System and method for observing the swimming activity of a person

• US 9,643,068 B2
• Filed: 03/29/2010
• Issued: 05/09/2017
• Est. Priority Date: 03/31/2009
• Status: Active Grant

• Current Assignee: Commissariat a L'Energie Atomique (AES Corporation), Movea SA (TDK Corporation)
• Original Assignee: Commissariat a L'Energie Atomique (AES Corporation), Movea SA (TDK Corporation)
• Inventors: Jallon, Pierre
• Primary Examiner(s): Breene, John
• Assistant Examiner(s): MORGAN, JEFFREY CHAD
• Application Number:
  US13/260,510
• Publication Number:
  US 20120072165A1
• Time in Patent Office: 2,598 Days
• Field of Search: 702141
• CPC Class Codes: A63B 2024/0056 for statistical or strategic analysis
• CPC Class Codes: A63B 2024/0071 Distinction between different activities, movements, or kind of sports performed
• CPC Class Codes: A63B 2071/0647 Visualisation of executed movements
• CPC Class Codes: A63B 2220/40 Acceleration
• CPC Class Codes: A63B 2220/836 Sensors arranged on the body of the user
• CPC Class Codes: A63B 2244/20 Swimming
• CPC Class Codes: A63B 69/00 Training appliances or apparatus for special sports training of parachutists B64D23/00
• CPC Class Codes: G06F 2218/00 Aspects of pattern recognition specially adapted for signal processing
• CPC Class Codes: G06V 40/28 Recognition of hand or arm movements, e.g. recognition of deaf sign language static hand signs G06V40/113

Abstract

A system for observing a swimming activity of a person includes a waterproof housing (BET) having a motion sensor (MS), and is furnished with fixing means (BEL) for securely fastening the housing (BET) to a part of the body of a user. The system has analysis means (AN) for analyzing the signals transmitted by the motion sensor (MS) to at least one measurement axis and which are adapted for determining the type of swimming of the user as a function of time by using a hidden Markov model with N states corresponding respectively to N types of swimming.

Claims

1. A system for observing a swimming activity of a person, comprising a waterproof housing comprising a motion sensor, and furnished with fixing means for securely fastening the housing to a part of the body of a user, characterized in that the system further comprises analysis means for analyzing signals provided by the motion sensor for at least one measurement axis of the motion sensor and which is configured to determine a type of swimming of the user as a function of time by using a hidden Markov model with N states corresponding respectively to N types of swimming, the model being characterized by a column vector μ

and a diagonal matrix Σ

, both of which are representative of a state of the hidden Markov model corresponding to said swimming and having dimensions equal to the number of measurement axes of the motion sensor.

2. The system as claimed in claim 1, wherein said motion sensor comprises at least one of an accelerometer, a gyrometer and a magnetometer.

3. The system as claimed in claim 1, further comprising a low-pass filter having a cutoff frequency between 0.5 Hz and 5 Hz.

4. The system as claimed in claim 3, wherein a probability density p_x(x(n)) of correspondence between the signals delivered by the motion sensor and a state of the hidden Markov model representing a type of swimming is defined by the following expression:

5. The system as claimed in claim 4, wherein said motion sensor comprises a triaxial accelerometer and the waterproof housing is fastened to a wrist of the user, the three axes of said accelerometer forming a right-handed trihedron, such that:

a direction of the first axis is a longitudinal axis of the forearm of the wrist to which the housing is fastened and is oriented toward the elbow; and

a third axis is vertically oriented downwards when the forearm of the wrist to which the housing is fastened is in a horizontal plane, the palm of the hand of the wrist to which the housing is fastened being directed downwards, and said housing being disposed on the outer face of the wrist.

6. The system as claimed in claim 5, wherein for breaststroke the three components μ

₁, μ

₂, μ

₃ of the column vector μ

are such that μ

₁ε

[−

0.45;

−

0.20], μ

₂ε

[−

0.1;

0.5], and μ

₃ε

[−

0.8;

0.45], and the three diagonal components Σ

₁, Σ

₂, Σ

₃ of the diagonal matrix Σ

are such that Σ

₁ε

[0.1,0.18], Σ

₂ε

[0.2;

0.6], and Σ

₃ε

[0.03;

0.2].

7. The system as claimed in claim 5, wherein for crawl the three components μ

₁, μ

₂, μ

₃ of the column vector μ

are such that μ

₁ε

[−

0.7;

−

0.8], μ

₂ε

[−

0.25;

−

0.45], and μ

₃ε

[−

0.4;

0.2], and the three diagonal components Σ

₁, Σ

₂, Σ

₃ of the diagonal matrix Σ

are such that Σ

₁ε

[0.2;

0.3], Σ

₂ε

[0.1;

0.3], and Σ

₃ε

[0.07;

0.5].

8. The system as claimed in claim 5, wherein for butterfly the three components μ

₁, μ

₂, μ

₃ of the column vector μ

are such that μ

₁ε

[−

0.8;

0.1], μ

₂ε

[−

0.45;

0.5], and μ

₃ε

[−

0.2;

0.4], and the three diagonal components Σ

₁, Σ

₂, Σ

₃ of the diagonal matrix E are such that Σ

₁ε

[0.2;

0.4], Σ

₂ε

[0.1;

0.5], and Σ

₃ε

[0.2;

0.8].

9. The system as claimed in claim 5, wherein for backstroke the three components μ

₁, μ

₂, μ

₃ of the column vector μ

are such that μ

₁ε

[−

0.2;

0.1], μ

₂ε

[0.3;

0.7], and μ

₃ε

[−

0.05;

0.4], and the three diagonal components Σ

₁, Σ

₂, Σ

₃ of the diagonal matrix Σ

are such that Σ

₁ε

[0.2;

0.4], Σ

₂ε

[0.1;

0.5], and Σ

₃ε

[0.2;

0.8].

10. The system as claimed in claim 5, wherein probabilities P(state_i, state_j) of said hidden Markov model of switching between two states i, j representing respectively a type of swimming are such that:

P(state_i,state_j)ε

[0.8;

0.9999], when i is different from j; and

P(state_i,state_j)ε

[0.0001;

0.2], when i is equal to j.

11. The system as claimed in claim 5, wherein said analysis means is internal or external to the housing, and the triaxial accelerometer comprises wired or wireless transmission means configured to transmit measurements of the system to said analysis means.

12. The system as claimed in claim 5, further comprising display means fixed to the housing or a remote display means.

13. The system as claimed in claim 1, wherein said fixing means is configured to fasten the housing to the wrist, to the ankle, to the neck or to the head of the user.

14. The system as claimed in claim 1, wherein the analysis means is further configured to determine a type of swimming of the user from among a set of at least two swimming strokes selected from at least breaststroke, crawl, butterfly, and backstroke.

15. The system as claimed in claim 1, wherein values of the column vector μ

and the diagonal matrix Σ

have ranges of values which characterize one of the N types of swimming.

16. A method for observing a swimming activity of a person on the basis of measurements obtained from a motion sensor fixed in a waterproof manner to a part of the body of a user, characterized in that signals obtained from the motion sensor for at least one measurement axis of the motion sensor are analyzed to determine a type of swimming of the user as a function of time by using a hidden Markov model with N states corresponding respectively to N swimming types, the model being characterized by a column vector μ

and a diagonal matrix Σ

, both of which are representative of a state of the hidden Markov model corresponding to said swimming and having dimensions equal to the number of measurement axes of the motion sensor.