Patent Application
20150369599
Under 35 USC 119, this application claims the benefit of the Jun. 18, 2014 priority date of French Application FR 1455576, the content of which is herein incorporated by reference in its entirety.
The invention relates to a method of location of a device which is displaced inside a three-dimensional space. The subject of the invention is also an information recording medium, an electronic unit and a device for the implementation of this method.
Such methods are particularly useful when a method of location of the device by GPS (“Global Positioning System”) is not possible. This is for example often the case inside buildings.
Known methods of location use the measurements of an inertial platform housed inside the device itself to measure its direction of displacement and the amplitude of its displacement in this direction from a previous position. Among these known methods, some use particle filters to estimate the position of the device in the three-dimensional environment. Accordingly, particle filters exploit the fact that there exist predefined constraints on the displacements of the device inside the three-dimensional environment. For example, a typical constraint is that a displacement cannot pass through a wall.
Thus, these known methods of location typically comprise:
a) the provision of a map of the three-dimensional space and of predefined constraints on the displacements of the device in this three-dimensional space,
b) the generation, by an electronic computer, of several distinct particles, each particle being associated:
c) the reception of measurements representative of the direction of displacement of the device and of the amplitude of this displacement from its previous position, these measurements being carried out by sensors onboard the displaced device,
d) the updating of the coordinates of the position of each particle as a function of the measurements received during step c) and of a predetermined displacement law for displacing this particle from its previous position Pk-1i to a new position Pki in a manner correlated with the measured displacement of the device, each displacement law comprising for this purpose at least one first measured variable whose value is dependent on the measurement of the direction of displacement received during step c) and a second measured variable whose value is dependent on the measurement of the amplitude of this displacement received during step c), and then
e) for each particle, if the latest displacement of this particle from the position Pk-1i to the position Pki satisfies the predefined constraints, the increasing of the weight associated with this particle with respect to the weights of the particles whose latest displacement infringes these predefined constraints,
f) the estimation of the position of the device on the basis of the positions of the particles and of the weights associated with these particles by allotting, during this estimation, more importance to the positions of the particles associated with the highest weights.
Such known methods of location of a device implementing a particle filter are for example disclosed in:
Such a method is also disclosed in detail in the thesis of J. Straub, “Pedestrian indoor localization and tracking using a particle filter combined with a learning accessibility map”, thesis, August 2010, Technical University of Munich. This thesis is downloadable at the following address: http://people.csail.mit.edu/jstraub/download/Straub10PedestrianLocalization.pdf.
Subsequently, this thesis is referenced through the term “Straub 2010”.
Prior art is also known from:
The invention is aimed at proposing a method of location of the device that is more precise.
Its subject is therefore a method of this type, as claimed in claim 1.
The method hereinabove makes it possible to exploit moreover the fact that there may exist zones of a map in which the directions of displacement are not equiprobable. This makes it possible to more rapidly and more precisely adjust the weight of the particles situated in the zone with favored direction of displacement and therefore to increase the precision of the estimation of the position of the device.
The embodiments of this method of location can comprise one or more of the additional characteristics of the dependent claims.
These embodiments of the method of location furthermore exhibit the following advantages:
The subject of the invention is also an information recording medium comprising instructions for the execution of the above method of location, when these instructions are executed by an electronic computer.
The subject of the invention is also an electronic unit for locating a device displaceable inside a three-dimensional space.
Finally, the subject of the invention is also a device directly transportable by a pedestrian who is moving in displacement inside a three-dimensional space, this device comprising:
The invention will be better understood on reading the description which follows, given solely by way of nonlimiting example and while referring to the drawings.
In these figures, the same references are used to designate the same elements. Hereinafter in this description, the characteristics and functions that are well known to the person skilled in the art are not described in detail.
To aid the pedestrian 4 to locate themself inside the building 2, the latter transports a location device 10 directly in their hand. The device 10 is capable of locating itself on a map of the building 2 without recourse to sensors other than those which it comprises internally. In particular, the device 10 can chart its position inside the building 2 without using a navigation system calling upon external charting beacons implanted in the environment of the building 2. These external beacons can be satellites or radio wave emitters fixed to the building 2. In particular, the device 10 can chart its position without using a GPS system (“Global Positioning System”).
The device 12 also comprises an inertial platform 17. The inertial platform 17 transmits to the computer 14, by way of an information transmission bus 19, measurements of the direction in which the device 10 is moving and of the amplitude of the displacement in this direction from the latest logged position of this device 10. For this purpose, for example, the inertial platform 17 comprises a three-axis accelerometer 18, a gyrometer 19 and a three-axis magnetometer 20. Here, the inertial platform 17 also comprises a barometer 21 for measuring the altitude of the device 10.
The device 10 is equipped with a screen 24 making it possible to display a graphical representation 26 of the map 16 and, on this graphical representation, a point PA representing the current position of the device 4 inside the building 2. This point PA is therefore situated in the graphical representation 26 at the site of the map 16 corresponding to the current position of the device 10 and therefore of the pedestrian 4.
For example, the device 10 is a smartphone or an electronic tablet programmed to execute the method of
The map 16 comprises several zones 30 to 35 whose juxtaposition covers the entire area of the story 8. These zones are parallel to the floor of the story and contained in one and the same plane called the “plane of the story”. This plane of the story is typically horizontal.
The zones 30 to 35 are contiguous and do not overlap. However, to increase the readability of
Conventionally, the edges of each zone are situated at the site of an obstacle which is impassable to the pedestrian 4 such as a wall. However, here, one and the same zone can encompass several rooms of the building 2. This is for example the case for the zone 33 which surrounds a room 40 and another smaller room 42. In
A zone can also encompass obstacles situated in the interior of a room which are impassable to the pedestrian 4. For example, an impassable obstacle is an interior partition or a pillar or any other element of the building 2 which the pedestrian 4 cannot cross. Such a zone is illustrated by the zone 31 which comprises two partitions 44 and 46 situated inside a room 48.
Here, each zone is a polygon. Hence, for each zone, the map 16 contains:
The XYZ frame is an orthogonal frame in which the X and Y directions are horizontal and the Z direction is vertical. In this embodiment, each zone is rectangular. Thus, only the coordinates of the two diagonally opposite vertices of a zone are recorded in the map 16 so as to economize on memory.
For each zone, the map 16 also comprises:
The position and the dimensions of each impassable object are here coded by a horizontal segment contained in the plane of the floor. Thus, each obstacle identifier is associated with a pair of points Ejd and Ejf. The points Ejd and Ejf mark, respectively, the start and the end of the segment [Ejd; Ejf], where j is the identifier of the impassable obstacle. The coordinates of the points Ejd and Ejf, in the plane of the floor, are known and contained in the map 16. Hence, for example, the zone 31 comprises eight obstacle identifiers IdO1 to IdO8. The identifiers IdO1 to IdO8 correspond, respectively, to the segments [Ea; Eb]; [Eb; Ec]; [Ec; Ed]; [Ed; Ef]; [Ef; Eg]; [Eh; Ei]; [Ej; ER] and [Ep; Em]. The position of the points Ea to Em is represented in
Each displacement law makes it possible to compute, on the basis of the measurements of the inertial platform 17, at an instant tk, the displacement of a particle Si from a previous position Pk-1i up to a new position Pki. This displacement is directly correlated with that of the device 10. Typically, this displacement between the positions Pk-1i and Pki is identical or very close to that of the device 10 between the instants tk-1 and tk. Subsequently, the superscript “i” is the identifier of the particle and the subscript “k” is the order number of the instant at which the direction and the amplitude of the displacement of the device 10 are measured.
The inertial platform 17 measures the angle θk, in the plane of the story, between the direction of displacement of the device 10 and the X direction. For this purpose, it is possible to use the measurements of the gyrometer 19 and of the magnetometer 20. Subsequently, the direction measured at the instant tk is called “direction θk”.
The inertial platform 17 is also capable of providing a physical quantity representative of the amplitude Ik of the displacement of the device 10 in the direction θk between the instants tk-1 and tk. For this purpose, it is possible to integrate the measurement of the accelerometer 18 between the instants tk-1 and tk and, if the measurement is zero, retain the previous measured speed vk-1. However, in the case of a pedestrian who is walking on a horizontal floor, to obtain a more precise estimation of the amplitude Ik, the computer 14 detects on the basis of the measurements of the accelerometer 18 the instant at which a foot of the pedestrian 4 comes into contact with the floor. On the basis of these successive instants, the computer 14 computes a frequency fk of the footsteps of the pedestrian 4. Next, the amplitude Ik of the displacement of the pedestrian 4 in the direction θk is computed by using the following footstep model: Ok=Afk+BT+C, where A, B, C and T are constant coefficients independent of the measurements of the inertial platform 17. Moreover, the coefficients A, B and C are coefficients which are independent of the morphological characteristics of the pedestrian. On the other hand, the coefficient T must be chosen equal to the height of the pedestrian 4. By default, the coefficient T is taken equal to the mean height of a human being, for example 1.78 m. The speed vk of displacement of the device 10 between the instants tk-1 and tk is obtained with the aid of the following relation: vk=Ik/Δt, where Δt is the duration of the time interval between tk-1 and tk. Typically, Δt is chosen equal to the duration of a footstep of the pedestrian 4.
Thus, when the pedestrian 4 moves by walking on the floor of the story 8, a displacement law is given by the following relations:
x
k
i
=x
k-1
i
+v
k
×Δt×cos θk;
y
k
i
=y
k-1
i
+v
k
×Δt×sin θk,
where (xki, yki) and (xk-1i, yk-1i) are the coordinates, in the plane of the floor, of the positions Pki and Pk-1i of the particle Si.
In the case of a method of location of the device 10 implementing a particle filter, it is beneficial to explore the largest possible number of trajectories with the particles. Thus, conventionally, the displacement of each particle is disturbed in a random manner. For example, accordingly, the usable displacement law is the following:
x
k
i
=x
k-1
i
+v
k
×Δt×cos θk+μxi;
y
k
i
=y
k-1
i
+v
k
×Δy×sin θk+μyi;
where μxi and μyi are random variables.
At each instant tk and for each particle Si, the values of these variables μxi and μyi are randomly drawn as a function of a predefined centered probability law, that is to say characterized by a zero mathematical expectation. Thus, the mean of the values of each random variable μxi and μyi at the various successive instants tk tends to zero as k increases. For example, this predefined probability law is the same for the random variables μxi and μyi and for all the particles Si. Subsequently, it is denoted Lpxy. This law Lpxy is characterized by a predetermined standard deviation σxy. Here, the standard deviation σxy is constant and independent of the measurements of the inertial platform 17 for an updating at each footstep. For example, the standard deviation σxy is greater than 5 cm or 10 cm and, preferably, less than 35 cm. For example, the law Lpxy is a uniform distribution or a Gaussian distribution.
In reality, there may also exist a measurement bias, called the direction bias, in the measurement of the direction θk. Such a direction bias may originate from a defect in the sensors of the inertial platform 17. This direction bias may also be caused by the fact that the pedestrian 4 rotates the device 10 in a horizontal plane. In a similar manner, there may also exist a measurement bias, called here the footstep bias, in the measurement of the amplitude Ik of the displacement of the device 10. This footstep bias may originate from a defect of the sensors of the inertial platform. In the example described here, it may also originate from a modeling error and more precisely from an error with respect to the default value of the coefficient T in the footstep model. Indeed, the actual height of the pedestrian 4 is unknown. Thus, if the pedestrian 4 is much shorter or much taller than 1.78 m, this introduces a systematic footstep bias in the measurement of the amplitude Ik. Typically, these biases are constant at least over a time interval that is long enough to be able to estimate them and correct them as described subsequently.
Here, to compensate and correct these direction and footstep biases, the displacement law used integrates corrective factors, respectively αi and εi, associated with each particle Si. For example, the displacement law is given by the following relations:
x
k
i
=x
k-1
i
+v
k
×Δt×(1+εki)×cos(θk+αki)+μxi;
y
k
i
=y
k-1
i
+v
k
×Δt×(1+εki)×sin(θk+αki)+μyi;
εki=εk-1i+μεi;
αki=αk-1i+μαi;
where:
The random variables μεi and μαi are used for the same reasons and in the same manner as the variables μxi and μyi introduced previously. Thus, a new value of the variables μεi and μαi is randomly drawn at each new instant tk and for each particle Si as a function, respectively, of a predefined probability law Lpε and of a predefined probability law Lpα. Typically, these laws Lpε and Lpα are the same for all the particles Si. Here, the mathematical expectations of the laws Lpε and Lpα are equal to zero. Consequently, just as for the random variables μxi and μyi, the mean of the values of each random variable μεi, and μαi at the various successive instants tk tends to zero as k increases.
Moreover, the function of the variables μεi, and μαi is only to slightly disturb the previous values εk-1i and αk-1i of the corrective factors εi and αi so that the values of the corrective factors εi and αi remain stable over time. For this purpose, the standard deviations σεand σα, respectively, of the laws Lpε and Lpα do not allow a fast variation of the values of the corrective factors εi and αi. Here, for this purpose, the standard deviation σε is chosen sufficiently small for the ratio Σσεk/T to be less than 10%/s and, preferably, less than 5%/s or 1%/s, where:
Step 96 is the step during which the coordinates of the particle Si are updated. This step is described in greater detail with reference to
In a similar manner, the standard deviation σα is chosen sufficiently small for the ratio Σσαk/T to be less than 10°/s and, preferably, less than 5°/s or 1°/s, where:
Here, the standard deviation σα is also constant. Thus, the previous ratio can also be written: (p−q)σα/T.
Just as for the variables μxi and μyi, the variables μεi, and μαi make it possible to explore a large number of possible values for the corrective factors εi and αi.
The displacement law described hereinabove is subsequently called the first displacement law. This first displacement law operates in most situations where the pedestrian 4 walks on a horizontal ground. On the other hand, in certain particular cases, there exist other more precise displacement laws. For example, the zone 35 is a stairwell comprising a staircase 50. In this zone 35, the length of the footsteps of the pedestrian 4 is imposed by the depth Lm of the stairs of this staircase 50. Thus, in the zone 35, it is preferable to use a second displacement law. Here, this second displacement law is identical to the first displacement law except that the product vki×Δt×(1+εki) is replaced with the measured variable nki. The variable nki is given by the following relation: nki=(Ent(|zki−zk-1i|/Hm))×Lm, where:
Here, it is assumed that the height Hm and the depth Lm are known in advance and recorded in the map 16.
The values of the measured variables zk and zk-1 are obtained, typically, on the basis of the measurements of the barometer 21. Here, only the zone 35 is associated with this second displacement law. All the other zones of the story 8 are associated with the first displacement law.
Inside the building 2, there exist zones such as the zones 30, 31, 33 and 34 in which the pedestrian 4 can move freely in all directions. Stated otherwise, in these zones, it is considered that all the directions of displacement are equiprobable. In this case, these zones are devoid of favored direction of displacement. Conversely, there exist zones of the building 2 where not all the directions of displacement are equiprobable. For example, the zone 32 is a long corridor parallel to the X direction. In this zone 32, the most probable directions of displacement for a pedestrian are parallel to the X direction. Indeed, it is less probable for the pedestrian 4 to move transversely to the longitudinal direction of the corridor. In this case, there is said to be a favored direction of displacement in this zone 32. Here, each favored direction is coded by an angle γ, in the plane of the floor, between this favored direction and the X direction. Moreover, in this embodiment, an angular tolerance σγ is also associated with each favored direction. This angular tolerance is generally expressed in degrees or in radians. For example, in the case of the zone 32, the angle γ=0° and the angular tolerance σγ=±30°. In the embodiment described here, the zone 35 is also associated with a favored direction for ascending and descending the staircase 50. For this favored direction, the angle γ=90° and the angular tolerance σγ is equal to ±20°.
The manner of operation of the device 10 will now be described with reference to the method of
To locate the position of the device 10 inside the building 2, the unit 11 implements a location algorithm known by the term “particle filter”. The manner of operation of a particle filter is well known. For example, the reader can refer to the prior art cited in the introduction of this patent application and, in particular, to Straub 2010. Thus, subsequently, only the new specifics of implementation of this algorithm are described in detail. In particular, the management of the changes of stories is carried out, for example, as described in Straub 2010.
The method starts with a step 90 when location of the device 10 is triggered. For example, location is triggered manually by the pedestrian 4 by interacting with the man-machine interface of the device 10. The computer 14 then generates an initial assembly of N0 particles Si, where i is an identifier of the particle making it possible to distinguish it from among the set of other particles generated. The number N0 of particles Si depends in particular on the initial knowledge that one has about the position of the device 10 in the building 2 and the area of this building. Typically, N0 is greater than 10 or 100. N0 is also generally less than 5000 or 1000. Each particle Si is initially associated:
During step 90, the initial position P0i and the initial values w0i, α0i and ε0i are initialized. Numerous schemes for initializing the positions P0i and the values w0i of each particle are known. For example, if the initial position of the device 10 is known to within plus or minus 1 m, the values P0i of all the particles Si are drawn at random inside a circle centered on the known position and of radius equal to 1 m. It is also possible to take each value w0i equal to 1/N0, where N0 is the initial number of particles generated.
Each value α0i is drawn at random in such a way that the distribution of the initial values α0i follows a predetermined probability law Lpα0 such as a uniform or Gaussian or other distribution. The law Lpα0 is generally not the same as the law Lpα which is used to obtain the values of the random variable μαi. It is the a priori knowledge that one has about the distribution of the direction bias which makes it possible to choose the probability law for the initial values α0i which most resembles the one observed in reality. For example, it is also this a priori knowledge about the distribution of the direction biases which makes it possible to fix the value of the standard deviation σα0 of the law Lpα0. For example, the standard deviation σα0 is chosen equal to 360° if one has no information about the direction bias. In another example, the standard deviation σα0 is chosen less than 45° if one has a little information about the direction bias. In contradistinction to the case of the previous random variables, this probability law Lpα0 does not necessarily have a zero mean.
Each initial value ε0i is chosen as described previously for the initial values α0i except that a predefined probability law Lpα0 is used for the footstep bias, instead of the law Lpα0. Moreover, the probability law Lpε0 is not necessarily identical to the law Lpα0. Indeed, generally, the direction bias and footstep bias are not correlated. Typically, the standard deviation σε0 of the law Lpε0 is equal 30% to within plus or minus 5% if one has no information about the footstep bias. In another example, the standard deviation σε0 is chosen less than 20% if one has a little information about the footstep bias.
Thereafter, during a step 92, the inertial platform 17 transmits its measurements to the computer 14 which receives them. On the basis of these measurements, the computer 14 computes the speed vk and the angle θk of the current displacement of the device 10 from its latest position. Here, new measurements of the speed vk and of the angle θk are computed each time that the computer detects that a foot of the pedestrian has just touched the floor.
During a step 94, the computer 14 identifies, for each particle Si, the zone inside which it is currently situated. Accordingly, for each particle Si, the computer compares the latest known position Pk-1i of this particle with the periphery of each zone of the map 16. For example, in the case of rectangular zones aligned with the XYZ frame, the computer 14 tests whether the following two inequalities are satisfied:
x
Aj
≦x
k-1
i and yAj≦yk-1i≦yBj,
where:
If these two inequalities are satisfied, then the particle Si belongs to zone j. Preferably, the computer 14 tests firstly whether the particle Si belongs to the same zone as that to which it belonged previously. If a particle Si does not belong to any zone, then a particular processing is triggered. For example, this particle is eliminated.
During a step 96, once the zone to which each particle Si belongs has been identified, each particle is displaced in a manner correlated with the measured displacement of the device 10 from its previous position Pk-1i up to a new position Pki. Accordingly, the coordinates of each particle Si are updated as a function:
The displacement law to be used is therefore that associated with the zone identified during step 94.
For example, if the particle Si is situated inside one of the zones 30 to 34, then the coordinates of the new position Pki are established using the first displacement law. On the other hand, if the particle Si is situated inside the zone 35, then the coordinates of the new position Pki are established using the second displacement law.
At each new execution of step 96, the new values for the variables μxi, μyi, μαi and μεi, are randomly drawn with the aid of the laws Lpxy, Lpα, and Lpε, respectively.
During a step 98, the computer 14 updates the weights wi of each particle Si. More precisely, the computer 14 decreases the weight wi of the particle Si if its latest displacement from the position Pk-1i to the position Pki has infringed predefined constraints associated with the zone inside which it is situated.
Typically, for each particle Si, the computer here verifies the following constraints:
Generally, here a constraint on the displacement of the particle Si is defined as being a condition which, if it is satisfied by the particle Si, is used to increase the weight wi of this particle Si with respect to the weight of the particles which do not satisfy this condition. Conversely, if this condition is not satisfied by the particle Si, then it is used to decrease the weight wi of this particle Si with respect to the weights of the particles which satisfy this condition.
To evaluate the constraint 1), for each particle Si, the computer 14 searches for whether there exists an intersection between the segment [Pk-1i; Pki] and each impassable obstacle of the zone inside which the particle is situated. This zone was identified during step 94. This intersection search is carried out solely with the impassable obstacles whose identifiers are contained in the list associated with the identified zone. Here, since each obstacle is coded by a segment, this intersection search amounts to searching for an intersection between two segments.
If an intersection exists, then a very low or zero value is assigned to the weight wi. A very low value is a value less than 0.2 or 0.1 for example. In the converse case, the value of the weight wi remains unchanged.
If the zone inside which the particle Si is situated is not associated with a favored direction, then the constraint 2) is not used to update the weight wi. In the converse case, the constraint 2) is used. To evaluate the constraint 2), the computer 14 computes a weight wθi on the basis of the deviation between the direction measured θk for the displacement of the particle Si and the favored direction of the zone inside which it is situated. The value of the weight wωi is all the larger the lower the angular deviation between the favored direction γ and the direction of displacement from the position Pk-1i to Pki. Moreover, here, the value of the weight wθi is established while taking account of the tolerance σγ associated with this favored direction. The value of the weight wθi is therefore larger if the angular deviation between the direction of displacement of the particle Si and the favored direction lies in the tolerance margin defined by the value of σγ and, in the converse case, the value of the weight wi, is smaller.
For example, here, the value of the weight wi, is computed with the following relation: wσi=p×exp[(θk+αki−γ)2/(2σγ2)], where p is a predefined constant coefficient. It will be noted that the value αki of the corrective factor αi of the direction bias is also used to compute this weight wσi.
Thereafter, the value of the weight wi is taken equal to its previous value multiplied by the value of the weight wθi thus obtained.
During a step 100, the computer 14 undertakes the normalization of the weights wi of all the particles Si so that the sum of all these weights is equal to one. For example, the computer 14 computes the sum W of all the weights wi and then divides each weight wi by the sum W.
During a step 102, the computer 14 re-samples the particles Si. This re-sampling step consists in eliminating the particles whose weights wi have become too low to replace them with particles associated with parameters whose weights are higher.
Numerous re-sampling techniques are known. For example, here, we apply the SIR (Sequential Importance Resampling) scheme described in the following book: B. Ristic, S. Arulampalam, N. Gordon, “Beyond the Kalman Filter, particle filter for tracking applications”, Artech House, 2004.
To summarize, this re-sampling scheme firstly consists in classing the particles into two groups: the particles to be regenerated, that is to say all the particles whose weight is below a predetermined threshold and the surviving particles, that is to say the other particles.
Thereafter, for each particle to be regenerated, the computer 14 eliminates the old particle and then generates a new particle to replace it. To generate a new particle, the computer 14 randomly draws a particle from the group of surviving particles. This chance drawing is carried out preferably in such a way that the probability of being drawn is proportional to the weight of the surviving particle.
Thereafter, the position Pki and the values αki and εki of the drawn surviving particle are assigned to the new generated particle. During this step, preferably, the computer adds a random disturbance on the position and on the values of the corrective factors by using, for example, the random variables μx, μy, μα and με. However, these disturbances are calibrated in such a way that the values αki and εki assigned to the generated particle remain very close to the current values for the corrective factors associated with the surviving particle drawn. Typically, the mean of the values αki which are assigned to the generated particles is closer to the mean of the values αki of the surviving particles than to the mean of the values αki of the eliminated particles. The same holds for the values εki assigned to the generated particles.
A new weight is also assigned to each generated particle and, optionally, to the surviving particle from which it arises.
During a step 104, the computer 14 estimates the position PA of the device 10 on the basis of the positions Pki and of the weights wi of all the particles Si. Numerous schemes are possible for doing this. For example, the position PA is taken equal to that of the particle Si having the highest weight wi. In another embodiment, the position PA is taken equal to the mean of the positions Pki of the particles Si on weighting each position Pki by the weight wi. Another scheme is also described in application WO 2012158441.
Finally, during a step 106, a point is displayed at the position PA on the graphical representation 26 of the map 16 to indicate to the pedestrian 4 his position on this map and therefore his position inside the building 2.
Steps 92 to 106 are repeated in a loop. As these iterations proceed, the precision of the estimation of the position of the device 10 increases since only the most probable positions Pki are retained.
In parallel with step 104, during a step 108, the computer 14 computes an estimation Eα of the direction bias and an estimation Eε of the footstep bias on the basis, respectively, of the current values αki and εki of the particles Si. For example, the estimation Eα is taken equal to the mean of the current values αki of all the particles Si. In a similar manner, the value of the estimation Eε is taken equal to the mean of the current values εki of all the particles Si.
The precision of the estimations Eα and Eε increases as the repetitions of steps 92 to 106 proceed. Indeed, it is very probable that the particle Si associated with an incorrect current value for the corrective factor αi or εi will rapidly infringe the constraints evaluated during step 98. Hence, the particles Si associated with incorrect current values for the corrective factors are preferably eliminated during step 102. Thus, as the iterations of steps 92 to 106 proceed, the particles Si associated with correct current values for the corrective factors are preferably selected as surviving particles during step 102. Hence, the estimations Eα and Eε converge toward the actual values of the direction bias and the footstep bias.
Moreover, since the values of the corrective factors αi and εi converge toward the actual values of the direction and footstep biases as the iterations of steps 92 to 106 proceed, the corrective factors correct these direction and footstep biases more and more precisely. Hence, the presence of these corrective factors in the displacement law makes it possible to substantially increase the precision of the estimation of the position PA even in the presence of such direction and footstep biases.
The estimations Eε and Eα can be displayed on the screen 24 or used by other applications to correct the direction and footstep biases.
Numerous other embodiments are possible. For example, certain more sophisticated sensors provide a mean value of the variable measured at an instant k as well as a standard deviation in this measurement (see for example Straub 2010). In this case, the measured values of the angle θk and of the amplitude Ik are obtained by drawing the values at random using a Gaussian probability law whose mean and standard deviation are equal to those transmitted by the sensors. This makes it possible in particular to take the measurement noise into account.
What was described previously in the particular case where it is a pedestrian who directly carries the device 10 may also be adapted to other situations. For example, the device 10 can be fixed on a trolley pushed by the pedestrian. In this case, the speed vk is obtained by integrating the acceleration measured by the trolley. Another solution consists in detecting the frequency of the impacts which occur each time a wheel of the trolley rolls over a join between two flagstones of a ground pavement. The method of location described also applies to the case where the device 10 is transported by a motorized robot which moves without the aid of the pedestrian 4. In this case, the amplitude of the displacement can be measured by measuring the number of wheel revolutions of a driving wheel of the robot.
The method described above applies to any type of three-dimensional space where the device 10 can be displaced. For example, it may also entail a space situated outdoors and outside any building. For example, this method is useful to locate a person in a place where location by GPS or on the basis of telephonic relay is impossible.
Other ways of coding the constraints on the displacement of the device 10 exist and are usable in the above method. For example, rather than recording the position of the walls, it is possible to record all the possible paths. If the displacement of a particle does not follow one of these possible paths, then its weight is decreased. Such an embodiment is described in greater detail in application WO 2012158441 and in application U.S. Pat. No. 8,548,738. Rather than recording the position of the walls, it is also possible to record the position of the doors. In the latter case, the weight of a particle is increased if the latter goes from one room to another by passing through a door. Such an embodiment is described in greater detail in Straub 2010.
Other predefined constraints can be used to update the value of the weight wi. For example, if an approximate measurement of the position of the device 10 is available by using other information, then the weight wi is increased if the position Pki is close to this approximate position and, on the contrary, decreased if the position Pki is far from this approximate position. For example, the approximate position is obtained on the basis of the power of a radioelectric signal received by the device 10 and of the known position in the XYZ frame of the emitter of this radioelectric signal. For example, the emitter is a Wi-Fi terminal.
Although the constraint consisting in testing the collision of a particle with an impassable obstacle is, in practice, the one used most often, it is not absolutely necessary if there exist other predefined constraints such as those described above which can be used.
As a variant, step 100 of normalizing the weight wi of the particles Si is omitted.
During the re-sampling of the particles, numerous other algorithms for determining the initial position of the regenerated particles are known and may be used instead of that described above. For example, the KLD (Kullbak-Leibler-Divergence) algorithm can be used.
Likewise, other algorithms are possible for associating a new current value αki and εki with each regenerated particle. Each new value αki is dependent on one or more of the values αki associated with the particles which have not been eliminated and is independent of the values αki associated with the particles which have been eliminated. The same holds for the new value εki.
In a simplified variant, the re-sampling step 102 is omitted. Even if the re-sampling is not undertaken, the method hereinabove makes it possible to increase the precision of the location of the position of the device 10 since the weight wi of each particle Si associated with an incorrect current value αki or εki decreases as the iterations of steps 92 to 104 proceed.
Other schemes for determining whether a particle is situated inside a zone are possible. For example, Straub 2010 describes an alternative scheme usable in the case of polygonal zones of more complicated shape than a simple rectangle.
Shapes of zones other than polygons are possible. For example, a zone can have the shape of a circle. In this case, the coordinates of its center and of its radius are recorded in the map 16. There is in reality no limitation on the shape of a zone as long as the coordinates of the periphery of this zone can be determined in the XYZ frame.
The above method has been described in the particular case where each zone is associated at one and the same time with impassable-obstacle identifiers, a displacement law and a favored direction. However, it may frequently happen that the displacement law is the same for several immediately contiguous zones. For example, here, this is the case for zones 30 to 34 which are all associated with the first displacement law. In this case, it may be more beneficial to record in the map 16 several stacked strata of zones for one and the same story. Each stratum comprises at least one zone and, typically, a set of several zones covering the entire area of the floor. The zones of one stratum are distinguished from the zones of another stratum by the type of property that it associates with this zone. For example, a first stratum comprises solely zones associated only with impassable-obstacle identifiers. A second stratum comprises solely zones associated only with a respective displacement law. Finally, a third stratum comprises only zones associated solely with a favored direction. When such a stack of strata is applied to the story 8, the zones of the first stratum are, for example, identical to the zones 30 to 35 except that they comprise only the identifiers of impassable obstacles. Here the second stratum is limited to two zones. One of them is identical to the zone 35 except that it is solely associated with the second displacement law. The other zone of this second stratum corresponds to the union of zones 30 to 34 and is solely associated with the first displacement law. Here the third stratum comprises four zones. The first and second zones are identical, respectively, to zones 32 and 35 except that they are solely associated with a respective favored direction. The third zone corresponds to the union of zones 30 and 31 and the fourth zone then corresponds to the union of zones 33 and 34 which are not associated with any favored direction.
The manner of operation of the method of location with a map comprising several strata is identical to that described above except that, for each particle S′, the zone of each stratum inside which it is situated is identified. Thereafter, it is these zones inside which it is situated which make it possible to identify the displacement law to be used and the predefined constraints to be tested so as to update the value of its weight wi. The use of several strata of zones may simplify the definition of these zones.
In another variant, one of these strata corresponds to an accessibility map such as described in Straub 2010 in chapter 5.1.2. However, preferably, the various contiguous boxes of the accessibility map of Straub 2010 that are associated with the same value of the degree Λ of accessibility are grouped together within one and the same zone.
Each zone can also be associated with additional predefined constraints for updating the weight wi of the particles situated in this zone. For example, it is possible to associate with each zone a coefficient wa of accessibility which represents the probability that a pedestrian enters this zone. Thereafter, when updating the weight wi of a particle Si situated in this zone, the weight wi is taken equal to its previous value multiplied by this coefficient wa.
Other displacement laws are usable. For example, the standard deviations σα and σε are not necessarily constant. In this case, they are constant over long durations and then modified for a brief time interval before returning to their previous values. For example, solely during this brief time interval, the ratios Σσεk/T and/or Σσαk/T are permitted to exceed the previously defined thresholds. Typically, the temporary increasing of the values of the standard deviations σα and σε is used to reinitialize the current values αki and εki. Generally, the values of the standard deviations σα and σε are constant for more than 90% of the time of use of the device 10 so as to prevent, during this 90% of the time, a fast variation of the values of the corrective factors εi and αi. It is possible to verify that the ratio Σσεk/T is maintained below a predetermined threshold over at least 90% of the time of use, for example, by taking the difference p-q equal to a constant. Thereafter, the ratio Σσεk/T is computed for each value of p corresponding to an iteration of step 96 occurring during this time of use. If at least 90% of the ratios thus computed are below the predetermined threshold, then this ratio is maintained below this predetermined threshold for more than 90% of the time of use. The same computation can be used for the ratio Σσαk/T. The time of use is for example a period of continuous use of the device 10 without stopping the execution of the method of
In another variant, the expectation of one of the probability laws Lpε or Lpα is non-zero. This then introduces an additional bias which is added to the real bias. The current value εki or αki can also be computed on the basis of a previous value other than εk-1i or αk-1i. For example, εki is computed on the basis of the previous value εk-ni, where n is a constant strictly greater than one. Thus, it is also possible to use the following relations to compute the current value of εki: εki=(εk-1i+εk-2i)/2+μεi or εki=εk-2i+μεi. The same thing applies to the computation of the current value αki.
Moreover, if it is known that the direction bias is zero or negligible, then the first and the second displacement laws are simplified by eliminating the corrective factor αi. Conversely, if it is known that the footstep bias is zero or negligible, then the first displacement law is simplified by eliminating the corrective factor εi. The corrective factor remaining in the first displacement law is thereafter estimated as described above with reference to
Displacement laws other than those described above may be used. For example, a displacement law specifically adapted to the displacement on an escalator or on a moving walkway or in an elevator can readily be designed and associated with the zone comprising this escalator, this moving walkway or this elevator. Moreover, if there exists a measurement bias in these particular zones, then it is desirable to introduce a corrective factor into the displacement law to compensate this measurement bias. If, initially, the value of this corrective factor is not known, then this value is estimated in the same manner as was described for the corrective factors αi and εi.
Step 106 or 108 is not necessarily carried out after each iteration of steps 92 to 104. For example, these steps are carried out only one time out of two.
The previous embodiments have been described in the particular case where all the processings to determine the position PA are carried out by the computer 14 of the device 10. As a variant, these processings are distributed over several distinct computers. One of these distinct computers can be that of a computer server mechanically independent of the device 10. For example, the device 10 transmits the measurements to this server and this server determines the position PA and then returns the position PA determined to the device 10 which displays it on its screen 24. In this case, the map 16 is also recorded in the memory of this server.
The selection of a specific displacement law as a function of the zone in which the particle is situated can be implemented independently of the other characteristics of the method of
Likewise, the use of favored directions of displacement associated with zones can also be implemented independently of the other characteristics of the method of
| Number | Date | Country | Kind |
|---|---|---|---|
| 1455576 | Jun 2014 | FR | national |
