JP2006177749A - Method and apparatus for calculating movement trajectory of periodic moving body - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a movement trajectory calculation apparatus for a periodic motion body capable of easily and accurately grasping a movement trajectory of an object that performs periodic motion such as a human body during walking.
An apparatus for calculating a movement trajectory of a periodic motion body based on a detection signal of an acceleration sensor 2 attached to the periodic motion body, wherein the detection signal is subjected to a filtering process to obtain a low frequency component. Filter processing means 10 for generating acceleration data from which acceleration has been removed, and obtaining a primary regression line for the provisional velocity data obtained by integrating the acceleration data, and correcting the provisional velocity data using the primary regression line. Thus, a speed data calculation means 12 for generating corrected speed data, and a quadratic regression curve for the temporary position data obtained by integrating the corrected speed data are obtained, and the temporary position is calculated using the secondary regression curve. Position data calculating means 14 for generating corrected position data by correcting the data is provided.
[Selection] Figure 1

Description

  The present invention relates to a method and apparatus for calculating a movement trajectory of a periodic motion body such as a movement trajectory of a body center of gravity during walking motion.

  Analysis of movement trajectories during human walking movements is expected to have various applications such as evaluation of walking ability of elderly people and grasping of medical conditions. As a method for grasping the movement trajectory of an object, a method for calculating the movement trajectory of an object by performing image processing on moving image data of the moving object is conventionally known (for example, Patent Document 1).

  However, the method for obtaining the movement trajectory based on the image processing of the imaged data has to be an environment where the object can be photographed by the camera, and the object such as the inside of the clothes cannot be visually recognized from the outside. However, there is a problem that it is easy to be subjected to restrictions on measurement conditions and measurement objects.

Patent Document 2 discloses a method of obtaining position data by second-order integrating acceleration data detected by an acceleration sensor attached to a human body, and identifying the type of motion from the position data. It has been shown that the identification accuracy can be improved by removing low frequency components.
Japanese Patent Laid-Open No. 9-35040 JP 2003-93566 A

  However, even if an attempt is made to determine the movement trajectory of the human body during walking based on the method described in Patent Document 2, the measurement error is large due to various factors such as the zero point drift of the acceleration sensor and the posture change of the subject. Therefore, it has been difficult to apply to the calculation of the movement trajectory.

  FIG. 6A shows an example of a measurement result when a 3-axis acceleration sensor is actually attached to the subject's center of gravity and a walking motion is performed. In FIG. 6A, each graph indicates detection signals in the left-right direction, the up-down direction, and the front-back direction, respectively, with respect to the subject. As shown in FIG. 6B, the detection signal is subjected to high-pass filter processing to remove low-frequency components, and then sequentially integrated to obtain the velocity and displacement. As shown in FIG. 6C and FIG. As shown in FIG. 6 (d), it diverged with the passage of time, and the movement trajectory could not be calculated.

  Therefore, an object of the present invention is to provide a method and an apparatus for calculating a movement trajectory of a periodic motion body that can easily and accurately grasp a movement trajectory of an object that performs periodic motion such as a human body during walking.

The object of the present invention is a method for calculating a movement trajectory of a periodic motion body based on a detection signal of an acceleration sensor attached to the periodic motion body, wherein the detection signal is subjected to a filtering process, A filter processing step for generating acceleration data from which low-frequency components have been removed; a primary regression line for provisional velocity data obtained by integrating the acceleration data; and the provisional velocity data using the primary regression line Is corrected to obtain a velocity data calculation step for generating corrected velocity data, and a quadratic regression curve for the temporary position data obtained by integrating the corrected velocity data is obtained. This is achieved by a method for calculating a movement trajectory of a periodic moving body, which includes a position data calculation step for generating corrected position data by correcting temporary position data.

  The object of the present invention is an apparatus for calculating a movement trajectory of a periodic motion body based on a detection signal of an acceleration sensor attached to the periodic motion body, wherein the detection signal is subjected to filter processing. Filter processing means for generating acceleration data from which low-frequency components have been removed, a primary regression line for provisional velocity data obtained by integrating the acceleration data, and using the primary regression line, the provisional regression line By correcting the speed data, a speed data calculating means for generating corrected speed data and a quadratic regression curve for the temporary position data obtained by integrating the corrected speed data are obtained, and the secondary regression curve is used. This is achieved by a moving trajectory calculation apparatus for a periodic moving body comprising position data calculation means for generating corrected position data by correcting the temporary position data.

  ADVANTAGE OF THE INVENTION According to this invention, the movement locus | trajectory of the target object which performs a periodic motion can be grasped | ascertained easily and accurately.

  Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram showing a schematic configuration of a moving trajectory calculation apparatus for a periodic moving body according to an embodiment of the present invention. As shown in FIG. 1, the movement trajectory calculation device 1 is connected to an acceleration sensor 2, and these are integrated so that they can be attached to a belt of trousers or the like. The acceleration sensor 2 is a triaxial acceleration sensor, is fixed to the periodic motion body, detects accelerations in three orthogonal directions, and outputs a voltage corresponding to the acceleration.

  The movement trajectory calculation apparatus 1 includes a filter processing means 10, a speed data calculation means 12, a position data calculation means 14, and a movement trajectory generation means 16, and can be configured by a portable computer, for example. .

The filter processing means 10 is a high-pass filter, and removes a low-frequency component that is equal to or lower than the cutoff frequency from the frequency components included in the detection signal input from the acceleration sensor 2. Specifically, conversion data a _i obtained by multiplying N sampling data of the detection signal d _i (i = 1, 2,..., N) by a voltage-acceleration conversion coefficient α is obtained, and this conversion data is obtained. Acceleration data A _i is generated by performing high-pass filtering on a _i . When measuring the movement trajectory of a human body during walking, the cut-off frequency can exemplify the reciprocal of a value obtained by doubling the walking cycle (seconds), and is usually about 0.1 to 1 Hz.

The acceleration data A _i output from the filter processing means 10 is input to the speed data calculation means 12. Velocity data calculating means 12 performs integration by sequentially adding the acceleration data A _i, obtain the velocity data v _i of the temporary. That is, when the sampling time interval is Δt, v _i = A _i · Δt + v _i−1 . However, it is assumed that the initial speed v ₀ = 0.

The acceleration data A _i output through the filtering means 10, which includes an error factor such as the direct current component can not be removed in the filter processing unit 10, when determining the velocity data v _i of the temporary, the error Are integrated and accumulated. The present inventors have found that the accumulated error term is included in the tentative speed data v _i is found that can be grasped as a linear value that can be approximated by a linear equation.

Therefore, the velocity data calculating means 12 calculates a primary regression line of the temporary speed data v _i by the least squares method, by correcting the velocity data v _i provisional using the primary regression line, corrected velocity data V _i Is calculated. That is, the coefficients c and d are obtained by the least square method assuming a linear regression equation of v _i = c · t + d with respect to the temporary velocity data v _i . The error term L _i included in v _i can be calculated from L _i = c · t + d using the obtained coefficients c and d, and the corrected speed data V _i is V _i = v _i −L _i. Can be calculated as The corrected speed data V _i is obtained as a relative speed with respect to the average speed of the periodic motion body (for example, about 3 km / h forward in the case of a human body walking motion).

The corrected speed data V _i obtained in this way is input to the position data calculation means 14. Position data calculating means 14 performs integration by sequentially adding the correction velocity data V _i, obtaining the position data x _i of the temporary. That is, assuming that the sampling time interval is Δt, x _i = V _i · Δt + x _i−1 . However, it is assumed that the initial position x ₀ = 0.
Alternatively, in consideration of the acceleration data A _i , the temporary position data x _i may be calculated from x _i = ½ · A _i · Δt 2 ⁺ V _i · Δt + x _i−1 . The integration of the acceleration data A _i and the corrected velocity data V _i can be performed by other known methods using, for example, the Simpson formula, in addition to the sequential addition.

The temporary position data x _i obtained in this way includes errors such as a direct current component that cannot be removed by the filter processing means 10 as described above. Therefore, the position data calculation means 14 obtains a quadratic regression curve of the temporary position data x _i by the least square method, and corrects the temporary position data x _i using this quadratic regression curve, thereby correcting the corrected position data. X _i is calculated. That is, for the temporary position data x _i , assuming a quadratic regression equation x _i = e · t ² + f · t + g, the coefficients e, f, and g are obtained by the least square method. The error term C _i included in x _i can be calculated from C _i = e · t ² + f · t + g using the obtained coefficients e, f, and g, and the corrected position data X _i is expressed as X _i = X _i -C _i can be calculated. The corrected position data X _i is obtained as a relative position with respect to the reference position when it is assumed that the periodic moving body has moved at an average speed.

The position data X _i obtained in this way is input to the movement locus generating means 16. The movement trajectory generation means 16 generates a movement trajectory from the time-series change of the position data X _{i in} each axis direction and outputs it to a monitor or the like.

The acceleration trajectory 2 was grasped, and the movement trajectory when actually moving along the side of the rectangular plate fixed on the desk was obtained. FIGS. 2A to 2C show acceleration data A _i , correction speed data V _i, and correction position data X _i , respectively. As shown in FIGS. 2B and 2C, the corrected velocity data V _i and the corrected position data X _i are included in the acceleration data A _i by the correction processing in the velocity data calculating unit 12 and the position data calculating unit 14. It can be seen that the measurement error is eliminated. As a result, the movement trajectory generated by the movement trajectory generation means 16 has a result that the movement trajectory of the periodic motion is accurately reproduced as shown in FIG. Fig. 3 shows the same test performed for a triangle or pentagon in addition to a square, and the calculated movement trajectory (dark line) was compared with the movement trajectory (light line) taken with a PSD camera. It is. In any case, a calculation result close to the actual movement trajectory is obtained. Since the acceleration sensor 2 is moved along a plane, the measurement results in only two directions are displayed in FIGS.

FIG. 4 shows the result of obtaining the movement locus of the body center of gravity during walking. In this experiment, the movement trajectory calculation device 1 is integrated with the acceleration sensor 2 and accommodated in the casing, and is attached to the belt portion on the back side of the subject so that the acceleration sensor 2 is installed at the body center of gravity position. Fixed. 4A to 4C show acceleration data A _i , correction speed data V _i, and correction position data X _i , respectively. The movement trajectory at this time is shown in FIG. 5A to 5C
Are the state of the subject as seen from the front, the state as seen from the right side, and the state as seen from above. It can be seen that in any state, the periodic motion is accurately reproduced.

  Thus, according to the movement trajectory calculation apparatus of the present invention, it is possible to accurately calculate the movement trajectory of the periodic moving body, which has been difficult in the past. In addition, by making this movement trajectory calculation device portable, it is less likely to be subject to measurement condition restrictions, and it is also easy to calculate movement trajectories of periodic movements that are impossible with image processing, such as the movement of parts inside clothing on the human body. Can be applied to a wide range of fields.

It is a block diagram which shows schematic structure of the movement track | orbit calculation apparatus of the periodic exercise body which concerns on one Embodiment of this invention. It is a figure which shows an example of the process which calculated the movement locus | trajectory of the periodic motion body with the said movement locus | trajectory calculation apparatus. It is a figure which shows the other example of the process in which the movement locus | trajectory of the periodic motion body was computed by the said movement locus | trajectory calculation apparatus. It is a figure which shows the further another example of the process which calculated the movement locus | trajectory of the periodic motion body with the said movement locus | trajectory calculation apparatus. It is a figure which shows the movement locus | trajectory of the periodic motion body obtained from the calculation process of the movement locus | trajectory of the periodic motion body shown in FIG. It is a figure which shows the process which calculated the movement locus | trajectory of the periodic motion body by the conventional method.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Movement locus calculation apparatus 2 Acceleration sensor 10 Filter processing means 12 Speed data calculation means 14 Position data calculation means 16 Movement locus generation means

Claims (4)

A method of calculating a movement trajectory of the periodic motion body based on a detection signal of an acceleration sensor attached to the periodic motion body,
A filter processing step of performing a filtering process on the detection signal to generate acceleration data from which low-frequency components are removed;
A velocity data calculation step for obtaining corrected velocity data by obtaining a primary regression line for temporary velocity data obtained by integrating the acceleration data, and correcting the temporary velocity data using the primary regression line; ,
Position data for generating corrected position data by obtaining a quadratic regression curve for the temporary position data obtained by integrating the corrected velocity data and correcting the temporary position data using the quadratic regression curve A moving trajectory calculation method for a periodic moving body comprising a calculating step. The method according to claim 1, wherein the periodic motion body is a human body. The method according to claim 1 or 2, wherein the detection signal of the acceleration sensor includes detection components in three orthogonal directions orthogonal to each other. An apparatus for calculating a movement trajectory of the periodic motion body based on a detection signal of an acceleration sensor attached to the periodic motion body,
Filter processing means for applying a filtering process to the detection signal to generate acceleration data from which low-frequency components have been removed;
A velocity data calculating means for obtaining corrected velocity data by obtaining a linear regression line for the temporary velocity data obtained by integrating the acceleration data and correcting the temporary velocity data using the linear regression line; ,
Position data for generating corrected position data by obtaining a quadratic regression curve for the temporary position data obtained by integrating the corrected velocity data and correcting the temporary position data using the quadratic regression curve A moving trajectory calculation apparatus for a periodic moving body, comprising: a calculating means.

Images