Search

CN-121997198-A - Method for classifying motions, motion classification system, computing unit and computer program product

CN121997198ACN 121997198 ACN121997198 ACN 121997198ACN-121997198-A

Abstract

The invention relates to a method (100) for classifying movements by means of a movement classification system (200) with a plurality of peripheral classification elements (201) and a central classification unit (203), wherein the peripheral classification elements (201) are arranged on different body parts (301) of a person (300) and classify the movements of the person (300) on the basis of sensor data, respectively, the method comprising receiving (101) classification data (205) of the plurality of peripheral classification elements (201) by the central classification unit (203), wherein the classification data (205) each comprise at least one peripheral classification value (207), classifying (103) the movements by the central classification unit (203) on the basis of the peripheral classification values (207) of the plurality of peripheral classification elements (201) and generating classification values (209), and outputting (105) the classification values (209) to a display unit (211) by the central classification unit (203). The invention also relates to a motion classification system (200), a computing unit and a computer program product.

Inventors

  • JOHN JOSHY
  • U. Darren
  • F. Nordstrom

Assignees

  • 罗伯特·博世有限公司

Dates

Publication Date
20260508
Application Date
20251107
Priority Date
20241107

Claims (15)

  1. 1. A method (100) for classifying motions by a motion classification system (200), the 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 a person (300) and classify the motions of the person (300) based on sensor data, respectively, the method comprising: -receiving (101), by the central classification unit (203), classification data (205) of the plurality of peripheral classification elements (201), wherein the classification data (205) each comprise at least one peripheral classification value (207); classifying (103) said motion by said central classification unit (203) based on said peripheral classification values (207) of said plurality of peripheral classification elements (201) and generating classification values (209), and -Outputting (105) the classification value (209) by the central classification unit (203) to a display unit (211).
  2. 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) is provided by at least a predefined number of peripheral classification elements (201), the classification value (209) solved by the central classification unit (203) corresponds to the corresponding one of the peripheral classification values (207).
  3. 3. The method (100) according to claim 1 or 2, wherein the classification data (205) further comprises at least one motion value (213), respectively, wherein the method (100) further comprises: -deriving (107), by the central classification unit (203), a count value (215) of a periodic partial movement of the movement based on the movement values (213) of the plurality of peripheral classification elements (201), and -Outputting (105) the count value (215) by the central sorting unit (203) to the display unit (211).
  4. 4. A method (100) according to claim 3, wherein the motion values (213) provided by the peripheral classification elements (201) respectively describe incremental motion segments of the periodic partial motion of the corresponding body parts (301) on which these peripheral classification elements (201) are arranged, wherein the deriving (107) of the count value (215) comprises: -selecting (109) a peripheral classification element (201) providing a peripheral classification value (207) that coincides with the classification value (209) of the central classification unit (203); accumulating (111) the motion values (213) provided by the peripheral classification elements (201) at different points in time and generating an accumulated motion value (213) 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 a value of one complete cycle of said periodic partial motion, then determining (113) an average motion value based on the cumulative motion value (213) of the selected peripheral classification element (201), and The count value (215) is incremented (115) by a value of "1" once the average motion value reaches or exceeds the value of one complete cycle of the periodic motion.
  5. 5. The method (100) according to any one of the preceding claims, wherein the method further comprises: If the peripheral classification value (207) provided by a peripheral classification element (201) corresponds to the classification value (209) determined by the central classification unit (203), determining (117) to a success value (217) for the corresponding peripheral classification element (201), and -Evaluating (119), by the central classification unit (203), the performance of the plurality of peripheral classification elements (201) based on the success value (217).
  6. 6. The method (100) according to any one of the preceding claims, wherein the deriving (107) further comprises: -determining (121) the movement speed of the periodic partial movement determined by the peripheral classification element (201) on the basis of the movement values (213) provided by the peripheral classification element (201) and on the basis of the time stamps (221) provided by the peripheral classification element (201) for the respective movement values (213), and-generating the speed value (219) by the central classification unit (203), wherein the time stamps (221) define the points in time at which the movement values (213) were determined by the peripheral classification element (201).
  7. 7. The method (100) according to any one of the preceding claims, wherein the determination of the classification value (209) and/or the determination of the count value (215) does not take into account the peripheral classification value (207) and/or the motion value (213) of one peripheral classification element (201) of the plurality of peripheral classification elements (201) if a time interval between the peripheral classification values (207) and/or the motion values (213) provided by the peripheral classification element (201) in time sequence exceeds a predefined boundary value.
  8. 8. The method (100) according to any one of the preceding claims, wherein the method further comprises determining (123) a pause in the movement by the central classification unit (203) if the number of peripheral classification elements (201) providing peripheral classification values (207) consistent with the classification values (209) determined by the central classification unit (203) is below a predefined number and the time interval between peripheral classification values (207) provided by at least one peripheral classification element (201) in time succession reaches or falls below a predefined time interval.
  9. 9. The method (100) according to any one of the preceding claims, wherein the method further comprises determining (125) that the motion classification system (200) is in an idle state if the number of peripheral classification elements (201) providing peripheral classification values (207) consistent with the classification values (209) determined by the central classification unit (203) is below a predefined number and the time interval between peripheral classification values (207) provided by at least one peripheral classification element (201) in time succession exceeds a predefined time interval.
  10. 10. The method (100) according to claim 9, wherein the method further comprises resetting (127) the classification value (209) and/or the count value (215) and/or the speed value (219) and/or the success value (217) by the central classification unit (203) if an idle state of the motion classification system (200) is resolved.
  11. 11. The method (100) according to any one of the preceding claims, wherein the classified movement is one of the list consisting of walking, running, hiking, mountain climbing, swimming, cross-country skiing, cycling, wherein the count value (215) relates to a periodic part of the movement in the list consisting of walking, swimming stroke, swimming leg-kicking, bicycle pedal rotation.
  12. 12. Motion classification system (200) with a central classification unit (203) and a plurality of peripheral classification elements (201), wherein the peripheral classification elements (201) each comprise at least one motion sensor (223), wherein the motion classification system (200) is configured for implementing the method (100) for classifying a motion according to any of claims 1 to 11.
  13. 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 a wearable device.
  14. 14. A computing unit (225) configured for implementing the method (100) for classifying a motion according to any one of claims 1 to 11.
  15. 15. A computer program product (400) comprising instructions which, when the program is implemented by a data processing unit, cause the data processing unit to implement the method (100) for classifying motion according to any one of claims 1 to 11.

Description

Method for classifying motions, motion classification system, computing unit and computer program product Technical Field The invention relates to a method for classifying motions and a motion classification system. The invention further relates to a computing unit and to a computer program product. Background Methods for classifying movements are known from the prior art. Disclosure of Invention The object of the present invention is to provide an improved method for classifying movements and a movement classification system. This object is achieved by the method for classifying movements and the movement classification system according to the invention. Advantageous embodiments are the subject matter of the expansion. According to one aspect, there is provided a method for classifying motions by a motion classification system with a plurality of 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 motions of the person based on sensor data, respectively, the method comprising: Receiving, by the central classification unit, classification data for the plurality of peripheral classification elements, wherein the classification data each comprises at least one peripheral classification value; classifying, by the central classification unit, the motion based on the peripheral classification values of the plurality of peripheral classification elements and generating classification values, and The classification value is output by the central classification unit to a display unit. This results in the technical advantage that an improved method for classifying movements can be provided. The method is implemented here by a motion classification system comprising a plurality of peripheral classification elements and a central classification unit. Here, a plurality of peripheral classification elements are arranged on the body part at the position in motion and are set for performing classification of the movement of the person based on the sensor values. The central classification unit is in data communication with and receives classification data from the plurality of peripheral classification elements. The classification data here comprise at least peripheral classification values, wherein each peripheral classification value demonstrates a classification of the movement performed by the corresponding peripheral classification element. The central classification unit creates a classification value for the motion based on a plurality of peripheral classification values for a plurality of peripheral classification elements. Thus, the classification value of the motion is based on a plurality of peripheral classification values for the respective peripheral classification element. This enables a higher accuracy in classifying the movements of the person. By arranging different peripheral sorting elements at different parts of the person's body, multiple movements of the individual body parts of the person can be taken into account in the sorting. Whereby the accuracy of the motion classification can be further improved. Subsequently, the respective generated classification values are output to the display unit. This enables a corresponding motion classification to be displayed to the person using the motion classification system. According to one embodiment, if one of the peripheral classification values provided by the plurality of peripheral classification elements is provided by at least a predefined number of peripheral classification elements, the classification value determined by the central classification unit corresponds to the corresponding one of the peripheral classification values. A technical advantage may thus be obtained in that a classification of movements can be made with a higher accuracy. Here, if a predefined number of peripheral classification elements provide the same peripheral classification value, the classification value generated by the central classification unit coincides with the peripheral classification value provided by these peripheral classification elements. According to one embodiment, the classification data each further comprises at least one motion value, wherein the method further comprises: Calculating, by the central classification unit, a count value of a periodic partial motion of the motion based on motion values of the plurality of peripheral classification elements, and The count value is output to the display unit by the central classification unit. A technical advantage is thereby obtained that in addition to the classification value, a count value of the periodic partial movement of the person can also be provided. Here, in addition to the peripheral classification value, a plurality of peripheral classification elements also provide a motion value. Here, the count value created by the central classification unit is based on the motio