can be fixed to different parts of the user's body 300. Here, the central classification unit 203 is implemented on the computing unit 225 of the motion classification system 200.
[0077] Alternatively, the central classification unit 203 may also be implemented on an external computing unit 225, such as on a smartphone, tablet, or personal computer.
[0078] In addition to running, as shown here, walking, hiking, mountaineering, swimming, cross-country skiing, cycling, or other sports that include periodic movements can also be classified by the method according to the invention and the sports classification system 200 according to the invention.
[0079] Here, count value 215 can display the count value of gait, swimming arm stroke, cycling pedal rotation, or other periodic partial motions.
[0080] Figure 2 A flowchart is shown for a method 100 for classifying motion according to one embodiment.
[0081] In order to classify the movement of users, in the first method step 101, the central classification unit 203 receives classification data 205 from multiple peripheral classification elements 201.
[0082] In another method step 103, the central classification unit 203 classifies the motion based on the peripheral classification values 207 of the plurality of peripheral classification elements 201 and generates classification values 209.
[0083] In another method step 105, the central classification unit 203 provides the classification value 209 to the display unit 211.
[0084] Figure 3 Another flowchart of a method 100 for classifying motion according to another embodiment is shown.
[0085] Figure 3 The implementation method in is based on Figure 2 The implementation methods described herein include all those in [the document / concept]. Figure 2 The methods and steps described in the document.
[0086] In the illustrated embodiment, the classification data 205 provided by the peripheral classification element 201 includes motion values 213 in addition to the parallel peripheral classification values 207.
[0087] In method step 107, the central classification unit 203 generates a count value 215 based on the motion value 213 of the peripheral classification element 201.
[0088] In another method step 111, motion values 213 provided by peripheral classification element 201 at different time points are accumulated, and cumulative motion values 213 are obtained for at least one selected peripheral classification element 201.
[0089] Here, It is the k-th motion value 213 of the i-th peripheral classification element 201. It is the (k-1)th motion value 213 of the i-th peripheral classification element 201, and It is the motion change from the (k-1)th motion value 213 to the kth motion value 213 that immediately follows in time, where the following condition is met: Therefore, in method step 109, a peripheral classification element 201 is selected such that the peripheral classification value 207 it provides is consistent with the classification value 209 of the central classification unit 203.
[0090] Here, the motion values 213 of the peripheral classification elements 201 describe the periodic motion of the body parts to which these peripheral classification elements 201 are located. The motion values 213 can describe incremental segments of the periodic motion. For example, the peripheral classification elements 201 can be configured to collect a predetermined number of motion values 213 within each complete cycle of the periodic motion.
[0091] Here, the motion value 213 can be expressed in radians, where a complete cycle is represented by radians. Motion values can also be expressed as percentages or angles.
[0092] The motion value 213 may also include phase information. This phase information can be used to correlate the motions of the various body parts on which the various peripheral classification elements 201 are arranged.
[0093] For example, it can be determined whether the movements are performed in phase or out of phase. For example, when walking or running, the in-phase swings of the right arm and left leg, and the left arm and right leg, can be identified.
[0094] In method step 109, a peripheral classification element 201 is selected whose motion value 213 describes the motion implemented in the same phase.
[0095] In another method step 113, if at least one cumulative motion value of at least one selected peripheral classification element 201 reaches or exceeds the value of a complete cycle of periodic partial motion, then the average motion value is obtained based on the cumulative motion value 213 of the selected peripheral classification element 201.
[0096] Here, the average motion value is obtained as a weighted average of the set of peripheral classification elements 201 in the following manner: in, It is the average motion value. It is the cumulative motion value 213 of the i-th peripheral classification element 201, and These are the weight values associated with the i-th peripheral classification element 201. Set I is the set of all peripheral classification elements 201 that have been used to calculate the average motion value. For this purpose, for example, peripheral classification elements 201 with high performance can be selected first.
[0097] in, It is the peripheral classification value 207 of the i-th peripheral classification element 201. It is classification value 209 for central taxonomic unit 203. It is the cumulative motion value 213 of the i-th peripheral classification element 201, and It is a vector display of all cumulative motion values 213 of the selected peripheral classification element 201.
[0098] weight value It can be scaled according to the performance of each peripheral classification component 201: in, Displays the size of the collection.
[0099] Here, the motion value 213 of the peripheral classification element 201 describes the motion segment of periodic motion, especially the motion segment of periodic motion of various body parts of the user 300.
[0100] By accumulating the motion values 213 provided sequentially over time by different peripheral classification elements 201, for each selected peripheral classification element 201, after accumulating a certain number of motion values 213 provided sequentially over time, a predetermined share of a complete cycle of the periodic motion obtained by that peripheral classification element 201 is achieved. This predetermined share can be determined as a quarter cycle, a half cycle, or a three-quarter cycle. Of course, other values are also possible.
[0101] Here, the peripheral classification element 201 can be configured such that motion values 213 are collected within a predetermined number of incremental measurement steps during the period of periodic motion.
[0102] For example, for different peripheral classification elements 201, ten motion values 213 can be collected for each cycle of the periodic motion.
[0103] Here, these ten motion values 213 are collected sequentially at different points in time and transmitted to the central classification unit 203. After accumulating the ten sequentially collected motion values, a complete cycle of the periodic motion is achieved.
[0104] Here, the average motion value is calculated based on all the cumulative motion values of the selected peripheral classification element 201.
[0105] As the current motion value 213 of the peripheral classification element 201 is continuously provided during the motion, and correspondingly, the cumulative motion value 213 of each peripheral classification element 201 increases over time, the average motion value also has a value that continuously becomes larger over time.
[0106] Subsequently, in another method step 115, once the average motion value reaches or exceeds a predetermined share of a complete cycle of the periodic motion, i.e., multiple times a complete cycle, the count value 215 is incremented by 1 count value.
[0107] The selected peripheral classification element 201 for the calculation of the count value 215 can change over time according to the corresponding provided peripheral classification value 207. In particular, the number of selected peripheral classification elements 201 can change over time based on the provided peripheral classification value 207.
[0108] Therefore, the count value 215 can be accurately obtained by calculating the average motion value.
[0109] Figure 4 Another flowchart of a method 100 for classifying motion according to another embodiment is shown.
[0110] Figure 4 The implementation method in is based on Figure 2 The implementation methods described herein include all those in [the document / concept]. Figure 2 The methods and steps described in the document.
[0111] In the illustrated embodiment, in method step 117, if the peripheral classification value 207 provided by each peripheral classification element 201 is consistent with the classification value 209 obtained by the central classification unit 203, then the peripheral classification elements 201 obtain a success value 217.
[0112] Based on a success value of 217, the feedback can be calculated. : in, Let 217 be the success value of the i-th peripheral classification element 201, where the expected success can be obtained by considering the peripheral classification element 201 with the highest success value.
[0113] In another method step 119, the central classification unit 203 evaluates the performance of multiple peripheral classification elements 201 based on success value 217.
[0114] Here, when the number of success values 217 is greater, the corresponding peripheral classification element 201 obtains a higher performance value, and when the number of success values 217 is less, the corresponding peripheral classification element 201 obtains a correspondingly lower performance value.
[0115] Figure 5 Another flowchart of a method 100 for classifying motion according to another embodiment is shown.
[0116] Figure 5 The implementation method in is based on Figure 3 The implementation methods described herein include all those in [the document / concept]. Figure 3 The methods and steps described in the document.
[0117] In the illustrated embodiment, in method step 121, the central classification unit 203 determines the velocity value 219 of the periodic motion based on the motion value 213 and the timestamp 221.
[0118] In method step 105, the corresponding motion value 213 and / or velocity value 219 can be provided to the display unit 211.
[0119] Figure 6 Another flowchart of a method 100 for classifying motion according to another embodiment is shown.
[0120] Figure 6 The implementation method in is based on Figure 2 The implementation methods described herein include all those in [the document / concept]. Figure 2 The methods and steps described in the document.
[0121] In method step 123, if the number of such peripheral classification elements 201 is less than a predetermined number: their peripheral classification value 207 is consistent with the classification value 209 of the central classification unit 203, and if the time interval between peripheral classification values 207 provided sequentially in time reaches or is less than a predetermined time interval, then the pause in the movement of the user 300 is determined.
[0122] Conversely, in method step 125, if the number of peripheral classification elements 201 that provide peripheral classification values 207 consistent with classification value 209 is again less than a predetermined number, and if the time interval between peripheral classification values 207 provided in quick succession exceeds a predetermined time interval, the idle state of motion classification system 200 is determined by central classification unit 201.
[0123] When the predetermined time interval is exceeded, it is considered that: before the excessively long period of time, no movement of the user 300 was obtained by the movement classification system 200, which is different from the situation that can be explained by the pauses between the movements of the user 300.
[0124] If the most recent transmission of the peripheral classification value 207 has been longer than a predetermined time, then this situation is instead identified as an idle state of the motion classification system 200.
[0125] In another method step 127, if the idle state of the motion classification system 200 is obtained, the central classification unit 203 resets the classification value 209 and / or count value 215 and / or velocity value 219 and / or success value 217 that were generated when classifying earlier motions.
[0126] Figures 2 to 6 The implementation of method 100 can also be combined with each other in a manner different from the above embodiments.
[0127] Figure 7 A schematic diagram of a computer program product 400 is shown, which includes instructions that, when implemented by a data processing unit, cause the data processing unit to implement a method 100 for classifying motion.
[0128] In the illustrated embodiment, the computer program product 400 is stored on a memory medium 401. Here, the memory medium 401 can be any memory medium known from the prior art.
Claims
1. A method (100) for classifying motions using a motion classification system (200), said motion classification system (200) having a central classification unit (203) and a plurality of peripheral classification elements (201), wherein, The peripheral classification elements (201) are arranged on different body parts (301) of the person (300) and classify the movements of the person (300) based on sensor data, respectively. The method includes: The central classification unit (203) receives (101) the classification data (205) of the plurality of peripheral classification elements (201), wherein the classification data (205) includes at least one peripheral classification value (207). The central classification unit (203) classifies the motion (103) based on the peripheral classification values (207) of the plurality of peripheral classification elements (201) and generates classification values (209); and The central classification unit (203) outputs (105) the classification value (209) to the display unit (211).
2. The method (100) according to claim 1, wherein, If one of the peripheral classification values (207) provided by the plurality of peripheral classification elements (201) has been provided by at least a predetermined number of peripheral classification elements (201), then the classification value (209) obtained by the central classification unit (203) is consistent with the corresponding peripheral classification value (207).
3. The method (100) according to claim 1 or 2, wherein, The classification data (205) further includes at least one motion value (213), wherein the method (100) further includes: The central classification unit (203) determines (107) the count value (215) of the periodic portion of the motion based on the motion values (213) of the plurality of peripheral classification elements (201); and The central classification unit (203) outputs (105) the count value (215) to the display unit (211).
4. The method (100) according to claim 3, wherein, The motion values (213) provided by the peripheral classification elements (201) respectively describe the increasing motion segments of the periodic partial motion of the corresponding body parts (301) on which these peripheral classification elements (201) are arranged, wherein the calculation (107) of the count value (215) includes: Select a peripheral classification element (201) such as (109): whose peripheral classification value (207) is consistent with the classification value (209) of the central classification unit (203); The motion values (213) provided by the peripheral classification element (201) at different time points are accumulated (111), and a cumulative motion value (213) is generated for at least one selected peripheral classification element (201). If at least one cumulative motion value (213) of at least one selected peripheral classification element (201) reaches or exceeds the value of a complete cycle of the periodic partial motion, then an average motion value (113) is obtained based on the cumulative motion value (213) of the selected peripheral classification element (201); and Once the average motion value reaches or exceeds the value of a complete cycle of the periodic motion, the count value (215) is increased by the value "1" (115).
5. The method (100) according to any one of the preceding claims, wherein, The method also includes: If the peripheral classification value (207) provided by a certain peripheral classification element (201) is consistent with the classification value (209) obtained by the central classification unit (203), then a success value (217) is obtained (117) for the corresponding peripheral classification element (201); and The central classification unit (203) evaluates (119) the performance of the plurality of peripheral classification elements (201) based on the success value (217).
6. The method (100) according to any one of the preceding claims, wherein, The process of obtaining (107) also includes: The motion velocity of the periodic partial motion obtained by the peripheral classification element (201) is obtained based on the motion value (213) provided by the peripheral classification element (201) and the timestamp (221) provided by the peripheral classification element (201) for the respective motion value (213), and the velocity value (219) is generated by the central classification unit (203), wherein the timestamp (221) defines the time points at which the motion value (213) was obtained by the peripheral classification element (201).
7. The method (100) according to any one of the preceding claims, wherein, If the time interval between the peripheral classification value (207) and / or motion value (213) provided sequentially by one of the peripheral classification elements (201) exceeds a predetermined boundary value, then the peripheral classification value (207) and / or motion value (213) of that peripheral classification element (201) will not be considered when obtaining the classification value (209) and / or the count value (215).
8. The method (100) according to any one of the preceding claims, wherein, The method further includes: if the number of peripheral classification elements (201) that provide peripheral classification values (207) consistent with the classification values (209) obtained by the central classification unit (203) is less than a predetermined number, and the time interval between peripheral classification values (207) provided sequentially by at least one peripheral classification element (201) in time reaches or is less than a predetermined time interval, then the central classification unit (203) obtains (123) a pause in the movement.
9. The method (100) according to any one of the preceding claims, wherein, The method further includes: if the number of peripheral classification elements (201) that provide peripheral classification values (207) consistent with the classification values (209) obtained by the central classification unit (203) is less than a predetermined number, and the time interval between peripheral classification values (207) provided sequentially by at least one peripheral classification element (201) exceeds a predetermined time interval, then the motion classification system (200) obtained (125) is in an idle state.
10. The method (100) according to claim 9, wherein, The method further includes: if the idle state of the motion classification system (200) is obtained, the central classification unit (203) resets (127) the classification value (209) and / or the count value (215) and / or the speed value (219) and / or the success value (217).
11. The method (100) according to any one of the preceding claims, wherein, The classified sport is one of the following: walking, running, hiking, mountaineering, swimming, cross-country skiing, cycling, wherein the count value (215) relates to the periodic part of the following: gait, swimming arm stroke, swimming leg kick, cycling pedal rotation.
12. A motion classification system (200) having a central classification unit (203) and multiple peripheral classification elements (201), wherein, The peripheral classification element (201) includes at least one motion sensor (223), wherein the motion classification system (200) is configured to implement the method (100) for classifying motion according to any one of claims 1 to 11.
13. The motion classification system (200) according to claim 12, wherein, The peripheral classification element (201) and / or the central classification unit (203) are configured as wearable devices.
14. A computing unit (225) configured to implement a method (100) for classifying motion according to any one of claims 1 to 11.
15. A computer program product (400) comprising instructions that, when executed by a data processing unit, cause the data processing unit to execute a method (100) for classifying motion according to any one of claims 1 to 11.