CN105651281A - Space positioning algorithm on basis of multi-direction human movement modes - Google Patents

Abstract

The invention discloses a space positioning algorithm on the basis of multi-direction human movement modes, and belongs to the field of positioning and navigation technologies. The space positioning algorithm includes firstly, acquiring original information of human movement by the aid of MEMS (micro-electromechanical systems) sensors; secondly, estimating step lengths, sampling heading, discriminating the movement modes and solving heights; thirdly, solving coordinates of human positions by the aid of coordinate calculation formulas to position pedestrians. Low-cost MEMS inertia measurement units integrated in portable handheld terminals are used as hardware platforms for implementing the space positioning algorithm. The space positioning algorithm has the advantages of applicability to actual multi-direction human movement modes and high application value.

Description

Space orientation algorithm based on the multi-faceted motor pattern of human body

Technical field

The present invention relates to MEMS inertial sensor navigation field, particularly to the application in pedestrian's reckoning and inertial navigation field.

Background technology

Technique study about human body location has a lot, but part localization method exists certain defect. satellite-signal cannot be normally received as being positioned at the closing spaces such as basement based on the assistant GPS of mobile communications network (A-GPS), WLAN (WLAN) signal is highly susceptible to other signal disturbing, positioning precision is relatively low, radio-frequency (RF) tag (RFID) location is subject to environmental disturbances, infrared confirming orientation technology power consumption more greatly and usually can be subject to the obstruct of indoor wall or object, practicality is relatively low, the cost that super wideband wireless (UWB) positions this kind equipment is higher, it is unfavorable for popularizing, during ultrasonic locating, ultrasound wave is decayed substantially in transmitting procedure, thus affecting its location effective range.

Inertia system do not rely on external information, also not to outside emittance, be a kind of autonomous navigation system, therefore good concealment and not by the impact of outside electromagnetic interference, entirely can work in aerial, earth surface or even under water temporally. Inertia system is provided that position, speed, course and attitude angle information, and produced navigation information seriality is good and data updating rate is high, and short-term accuracy is high, good stability.

Currently utilize inertial sensor location technology, the single advance walking movement pattern that the resolving of human body coordinate is all only considered, do not meet the multi-faceted motion that human body is actual, therefore, need the coordinate that reliable algorithm resolves human body that human body is positioned the actual multi-faceted motion positions of human body.

Summary of the invention

In view of this, technical problem solved by the invention is to provide a kind of space orientation algorithm based on the multi-faceted motor pattern of human body. Expand MEMS inertial sensor independent navigation application.

To achieve these goals, the technical solution adopted in the present invention is:

Under various different motion pattern provided by the invention, the hardware platform of Autonomous Navigation Algorithm is MEMS Inertial Measurement Unit integrated in handheld terminal.

During various motor patterns provided by the invention switching, the real-time update condition of autonomous location algorithm, modulus value and three axis accelerometer single shaft data phase situation of change is counted including 3-axis acceleration;

Described 3-axis acceleration counts modulus value and three axis accelerometer single shaft data phase, is used for judging motor pattern switching type; Utilize the coordinate that multi-faceted locomotion pattern space location algorithm coordinate prediction equation calculates position of human body that pedestrian is positioned.

Compared with prior art, it is an advantage of the current invention that: user is not subject to the restriction of motor pattern when location, being free to multi-faceted motion, the motor pattern of the walking that is not limited to advance, for playing great impetus based on MEMS inertial sensor navigation in human body real life.

Accompanying drawing explanation

In order to make the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, the present invention is described in further detail, wherein:

Fig. 1 is multi-faceted locomotion pattern space location algorithm flow process.

Fig. 2 is space orient models.

Fig. 3 is forward travel, setback model.

Fig. 4 is left transverse movement, right transverse movement model.

Detailed description of the invention

Below in conjunction with accompanying drawing, the present invention is further described.

The flow process of this algorithm is as shown in Figure 1.

Use three axis accelerometer modulus value and three axis accelerometer single shaft data phase, it is judged that human motion pattern and gait. The course of pedestrian is tried to achieve by EKF algorithm fusion three-axis gyroscope, three axis accelerometer and three axle magnetometer data. Barometer is utilized to resolve human height. Space orient models is Fig. 2 such as.

It it is hereafter the embodiment of multi-faceted locomotion pattern space location algorithm coordinate prediction equation.

When human motion pattern is only advanced walking and when retreating walking, two dimensional motion model is Fig. 3 such as, and pedestrian advances from A point and walks to B point, then walks C point from B point backing. Carry out coordinate by formula (1) and position resolves.

$\left\{\begin{array}{c}E\left(k\right)=E(k-1)+S_k{L}_k{\mathrm{cos\θ}}_k\\ N\left(k\right)=N(k-1)+S_k{L}_k{\mathrm{sin\θ}}_k\\ H\left(k\right)=H(k-1)+V_k\end{array}\right.---\left(1\right)$

In above formula, L_kFor the step-length of each step, �� is the course of each step, V_kFor the height that the k moment increases. S_kFor motor pattern mark, S when currently walking_k=1, retreat S during walking_k=-1, other motor patterns S_k=0. E (k) represents that east orientation, N (k) represent north orientation, and H (k) represents height.

When human motion pattern only has left horizontal walking and right transverse direction to walk, two dimensional motion model is Fig. 4 such as, and pedestrian is from the right laterally walking of A point to B point, or pedestrian is from the left horizontal walking of A point to C point. Carry out coordinate by formula (2) and position resolves.

In above formula, L_kFor the step-length of each step, �� is the course of each step, V_kFor the height that the k moment increases. U_kFor motor pattern mark, U during left laterally walking_k=-1, U during right laterally walking_k=1, other motor patterns U_k=0.

When human motion be that original place is static, original place rotates, walking of advancing, retreat the multi-faceted hybrid motion pattern of walking, left laterally walking and right laterally walking time, carry out coordinate by formula (3) and position resolve.

In above formula, L_kFor the step-length of each step, �� is the course of each step, V_kFor the height that the k moment increases. S_kFor forward and backward walking movement mode flags, U_kFor left and right horizontal walking movement mode flags. Time static, S_k=0, U_k=0; During advance walking, S_k=1, U_k=0; When retreating walking, S_k=-1, U_k=0; S during left laterally walking_k=0, U_k=-1; S during right laterally walking_k=0, U_k=1.

Claims (2)

1. based on a space orientation algorithm for the multi-faceted motor pattern of human body, it is characterized in that, including 5 azimuth motion pattern discrimination algorithms of human body actual motion, mode handover procedure space orientation algorithm etc.;

Described multi-faceted motor pattern refers to, 5 azimuth motion being likely to occur in human body actual motion, including static, forward travel, setback, left transverse movement and right transverse movement;

Described static finger, stands in original place;

Described forward travel refers to, forward is careful, hurries up, jogs and is hurried up;

Described setback refers to, reversely moves back slowly, rewind;

Described left transverse movement refers to, left transverse direction strides walking;

Described right transverse movement refers to, right transverse direction strides walking;

Described mode handover procedure space orientation algorithm includes the location of following motor pattern switching mode:

Static, forward travel, setback, left transverse movement, right transverse movement.

2., based on a space orientation algorithm for the multi-faceted motor pattern of claim 1 human body, it is characterized in that comprising the steps: that three-axis gyroscope, three axis accelerometer, three axle magnetometers and barometer are carried out data acquisition and calibrate for error and pretreatment by (1); (2) attitude angle of pedestrian is tried to achieve by EKF data fusion, obtain the course of each step of pedestrian, utilize acceleration information that the gait of pedestrian is detected, and on the basis of gait detection, estimate the step-length of each step of pedestrian, use three axis accelerometer to distinguish static, forward travel, setback, left transverse movement simultaneously, right transverse movement 5 type games motor pattern, barometer resolves human height; (3) by the course of pedestrian, motor pattern, step-length and the position coordinates highly being obtained current line people by the calculating of coordinate prediction equation.