Description

PEDESTRIAN NAVIGATION METHOD AND APPARATUS FOR USING GEOGRAPHIC INFORMATION SYSTEM

Technical Field

[1] This invention concerns a self-contained pedestrian navigation system.

[2] GPS(Global Positioning System) provides the exact 3D position and time everywhere in the world, but we sometimes can not use this information in valley or inside a building because the GPS signal is very weak.

[3] To solve these problems, we use the navigation system combined with INS or other ground based radio navigation such as VOR, ILS for vehicles such as aircraft. For pedestrians we can use personal navigation system using mobile phone by estimating the distance between the mobile phone and a relay station by measuring the time of arrival, but it is not operable in mountains or rural regions where relay stations were not sufficiently established.

[4] Therefore the purpose of this invention is to provide the exact 3D position of a pedestrian in regions where GPS signal is not available, without relying on an infrastructure such as mobile phone relay station, and also to provide a method to determine the exact position by using GIS (Geographic Information System) based on the trajectory obtained by this invention. Consequently, many LBS application such as navigation for blind or climb route guidance in valley will be feasible.

Background Art

[5] The very simple pedestrian navigation (so called dead reckoning) is possible to estimate the total displacement by multiplying the number of strides of a pedometer and the average stride length, or to estimate the net directional displacement by using pedometer combined with digital magnetic compass and gyroscope. More sophisticated pedestrian navigation reflects the variation of stride length according to the slope by estimating the gradient of a slope by using the change of barometric altitude, or calculates the length of each stride by integrating twice the directional acceleration of accelerometer mounted on the user's leg or foot. But the error of the system which calculates stride length with foot's acceleration is too big, even though using very expensive inertial sensor. So, the stride length can be estimated by applying a scaling multiplier in consideration of pedestrian's walking characteristics by distinguishing the pattern of the acceleration wave of an accelerometer mounted on the user's waist or torso. Or the stride length is predicted by the relationship between the acceleration pattern of the user's body and the distance traveled by the user, which was measured during GPS receiver was operable.

[6] However, there exists a limit in these methods to estimate or to predict user's stride length. The estimated stride length may be correct when the user travels in flat land without any obstacle, but there are big discrepancies when he can not walk with his own manner of walking in a steep path with obstacles or in a congested pathway of subway station where he must walk in avoiding collision with other pedestrian.

[7] Although this invention is a kind of self contained navigation system, a method to measure directly the displacement of each step, which is not using empirical data such as dead reckoning, is selected.

[8] As it is possible to produce an economic GIS for pedestrian based on the records of trajectory of the user's position by measuring user's stride length directly, it is possible to apply map-matching technique in pedestrian navigation.

[9] The map-matching technique is a method to estimate the user's position, which is expected more accurate, within the map produced with a very exact sensor, with the user's position data provided by less accurate sensor. Therefore the map-matching technique is widely used in CNS (Car Navigation System). Disclosure of Invention Technical Problem

[10] The primary goal of this invention is to determine the pedestrian's location by measuring user's horizontal and vertical displacement at each step, based upon the characteristics of human locomotion (in which, at all times at least one foot is in contact with the surface and during a brief phase both feet are in contact with the surface in walking, never both feet are in contact with the surface and during a very brief phase one foot is in contact in running), to obtain the same degree of accuracy in positioning as GPS receiver where GPS signal is not available, without relying upon any infrastructure. Technical Solution

[11] This invention is fundamentally composed of the basic sensor module (which is composed of a digital clock, a 3 axis accelerometer, a digital magnetic compass, a processing unit and a memory unit), and the goniometers to measure the joint angle changes of human legs for calculating user's stride length, or the sensor to measure the distance from a reference point of human body to user's feet.

[12] Human locomotion is very complicated, but many experimental results of gait science reveal us as followed: As a user walks faster, both the step size and the frequency of steps increase. The frequency of steps and the step size are very closely correlated and we prefer running to walking in case that the walking speed exceeds a certain point (maybe 2m/s). The mechanics of walking completely changes, as the slope becomes greater than 7% and so on.

[13] The methodology of this invention is fundamentally based upon the techniques to calculate the distance vector by measuring the angles between horizontal plane and the upper leg, the upper leg and the lower leg, the lower leg and the feet, or to measure the distance vector from a certain reference point placed near user's torso to the user's foot with a sensor using radio wave and acoustic wave.

[14] Human body repeats the movements of acceleration and deceleration along forward-backward direction and up-down in a cycle of walking or running, but in the view of long period motion over 5 cycles the trunk's motion can be regarded as constant speed motion.

[15] Under the assumption that human's trunk is vertical to the horizon to maintain the balance; we can regard the plane of waist belt as horizontal plane. Therefore we can estimate the gradient of limbs to the horizon by measuring the angles between the plane of waist belt, the upper leg, the lower leg, and the foot.

[16] Or, it is possible to estimate the gradient of upper leg to the horizon by measuring the angle between the horizontal plane of earth's magnetic field and the upper leg with the goniometer combined with the digital magnetic compass.

[17] The acceleration measured by the accelerometer mounted on the body is composed of the horizontal acceleration which repeats the acceleration and deceleration during walking, and the vertical acceleration created by the gravity.

[18] Digital magnetic compass can get the azimuth of walking by using the fact that the direction of resultant force detected by accelerometer is vertical for long period, and the change of resultant force during the short period is the horizontal acceleration and deceleration in walking. Also we can count the number of steps by counting the horizontal change of acceleration and deceleration as one period which is equal to one step.

[19] The difference of the angle between the horizontal magnetic field and the long period resultant force can be regarded as the slope of the ground where the walker stays. Or, it is possible to define the vertical plane of the gravity detected when the walker took a stand, as the horizontal plane.

[20] For a precise pedestrian navigation, the tilt angles of three elements of walking

(upper leg, lower leg, and foot) of the limbs to the level are necessary.

[21] In practice and for the convenience of put on/take off, it is enough to obtain the accuracy for pedestrian navigation by measuring the angles of upper leg and lower leg with the concept of effective leg length.

[22] The modeling of walking based on the parameter I (length of upper leg) and I

(effective length of lower leg) is general in gait analysis to describe the human walking in numerical formula with the tilt angles of upper and lower leg. The I will be changed into different value by the mode such as level walking, a gentle slope walking,

a steep slope walking(upward and downward walking). [23] Human locomotion mode can be classified by slow walking, normal walking, speedy walking, running, level walking, a gentle slope walking, and upward and downward walking at steep slope. Prescribed effective length of leg(f ) is applied to each locomotion mode. [24] Locomotion mode can be distinguished by the horizontal and vertical wave of three axis accelerometer with the electric clock. It is possible to select the prescribed effective length of lower leg(f ) and measure the tilt angle of upper leg and lower leg at the point that the foot contacts the ground. [25] In level ground as described in Fig. 2, the horizontal displacement H can be calculated by using the effective leg length I , I and θj,θ ₂ (measured tilt angles of upper leg and lower leg) [26]

H = 2 ( Z ₁ + I ₂ ) sin( ^θ z „ ^θi )

2

[27] In gentle slope ground as described in Fig. 3, the horizontal displacement H and vertical displacement V can be calculated from the geometrical relation. [28]

r = 2(/ _{1 +} / ₂ ) sin(- ) sin(- - |)

[29] At steep slope ground as described in Fig. 4, the horizontal displacement H and vertical displacement V can be calculated as following. [30]

/ ₄ = /( I ₁ + / ₂ cos6> _? ) ² + / ₂ ² siii ² 6> ₃

-J 1? + /, ² + 2 / 1, L 2 cos&

θ. = 6> - sin ^[ -í-

[31]

= ( Z ₁ + I ₂ +/ ₄ ) sin(^L^)

r = ( Z, + Z ₇ ) cos(^L^L)- / cos(^L^L)

Z. Z.

[32] Not only at level ground but also at steep slope, the east- west and south-north directional horizontal displacement can be calculated by multiplying the sin and cos value of azimuth angle from the digital magnetic compass, by the horizontal displacement H.

[33] By accumulating the east- west and south-north directional horizontal displacement for each step, the three dimensional trajectory of a pedestrian can be obtained. So, 3D information of new climb-path can be generated, which is not described in the existing map.

[34] Though the method of measuring the tilt angles of limbs to calculate the displacement has the advantage of not affecting the outside environment, it has a drawback of the error caused by the deviation of the mounting place of goniometers to the body or applying the electromyography sensors.

[35] Therefore, we can consider another approach.

[36] It can be considered that the sole of the foot is in contact with the ground, when the measured pressure of a pressure gauge attached to the sole is above a threshold value, or when the measured acceleration of an accelerometer attached to the instep is above a threshold value. When the sole of the foot is in contact with the ground, the transmitter of radio wave and ultrasound generates a pulse including the identification code for both left foot and right foot.

[37] Three receivers of radio wave and ultrasonic wave are mounted separately on the waist belt of the pedestrian. Two transmitters are mounted on the left foot and the right foot separately.

[38] A distance vector of the left foot or the right foot from a reference point of pedestrian's body can be obtained by three distances between the receiver and the transmitter, calculated by the time difference of arrival of a radio wave pulse and an ultrasonic pulse

[39] It will cost much to reduce the size of ultrasonic transmitter even though the ultrasound is not audible. To reduce the size of the device, it is possible to take an advantage of very small sized and high performance microp*hone. When the sole of the foot contacts with the ground, the different sounds for left foot and right foot can

be generated by ringing a bell or castanets.

[40] Under the assumption that the coordinate reference point is in the ellipsoidal plane of the belt in figure 5 and the positions of each receivers are (x ,y ,0),(x ,y ,0),(x ,y ,0), the positions of left foot (x ,y ,z ) and right foot (x ,y ,z ) can be determined left left left right right right as they satisfy the following relationship, where (δt ,δt ,δt ) are the time difference of arrivals for radio wave and ultrasonic wave, and (d ,d ,d ) are the distances of the foot measured by the receivers mounted at the points as (x ,y ,0),(x ,y ,0),(x ,y ,0). [41]

where / = 1, 2, 3, k = left, right

[42] It is fundamental principle that three receivers are necessary to determine three dimensional positions of left foot and right foot. [43] But two receivers are enough to determine the positions (x ,0,z ), (x ,0,z ), right right right right under the assumption that y , y = 0, as the human body maintains the balance. left right

[44] During walking, at least one foot always contacts with the ground, but during running both foot can be on the air in a running cycle, so it is necessary to estimate the moving distance during flying.

[45] In fig. 6, the model of human running is described by spring (leg) with mass. When the left foot contacts with the ground, the kinetic energy of motion changes into the elastic energy as the spring shrinks. At the moment the left foot jumps to the sky, the preserved elastic energy changes into the kinetic energy again.

[46] Under the assumption that a period of running step is composed of t (time of contact) and t f (time of flight), it is possible to measuret C andt f , and also to measure I

(length of effective leg vector) at the moment the left foot contacts with the ground, and the tilt angle θ of the effective leg vector with the level. It is also possible to measure the tilt angle θ of the effective leg vector with the level, at the moment the foot takes off the ground.

[47] In steady running, it is known that θ = π- θ because of the symmetry.

[48] So, the horizontal displacement H in a cycle of running can be obtained as follow, with i, θ , and T.

[49] T= t + t

[50] where t = 2 v /g , g = gravity

[51] t _c = £( 2 Q ₁ - ^ V v _x

[52] v /v = tan( θ - π/2 + δ ^* ) => tan( θ - π/2 )

[53] T= t'+t = 2 v /g +(2θ -π)/v = 2 v /g tan(θ - π/2 ) +£(2θ -π)l v c f z l x x 1 I x

[54]

[55] With the GIS information constituted by the three dimensional trajectory obtained by the apparatus described above, the walker can figure out the slope of the ground without measuring the tilt angle of upper and lower leg.

[56] In this case that a pedestrian walks the same way he had passed and he has the GIS information of the trajectory he had passed, it is possible to estimate his position with the basic sensor module (composed of a digital clock, a 3 axis accelerometer, a digital magnetic compass, a processing unit and a memory unit) by using Map-matching methods (Car Navigation System adopts generally) without putting on goniometers. It becomes more convenient for carrying.

[57] It may be not sufficient for a blind person in the center of a metropolis where many obstacles exist, to use these methods even though it can provide good position accuracy for normal persons.

[58] The self-contained navigation which measures or estimates the pedestrian's displacement with the sensors such as a digital magnetic compass, an accelerator, and a goniometer, without relying upon an infrastructure, the intrinsic errors of the sensor itself or caused by the deviation of mounting place should become great by accumulation.

[59] Therefore, pedestrian coordinate-correction points can be established to improve this problem at the areas unavailable of GPS signal.

[60] The position accuracy can be recovered by substituting the position of the coordinate-correction points which contains three dimensional WGS- 84 coordinate obtained by surveying, for the calculated position of the self-contained navigation device, when a blind pedestrian connects his navigation device to the coordinate- correction point.

Advantageous Effects

[61] The availability of using LBS(Location Based System) based on WGS 84 on which the GPS coordinate is based, become seamless in the world, because it is possible to acquire the exact 3 D position with this invention even in the area where the GPS signal is not available such as in deep valley or inside a building. Brief Description of the Drawings

[62] Figure 1 is to explain the gait mechanism of humankind,

[63] figure 1-1 is a human body's conceptual diagram widely used in Gait-behavior

Analysis,

[64] figure 1-2 is a conceptual diagram of the gait in a view of three components of legs.

[65] figure 1-3 is a conceptual diagram of the gait in a view of two components of legs.

[66] Figure 2 is a calculation diagram of horizontal displacement as a gait model of a flat land.

[67] Figure 3 is a calculation diagram of both horizontal displacement and vertical displacement as a gait model of a gentle slope.

[68] Figure 4 is a calculation diagram of both horizontal displacement and vertical displacement as a gait model of a steep slope.

[69] Figure 5 is calculation diagram to determine distance vector between left foot and right foot using radio wave and ultrasonic wave, in which transmitters are attached at both feet and three receivers are mounted on the belt.

[70] Figure 6 is calculation diagram to determine horizontal displacement H when a person runs on the road. Best Mode for Carrying Out the Invention

[71] The solution enhancing the positional accuracy and the convenience of put on and take off the apparatus with this invention is as follows. A power source and an apparatus combined with an operation processor, a 3-axis accelerometer, a digital magnetic compass are installed in a backpack of the climber, and three microphones are fixed in the near the vertical plane of the backpack with a proper space.

[72] Two transmitters generating radio wave pulse and ultrasonic pulse when the foot contacts with the ground, are mounted at the instep of the left and right foot.

[73] After loading this equipment, the inclination angle is measured and memorized during an erect stance.

[74] A gait mode is determined by the signal pattern of the 3-axis accelerometer and the distance vector from the reference point to the foot is calculated.

[75] Therefore, horizontal displacement and vertical displacement are calculated and corresponding three dimensional positions is calculated by coordinate transformation. Mode for the Invention

[76] When new mountain pass is developed, the developer records the trajectories by

GPS when GPS signal is available, otherwise path data is recorded from this invention. GIS such as a guidance of mountain path can be constructed by three dimensional trajectory data.

[77] Thereafter precision guidance can be provided for users of GIS such as mountain path, with the displacement information by calulating with the tilt angle of limbs or measuring the distance vector of climber's foot, or by estimating horizontal displacement with the basic sensor module of this invention and applying map-matching method to get precise guidance.

[78] Errors might be accumulated unnecessarily in case of staying in a certain area longtime such as making camp in valley where GPS signal is weak.

[79] Therefore, the correction points having accurate WGS-84 coordinate need to be installed at such area to recover the position accuracy. Industrial Applicability

[80] This invention can be applied to acquire accurate position of guidance system for blind pedestrian. When a blind person checks the ticket, position accuracy can be maintained by the correction point installed around the ticket gate of subway.