WO2016139615A1

WO2016139615A1 - Method and system for real-time location

Info

Publication number: WO2016139615A1
Application number: PCT/IB2016/051200
Authority: WO
Inventors: Simone PEIRANI; Maurizio Valle; Saverio Pagano
Original assignee: Universita' Degli Studi Di Genova
Priority date: 2015-03-04
Filing date: 2016-03-03
Publication date: 2016-09-09

Abstract

System and method for real-time location of inanimate objects and/or persons (10) and/or animals within an environment (9), each object or person or animal being associated to a transmitting mobile device (1), there being provided a plurality of transceiver stations (2) placed in predetermined positions within the environment and a monitoring unit (3) in communication with said stations (2). The method provides the following steps: transmitting a signal by the mobile device (1); receiving the transmitted signal by a plurality of stations (2) reached by the signal and for each station calculating (2) a parameter of the distance from the mobile device (1), which distance parameter is obtained from the reception characteristics of the signal by the station (2); selecting a reference station (2') among the stations (2) reached by the signal, which reference station (2') is identified as the closest to the mobile device (1) on the basis of the distance parameter; selecting a set of stations (2), among the stations (2) reached by the signal, which set comprises the reference station (2') and three or more stations (2) closest to the reference station (2'); identifying a plurality of clusters corresponding to all the possible combinations of at least three stations (2) of the set of stations; for each cluster calculating the position of the mobile device (1) on the basis of the distance parameter of each station (2) of the cluster; calculating the position of the mobile device (1) by working out the mean of the positions calculated for each cluster.

Description

Method and system for real-time location.

The present invention relates to a method and a system for real-time locating inanimate objects and/or persons and/or animals within an environment.

The method provides each object or person or animal to be associated to a transmitting mobile device, so called beacon node, there being further provided a plurality of transceiver stations, so called anchor nodes, placed in specific positions within the environment and a monitoring unit in communication with said transceiver stations.

Real time location systems (RTLSs) generally are based on wireless communication technologies. A RTLS is composed of a set of wireless transmitters and receivers. There are two types of transmitters and receivers: fixed and mobile. Fixed transmitters and receivers act as a reference for the mobile transmitters, determining their position. Currently the technologies most used in RTLSs are:

- GPS (Global Positioning System) : it is basically a series of satellites (fixed transmitters in earth orbit) , that constantly transmit information to mobile receivers on the earth surface. Receivers calculate their positions on the basis of satellite coordinates, therefore the reached accuracy increases as the number of references to the satellites increases. It is necessary to have at least three satellite references for calculating the position. - AGPS (assisted GPS) : it is a mixed technology usually implemented in mobile devices having a GPS receiver. It uses signals provided by satellites and by mobile telephone networks. It is mainly used in two cases: when the GPS signal is low and when the mobile device starts the execution of the GPS device, that is when the device takes its position in the mobile telephone network to assist GPS.

GSM (Global System for Mobile Communications) /GPRS (General Packet Radio Service): these are locating services offered by mobile telephone providers. The operation is based on using the same network of antennas providing the telephone service. In this case, the position can be calculated both in the mobile device and by the service provider, since both earth antennas and the antennas of the mobile devices can act as transmitters and receivers. There are several parameters used for calculating the position, for example the time of arrival of the signal, the angles of incidence, triangulation of signals or relevant cells.

All such systems cannot be used in indoor environments, where GPS signal has not enough strength and where greater accuracies are required for example in order to discriminate the position inside rooms or areas .

GPS is operatively limited to outdoor environments, where satellite signals can be received, and therefore it cannot be used in tunnels, houses, offices, etc... GSM can be used only where there is mobile telephone coverage. Moreover, the obtained accuracy is very poor, with error ranges even in the order of magnitude of kilometers. It is possible to increase accuracy by using several antennas .

Currently RTLSs specifically designed for indoor applications are known, they can be divided into different categories on the basis of two main parameters: technology and frequency band used; the parameter used for calculating the ranging, that is the distance between the transmitting mobile device and the transceiver station. Such methods are widely used and provide a relatively efficient operation, but have some drawbacks as mentioned above.

The technologies most used are:

Wi-Fi (Wireless Fidelity) : it uses wireless devices for calculating the location of the transmitting mobile devices. This type of systems use a network of static devices (transmitters and receivers) operating as a reference for mobile devices. The static devices calculate the position of the mobile devices by the received signals. There are several techniques for processing such signals for determining the position of the mobile devices, among which triangulation , hyperbolic positioning, etc... The used frequency is 2,4 GHz, and it is based on IEEE 802.11b/g/n standard. This system does not require a dedicated network infrastructure since it is possible to use Wi-Fi infrastructure already existing in the location. The locating accuracy however is not lower than 3 to 5 m and the system provides a high power consumption.

ZigBee: it is based on IEEE 802.15.4 standard, it works at a frequency of 2,4 GHz. Its advantage is to have low power consumption and to use mesh networks , but at the same time it requires a dedicated network infrastructure and it provides locating accuracies not less than 3-5 m.

Bluetooth: it is based on IEEE 802.15.1 standard, it works at a frequency of 2,4 GHz. The system is characterized by low power consumption and a short range, even if in the last versions radio coverages also in the order to many tens of meters are guaranteed, and it requires a dedicated network infrastructure and locating accuracies not less than 3- 5 m.

UWB (Ultra Wide Band) : such system uses UWB technology (Ultra Wide Band) as the communications protocol. The operation is similar to the Wi-Fi-based system, but the accuracy is considerably improved. The system requires a dedicated network infrastructure, but it provides locating accuracies less than 1 meter, it has low power consumption and it operates at medium- short range (maximum of 25 m) . It works at frequencies of 3 to 10,5 GHz.

RFID (Radio Frequency Identification) : its operation is based on a network of readers and tags. Readers transmit a constant radiofrequency signal received by the tags . Then the tags reply to the readers by sending an identification number. In this type of localization, each reader covers a surface through its own transmitted radio requency signal, within what is called as area coverage. When a tag passes through a specific coverage area, the system defines that the tag is physically in that area. The accuracy of these systems is limited to the reading area that can be covered by the readers. If a higher accuracy is needed, it is necessary to decrease the reading area and to install closer devices, but this considerably increases the cost of the system distribution, even if the cost of the devices is reduced. Other drawbacks of such system are the need of a dedicated network infrastructure, proximity localization, short range (up to one meter) .

The current RTLSs, regardless of the technology being used, do not guarantee satisfactory performances when the GPS is not able to operate in hostile environments for radio transmissions, for example environments where there are several obstacles such as objects, furniture, walls, people. A typical example of a hostile environment for radio transmission is building yards where many obstacles made of iron are present, such as for example shipyards.

The currently known systems mainly use the simple trilateration or multilateration for calculating the position of the transmission mobile devices. It has been found that these techniques are insufficient for reaching optimal locating accuracies, particularly less than 30 cm.

Finally the currently known systems are affected by the Non Line of Sight (NLOS) condition, that is above all in indoor environments performances decrease for all those cases when an obstacle is present between the transmitting mobile device and one or more transceiver stations.

Moreover there is a serious problem related to building yards and similar working environments, that is injuries and deaths in the workplace. The increase of such cases over the last years, has made more and more clear the need of carrying out proactive measurements on safety & security in shipyards and building yards or the like. Moreover this trend is expected to increase in the future years .

In the specific field of safety measurements in working places there are precise regulations but rescue procedures are often left completely to the human operator, due to the lack of reliable technologies in this field, such as indoor localization.

The present invention aims at overcoming the above drawbacks of currently known systems, by providing a method as described hereinbefore, which further provides the following steps:

a) transmitting a signal by the transmitting mobi1e device ;

b) receiving the transmitted signal by a plurality of stations reached by the signal and for each station calculating a parameter of the distance from the mobile device, which distance parameter is obtained from the reception characteristics of the signal by the station; c) selecting a reference station among the stations reached by the signal, which reference station is identified as the closest to the mobile device on the basis of the distance parameter;

d) selecting a set of stations, among the stations reached by the signal, which set comprises the reference station and three or more stations closest to the reference station;

e) identifying a plurality of clusters corresponding to all the possible combinations of at least three stations of the set of stations;

f) for each cluster calculating the position of the mobile device on the basis of the distance parameter of each station of the cluster;

g) calculating the position of the mobile device by working out the mean of the positions calculated for each cluster.

The signal transmitted from the mobile device is of the radio-frequency type and it is possible to use for the communication one of the technologies and of the frequency bands described above.

The distance parameter or ranging is calculated from the received signal and it can be directly calculated by the transceiver stations.

The position of the stations is known a priori, it being calculated upon distributing the stations, therefore the stations that are closer to the reference station are automatically defined once the reference station is identified, the distances among the stations being already known or it being possible to directly calculate them from the positions.

The high accuracy of the locating method is guaranteed by the selection of the reference station and by the division into clusters of the set of stations that have been considered, which clusters are used for having the partial localization estimates whose mean is finally worked out to obtain the final position of the mobile device.

The clustering strategy is based on the Euclidean distance of the stations from the reference station. The goal is not to involve in the locating process all the stations interrogated by the mobile device, but only the set comprising the stations considered as the most suitable for estimating the position of the mobile devices. The process selecting the set therefore occurs by means of the distance parameter and the Euclidean distance among the stations.

For each cluster the calculation of the position of the mobile device can take place by any currently known method, preferably by trilateration or multilateration .

In one preferred embodiment the environment is divided in sub-areas and the set of stations is selected only among the stations reached by the signal comprised within the same sub-area of the reference station .

This is very advantageous since it allows the sub- area where the mobile device is located to be identified, by selecting the reference station and therefore the reference sub-area. The analysis of the position, based on the detections of the other stations of the set, is therefore limited to the identified sub- area. This allows for example a room or a space to be considered, without involving the stations of other rooms in the calculation of the position, that is corresponding to different sub-areas, which stations can be also closer the reference station, but whose detections can be altered by the presence of walls or columns. The process selecting the reference station results in the non-ambiguity of the sub-areas of the indoor environment to be monitored.

The clusters can comprise any number of stations between three, that is the least number necessary for carrying out the localization, and the total number of stations of the set. Experimental tests have shown that a set composed of four stations divided into clusters composed each one of three stations provide the best results as regards locating accuracy.

In one embodiment, if the stations reached by the signal and within the same sub-area as the reference station are less then three, three stations reached by the signal are selected on the basis of the distance parameter and the position of the mobile device is calculated on the basis of the distance parameter of each station.

Thus, if the number of stations reached by the signal in the reference sub-area is lower than the least number necessary to carry out the localization, the method performs an exception to the fact of limiting the analysis to the reference sub-area, and it considers the stations more suitable for the localization that belong also to other sub-areas. In a first embodiment the distance parameter is the strength of the received signal (RSS- Receive Signal Strength) . Thus the ranging is calculated by measuring the strength of the signal received by the stations. The advantage of such type of calculation is the immediate implementation on any radio-frequency device .

According to one improvement the strength values of the received signal are weighted by the standard deviation of a range of signal strength values related to the distance, which standard deviation is calculated beforehand. The standard deviation can be plotted by experimental tests where several samples of the strength of the received signal at a plurality of known distances are acquired. The weighting of the values of the received signal strength preferably is the division of the detected value by the standard deviation, to obtain a more reliable value of the received signal strength .

In one variant embodiment, the distance parameter is the time of flight of the signal (ToF) . In this case the system uses the time of flight the radio-frequency message takes starting from the mobile device to the station or vice versa. With respect to the received signal strength, such mode is less sensitive to the obstacles between transmitter and receiver, and it has high locating accuracies, even if it requires ad hoc hardware for accurately managing the synchronization between the mobile device and the stations in order to obtain a homogeneous time reference. It is possible to provide also the values of time of flight of the signal to be weighted by the standard deviation. Experimental studies however have confirmed that such procedure does not considerably improve measurement accuracy, since the TOF method is already reliable enough.

According to a particularly advantageous improvement it is possible to add one or more further stations, said further station being used a single time as the transmitting mobile device for calculating the position of the further station, and the further station being used after said calculation of the position as a transceiver station. Therefore the further stations firstly are placed and used as transmitting mobile devices, their position is defined, which position therefore is known to the system, and finally they are converted into simple transceiver stations, starting to operate in the system like other stations. Thus the system gains an intrinsic scalability and it can be enlarged in any moment as one desires, for example when the environment to be monitored is subjected to a change in the shape and to an increase in the dimensions. Such characteristic is particularly usable in combination with the ToF procedure.

According to another variant embodiment, the distance parameter is the angle of arrival (AoA) , that is the angle where the signal arrives to the station. In this case the locating accuracies are better than systems using RSS but are lower than systems using ToFs . Moreover such mode is sensitive to the possible presence of obstacles between the transmitter and the receiver, and it requires an array of antennas on the station to extract information about the angle of arrival of the radio-frequency message.

The present invention further relates to a system for real-time locating inanimate objects and/or persons and/or animals within an environment, comprising one or more transmitting mobile devices, each mobile device being associated to a individual object or person or animal, a plurality of transceiver stations arranged in the environment in predetermined positions and a monitoring unit in communication with said transceiver stations. The monitoring unit can also communicate with the mobile device. The mobile device is intended to transmit a signal, the stations reached by the signal are intended to receive the transmitted signal, and the monitoring unit is intended to calculate the position of the mobile device. Such calculation provides: to select a reference station among the stations reached by the signal, on the basis of the reception characteristics of the signal; to select a set of stations, among the stations reached by the signal, which set comprises the reference station and three or more stations closest to the reference station; to identify a plurality of clusters corresponding to all the possible combinations of at least three stations of the set of stations; for each cluster to calculate the position of the mobile device on the basis of the reception characteristic of the signal of each station of the cluster; to calculate the position of the mobile device by working out the mean of the positions calculated for each cluster.

In one embodiment the mobile device and the stations are of the Ultra Wide Band (UWB) type. The use of the UWB technology in combination with the system and method shown above allow the locating accuracy to be less than 30 cm, reaching also few cm in some cases.

In a further embodiment the stations are arranged each one every 50 m² of the environment.

By the characteristics described above the system allow the presence of several operators/tools changing in number over time to be dynamically managed and it is able to provide the detected position with a high accuracy. The method guarantees high locating accuracies since it does not perform a simple trilateration on the stations closest to the mobile device, but is allows the triads of stations more suitable for locating the mobile device to be selected by implementing statistical and heuristic methods. The result is a drastic reduction in the possibility of errors on the position of the mobile devices.

A further strong point is the possibility of using the algorithm with any type of physical level of wireless sensor network (WSN) .

The accuracies obtained by the system of the present invention pave the way to different application scenarios such as the management of emergencies in complex environments, increase of safety/security in building yards, checking children in crowded areas, locating hospital apparatuses, tracking of moving vehicles such to avoid collisions, tracking patients under risk of escaping from care facilities, searching for products in supermarkets, searching for counters in public offices, etc.

As regards safety applications the system provides a real help to people having to operate after an accident, in order to provide information about who is still within a critical area to be evacuated. Thus rescuers can focus the efforts on well-defined areas, without wasting energies, time and without exposing rescuers safety to a risk in case of useless searches, besides, obviously, speeding up rescues for people in need.

Besides "safety" applications there are "security" applications, those applications where it is necessary to know the position of objects and/or people such to prevent them from entering not authorized areas or under potential dangerous situations if particular activities, with a high risk component, are temporarily taking place in specific areas.

Further advantages of the system are the possibility of providing wearable transmitting mobile devices with low power consumption, and the easiness of installing them without the need of specialized technical personnel.

These and other characteristics and advantages of the present invention will be more clear from the following description of some embodiments shown in the annexed drawings wherein: Fig.l is a diagram of the system;

Fig.2 is a possible graphical interface of the system.

Figure 3 is a flow diagram of the general principle of the present invention.

Figure 4 is a flow diagram of a tracking method comprising the method according to figures 1 to 3.

Figure 5 is a flow diagram of one embodiment of the method for converting the digital map in order to differentiate admissible and inadmissible areas.

Figure 6 is a flow diagram of one example for determining the start and final variables of the path to be tracked.

Figure 7 is a flow diagram of one example for carrying out the run time phase providing a check at predetermined intervals and with predetermined criteria of the tracking positions determined by inertial and Particle filter algorithms.

Figure 1 shows a diagram of the real-time location system of the present invention. The system provides in real-time the position of one or more persons 10 within an environment 9.

The system comprises one or more transmitting mobile devices 1, each mobile device 1 being associated to an individual person 10, a plurality of transceiver stations 2 placed in predetermined positions within the environment 9 and a monitoring unit 3 in communication with the transceiver stations 2.

The environment 9 is divided into sub-areas 90, 91, 92 that can be reached by persons 10. The mobile device is intended to transmit a radio- frequency signal; the transmission of the signal can be continuous or it can be timed at suitably dimensioned intervals .

Stations 2 reached by the signal perform a calculation of the received signal strength (RSS) or of the time of flight (ToF) .

Stations 2 are arranged in a number of at least three for each sub-area 90, 91, 92 preferably one station every 50 m² of the environment.

The monitoring unit 3 comprises a gateway 30 for the communication with the stations 2, a calculation unit or location engine 31 and a graphical interface 32.

The ToF/RSS collected by the stations 2 are directed through the gateway 30 to the location engine 31 containing the algorithm for estimating the position of the mobile devices to be monitored.

The location algorithm can be therefore summarized in the following steps:

1. Among the N ToF/RSS received at time t, the lowest/highest one is selected. This means the station 2 being the one closest to the mobile device 1 and therefore this means a more reliable measurement.

2. Selecting the station 2 that has measured the

ToF/RSS selected at point 1. Such station is defined as reference station 2' or ANref . In the figure the reference station ANref , denoted as 2' , clearly is the station closest to the mobile device 1.

3. Selecting the sub-area 90 where AN_ref is installed. Such area is defined as S* . This means that among all the received ToF/RSS only the ones related to the stations of that sub-area 90 are selected. They relate to the stations that are closest to the mobile device and are considered as more reliable than the possible other received ToF/RSS.

4. If among the ToF/RSS received at time t there are also those related to stations in S* or at least in the least number for estimating the position of the mobile device, n denotes the number of these ToF/RSS and the algorithm goes on at points 4-5-6-7-8. Otherwise it moves to step 9.

For each station related to the sub-set n of ToF/RSS the Euclidean distance with AN_ref is calculated such as shown in Eq. (1) :

n. (1) where x_Rp , y_Rp and z_Rp are coordinates of AN_ref , X_j , i and Z_j are coordinates of the stations in S* related to the sub-set n of ToF/RSS. The distance between the reference station 2' AN_ref and the other stations 2 of the sub-area 90 is shown in the figure by broken lines.

5. If n > 3 the sub-set M of ToF/RSS is generated by selecting those related to the three stations whose Euclidean distances with AN_ref are the smallest. A number of ToF/RSS higher than 3 is obtained on which the estimation of the mobile device position is based. Obviously if n = 3 the clustering process based on the Euclidean distance is not applied since we have the least number of ToF/RSS to estimate the position of BN by trilateration/optimization (ToF/RSS related to AN_ref+ 2 ToF/RSS related to stations with the smallest Euclidean distance) .

6. If n>3, M is involved in a clustering process such as shown in Eq. (2) :

3< k < M+l. (2) where k is the dimension of the cluster, M + l has to be at least equal to 4 such to have g ≥ 4.

The dimension of the cluster therefore can be selected as one desires. The best experimental results are obtained with k = 3.

7. For each cluster calculated at step 6 the position of the BN is calculated, minimizing the cost function shown in Eq. (3) :

Jv —∑i=i(^i ^xv,i ^x0v,i) Ύν,ϊ Ύθ,ν,ί) (^zv,i ^z0,v,i)²)²

v=l,... ,g

(3) where xov,±, yov,± and zov,± are the coordinates of the i-th station of the v-th cluster, Xv,±, y ,i e z_Vf± are the coordinates of the estimate of the position of the mobile device carried out by the v-th cluster and R± is the value of the i-th ranging calculated on the basis of the i-th ToF/RSS .

8. As the estimate of the position of the final BN s (x,y,z) the mean of the estimates of the positions of each cluster calculated in step 7 is considered, such as shown in Eq. (4) :

(4)

9. If, on the contrary, among the ToF/RSS received in step 4 there is not the sufficient number (at least 3) of those related to stations in S*, a normal minimization process is carried out such as shown in Eq. (3) where the 3 smallest/highest received Tof/RSS are used. In this case no clustering is implemented based on the Euclidean distance.

Figure 2 shows a graphical interface 32 for a possible application of the system. The management of the evacuation in a building yard is shown. The rectangle 93 on the right is the assembly point the workers have to reach in case of evacuation in the building yard in emergency situation. Some markers 1' , that for example can be green coloured, are the workers that have reached the assembly point 93. The remaining mobile devices, that have not reached the assembly point 93 yet, are represented by markers 1' ' that can be red coloured. The infrastructure network, composed of the stations 2 whose position is accurately known, is represented by markers that for example can be grey coloured.

It is possible, by the system, to count how many workers have reached the assembly point (green markers) and to known how many persons are still in the building yard to be evacuated (red markers) . All this allows rescuers to be directed towards said sectors without being dispersed in the whole building yard, thus increasing the possibility of rescuing persons still alive within a short time.

In the example of the figure, when a user of the graphical interface selects a mobile device corresponding to a worker the name of the worker and his/her position is displayed. The positions of the stations 2 and the list of the workers, divided into workers already at the assembly point 94 and workers still to be rescued, are visible on the left.

With reference to figure 3 it summarizes the main steps of the method according to the present invention: step 300 provides a plurality of transceiver stations 2 arranged in the environment at predetermined positions and a monitoring unit 3 in communication with said stations 2 to be provided.

Step 310 provides a transmitting mobile device 1 to be provided that is to be associated to at least one or to each one of the following entities: inanimate objects and/or persons (10) and/or animals that are situated and/or move within an environment (9) .

At least one of the mobile devices transmits a localization signal 320 that is received and processed by one or more of the stations 2 as shown in step 330. Each station calculates 340 a parameter of the distance from the mobile device 1, which distance parameter is obtained from the reception characteristics of the signal by the station 2.

Step 350 provides the selection of one reference station 2', among the stations 2 reached by the signal, which reference station 2' is identified as the closest one to the mobile device 1 on the basis of the distance parameter .

Step 360 provides the selection of a set of stations 2, among the stations 2 reached by the signal, which set comprises the reference station 2' and three or more stations 2 closest to the reference station 2' ;

At step 370 a plurality of clusters corresponding to all the possible combinations of at least three stations 2 of the set of stations is identified and at step 380 the calculation for each cluster of the position of the mobile device 1 is performed on the basis of the distance parameter of each station 2 of the cluster as well as in 390 the calculation of the position of the mobile device is performed by working out the mean of the positions calculated for each cluster .

According to one embodiment, the real-time location method and system according to the present invention can be used in a navigation system and method wherein at least the starting position of the user and/or also at least the position of a destination or target is determined by the real-time location method and system of the present invention. As it will be noted below the navigation method and the corresponding system further have at least one or more further steps for determining the position of the user when moving along the path towards the destination or target in different moments of the displacement along a path and/or in different points of said path.

The number of such steps is variable and it is defined beforehand in a fixed manner or it is determined by providing a given predetermined number of steps for detecting the position and also by a given number of additional steps for determining the position that are activated depending on the occurrence of one or more predetermined conditions detectable by the system.

In one embodiment, the tracking of the position of the user along the path occurs by a combination of position detecting steps by the method and system of the present invention described above and a combination of steps tracking the displacement of the user that are performed by using a A* algorithm (A-star) in combination with Pedestrian detection and particle filter algorithms.

The described algorithms are algorithms known for example, for algorithm A star, from: https://en.wikipedia.Org/wiki/A* search algorithm;

For particle filter algorithms from: IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 50, NO. 2, FEBRUARY 2002, A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking, M. Sanjeev Arulampalam, Simon Maskell, Neil Gordon, and Tim Clapp;

For Pedestrian detection algorithms from Pedestrian tracking using inertial sensors, JOURNAL OF PHYSICAL AGENTS, VOL. 3 NO. 1 JANUARY 2009, Raul Feliz, Eduardo Zalama, Jaime Gomez Garcia-Bermejo .

The combination of the tracking by one or more of said algorithms used in combination with each other and the RTLS of the present invention described above can be implemented in dedicated devices (generally called as TAGs) or commercially available devices (smarthphones or tablets) .

The problem faced in this case is to define a "shortest path". The shortest path is defined as the set of points minimizing the distance between a general starting point start (x,y) and a final destination point final (x,y) .

At least the estimated starting position and/or also at least the final position is determined by the method and system described above with reference to figures 1 to 3. The starting position can be used as the starting point for the user to navigate towards a destination point. It is possible to have two different application scenarios:

a. safety: the position of the rescuer/user is the starting point, the position of the person potentially at risk is the destination/target point. The position of the user/rescuer and/or the position of the destination/target are determined, when possible, by the system according to the present invention and the rescuer displays on the smartphone or tablet the shortest path to reach the destination;

b. consumer/entertainment : the user position is the starting point and the destination desired to be reached is the destination point (for example the department inside a hospital, an office inside an administrative building etc..) . The user displays on the smartphone or tablet the shortest path to reach the destination point.

The navigation occurs in real-time, by correcting and updating the shortest path depending on the position of the user/rescuer at each moment t.

In one embodiment of the navigation method shown also in the flow chart of figure 4 and using the method of the present invention, the following steps are provided:

i) providing a digital map of the site or area containing the path, that is the starting point and the destination point, in the form of an array of pixels, step 400;

ii) converting the individual pixels such that they take only values 0 and 1 according to a specific criterion, step 410;

iii) defining at least a starting point and/or at least a point of arrival by means of the real-time location system (RTLS) according the characteristics described above; step 420;

iv) calculating the shortest path between the starting point and the destination point by a so called A* algorithm (A star algorithm) : step 430; v) during the displacement along the path, at each moment t, tracking the position of the user along the path by means of inertial and statistical algorithms; vi) defining each new position of the user determined at moment t as a new starting position of the path of the user and calculating a new shortest path by means of the A* algorithm having said new position as the starting point;

vii) repeating steps v) and vi) till reaching the destination point or a position near the destination point according to a threshold of maximum distance from said destination point beyond which the destination point is considered as not being reached.

In one embodiment, the conversion of the digital map occurs by finding obstacles or areas inaccessible by the path, that is areas where the presence or the passage of the user is inadmissible and by means of markers differentiating the pixels of the digital map corresponding to such areas from pixels corresponding to the areas of the map where the passage or presence of the user is possible.

Therefore the obstacles are defined such as walls and/or areas where the tracked user cannot be present and where therefore the locating method cannot estimate the user position. These inadmissible areas are also areas where the shortest path CANNOT pass through. The admissible areas where the user can be present or can move are those areas where the method can estimate the user position and where the shortest path can pass through. According to one embodiment, the markers that di erentiate the admissible areas from the inadmissible areas are composed for each pixel in the digital map of a parameter that can take only two values such as for example 0 and 1 and the conversion of the digital map therefore provides a numerical matrix wherein each element is a pixel of the digital map and wherein each element can take a value 0 or 1 depending on whether the area is considered admissible or inadmissible.

One embodiment of the conversion process provides the following steps:

111) the general colour image of the map being defined as S, it is converted into a black and white image I: step 500 and 510.

112) defining as pi (x,y) each pixel of I as the triad of values (x, y, z) , where the pair (x,y) identifies the coordinate of the pixel and z is the RGB value taken by the pixel in the coordinate (x,y) : step 520.

113) defining a threshold value th that identifies the RGB value such that pi(x,y) is considered as an obstacle or admissible area, in this manner:

a. if pi(x,y) >= th then it is a inadmissible area or obstacle;

b. if pi(x,y) < th then it is admissible area:

step 530.

114) for each pixel pi(x,y) of I:

a. if pi(x,y) is inadmissible area or obstacle then pi(x,y) = 1; b. if pi(x,y) is admissible area then pi(x,y) = 0; step 540

ii5) providing the matrix I composed only of 1 and 0 to the A* algorithm for calculating the shortest path: step 550.

According to one embodiment, the determination of starting point or initial point and of the destination or target point for A*algorithm provides the following steps :

iiil) associating to the input variable first_start (x,y) the user position provided by following the steps 300 to 390 of the location method according to the description above at time tO (starting/initial time) step 600;

iii2) checking that start (x,y) is within an admissible area by using the matrix I: step 610;

iii3) repeating step iiil) if condition iii2) is not met and repeating the loop till meeting such condition ;

iii4) associating to the input variable final (x,y) the destination point, step 620;

1115) checking that final (x,y) is within an admissible area by using the matrix I step 630;

1116) repeating step iii4) if the condition iii5) is not met and repeating the loop till meeting such condition ;

1117) providing said variables to the A* algorithm for calculating the path, step 640.

According to a characteristic, the A* algorithm determines the shortest path by evaluating only the points considerd as admissible areas to prevent the path from passing through walls, barriers or other obstacles preventing the user from passing or being situated therein.

The process for determining the shortest path is shown in details in the following document https://en.wikipedia.Org/wiki/A* search algorithm that is a part of the present description.

The maximum resolution of the shortest path that can be obtained is the density itself of Ι(χ,γ) that is one point for each ρί(χ,γ) .

According to one embodiment, for the calculation of the travelled tracked path, that is the displacement of the user along the path and the estimate of his/her position in real-time during the displacement along the path the invention provides to use a combination of pedestrian algorithm and an algorithm called as

Particle filter.

The so called pedestrian algorithm or inertial algorithm uses the Inertial Measurement Unit (IMU) hardware present in the mobile device of the user for example a TAG or smartphone.

The Particle Filter algorithm allows the accuracy of the user position localization at each moment t to be improved according to obstacles and admissible areas.

By the combined action of Pedestrian algorithms and Particle Filter algorithms it is possible to update and track the displacement of the user with no need of using the localization method described above of the present invention that anyway requires a given number of anchor nodes (AN) that therefore can be reduced.

In particular, according to an improvement, the invention provides a given predetermined number of anchor nodes arranged along the path and intended to provide information about the position of the user tracked at predetermined space and/or time intervals along the path and/or in the duration of the displacement, while in the intermediate time moments between said predetermined time moments and/or in the displacement intermediate positions between the positions defining the space intervals, the user position is determined by using the combination of pedestrian and particle filter algorithms.

Still according to an improvement, in the positions defining the space intervals and/or at the predetermined moments defining the time intervals, the position is determined both by the localization method of the present invention by using anchor nodes , and by tracking operation with the combination of pedestrian or inertial algorithm and particle filter algorithm and a comparison of the point coordinates determined by the two methods is perfomed, since as the new starting point for the calculation of the remaining shortest path by the A* algorithm the corrected coordinates of said point are used based on said comparison or possibly the coordinates determined by using the anchor nodes replacing those determined by the combination of Pedestrian or inertial algorithms and the Particle filter are used. As it is clear from what mentioned above, the use of inertial algorithms that use accelerometers , magnetometers and gyroscopes allow a quite spread distribution of ANs (Anchor nodes) to be provided, namely a minimization of the distribution of ANs.

The use of the localization system based on signals of anchor nodes is only used in a first startup phase of the system where the first localization of the user occurs, defining the input variable first_start (x,y) .

It is also possible to provide a second phase called as reset phase taking place when the TAG, smartphone or tablet records a RSSI or TOF exceeding a specific predetermined threshold, called as alert_threshold. According to one example the Pedestrian and Particle Filter algorithms are used only in the run-time phase.

The run-time phase provides the following steps: 2. the run-time phase is present when t0<=t<=t0+n, where n is the time necessary for the reset phase to be present .

a. if a RSSI or TOF are different from the alert threshold of alert_threshold, particularly if RSSI is < said threshold and TOF is > said threshold then the run-time phase is followed and the new position is calculated by means of inertial algorithms using the user acceleration and orientation with respect to the magnetic north.

It is possible to choose among different inertial or pedestrian algorithms such as for example one or a combination of the following algorithms described in the following documents:

2015 IEEE Tenth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP) Singapore, 7-9 April 2015, "An Enhanced Pedestrian Dead Reckoning Approach for Pedestrian Tracking using Smartphones";

2013 International Conference on Indoor Positioning and Indoor Navigation, 28-31st October 2013, "Particle filter and smoother for indoor localization" ;

2013 International Conference on Indoor Positioning and Indoor Navigation, 28-31 th October 2013, "Pedestrian Dead Reckoning Using Adaptive Particle Filter to Human Moving Mode";

IEEE SENSORS JOURNAL, VOL. 15, NO. 5, MAY 2015 SmartPDR: "Smartphone-Based Pedestrian Dead Reckoning for Indoor Localization";

"Low-Cost Bluetooth Foot-Mounted IMU for Pedestrian Tracking in Industrial Environments";

Mongolia University, Hohhot, 010021, China, "An Accurate Step Detection Algorithm Using Unconstrained Smartphones Xiaokun Yang and Baoqi Huang Inner " ;

- Dept. Pervasive Computing, Tampere University of Technology, Tampere, Finland. Dept. Biomedical Engineering and Computational Science, Aalto University, Espoo, Finland, "PEDESTRIAN LOCALIZATION IN MOVING PLATFORMS USING DEAD RECKONING, PARTICLE FILTERING AND MAP MATCHING"; 2015 International Conference on Indoor Positioning and Indoor Navigation (IPIN) , 13-16 October 2015, Banff, Alberta, Canada, "Smartphone-based Indoor Positioning Utilizing Motion Recognition".

At each new predetermined moment t a position of the user is determined which is the start position for determining the shortest path on the remaining part of the path .

The algorithms mentioned above allow the steps made by the user to be counted and the relevant travelled distance to be calculated. Thus it is possible to perform an accurate tracking of the user and there is no need of calculating a new first_start (χ,γ) at each moment t by using the reception signals of the anchor nodes.

The position determined by inertial algorithms start (x,y) is corrected or it is kept as valid by considering the obstacles and the admissible areas as it follows:

· if start (x,y) is not within an admissible area, then start (x,y) is corrected by moving the point within the admissible area closest to start (x,y) .

This correction process is performed by using the Particle Filter.

If start (x,y) is within an admissible area, then start (x,y) is considered as valid and it is used for calculating again the shortest path by the A* algorithm.

If RSSI or TOF are lower or higher than the alert threshold respectively, then the reset phase is followed and the user position is calculated on the basis of the method according to the present invention by using the available anchor nodes. Such value being used as the new variable first_start (x,y) for the calculation of the shortest path on the remaining path part .

At the same time the errors accumulated by the Pedestrian and Particle Filter algorithms are cleared and such algorithms are used for a new runtime phase.

Figure 7 shows an example of the method steps providing to check the positions determined by inertial algorithms and by particle filter algorithm.

At step 700 a time interval and/or a space interval during the displacement of a user along the path is provided;

At the moments or positions corresponding to the final and/or starting point of each time and/or space interval determining the user position by the method according to steps 300 to 390: step 710.

determining the user position by means of inertial and Particle filter algorithms at the time moments intermediate between the starting and final moments of the time intervals and/or at the positions intermediate between the starting and/or final points of the space intervals: step 720.

An alert threshold is defined with which the values of RSSI or TOF are compared for checking whether said threshold is exceeded: step 730.

Step 740 defines whether the current position determined by inertial algorithms falls within an admissible area or not. If this situation occurs and therefore the detected position is within an inadmissible area, then such position is modified by moving the coordinates of the position to the closest admissible area and by using such point as the starting point for calculating the remaining shortest path up to the destination point: step 760.

If the values of RSSI or TOF exceed the threshold, step 770, then the current position is determined on the basis of the detections by means of anchor nodes and according to steps 300 to 390, such position replaces the one determined by the inertial algorithm and it is used for the step 760 determining the remaning shortest path up to destination. The described steps are obviuously repeated till reaching the destination or till ending the process for other reasons .

The invention also relates to a system for performing said method comprising a processing unit provided in the mobile device, one or more inertial sensors detecting the orientation and movement accelerations of the user that is of the mobile device, transmitting and receiving unit, a memory for storing data and programs . A program implementing the inertial algorithms and for the comparison, a control program which programs are executable by the processor; at least one user interface system both of the speech type, touch type, by means of buttons or other input members for example a touch screen with a graphical interface.

Claims

1. Method for real-time location of inanimate objects and/or persons (10) and/or animals within an environment (9) , each object or person or animal being associated to a transmitting mobile device (1) , there being provided a plurality of transceiver stations (2) placed in predetermined positions within the environment and a monitoring unit (3) in communication with said stations (2) ,

characterized in that

it provides the following steps:

a) transmitting a signal by the mobile device (1) ; b) receiving the transmitted signal by a plurality of stations (2) reached by the signal and for each station calculating (2) a parameter of the distance from the mobile device (1) , which distance parameter is obtained from the reception characteristics of the signal by the station (2) ;

c) selecting a reference station (2') among the stations (2) reached by the signal, which reference station (2') is identified as the closest to the mobile device (1) on the basis of the distance parameter;

d) selecting a set of stations (2) , among the stations (2) reached by the signal, which set comprises the reference station (2') and three or more stations (2) closest to the reference station (2');

e) identifying a plurality of clusters corresponding to all the possible combinations of at least three stations (2) of the set of stations; f) for each cluster calculating the position of the mobile device (1) on the basis of the distance parameter of each station (2) of the cluster;

g) calculating the position of the mobile device (1) by working out the mean of the positions calculated for each cluster.

2. Method according to claim 1 , wherein the environment (9) is divided into sub-areas (90, 91, 92) and the set of stations (2) is selected only among the stations (2) reached by the signal comprised within the same sub-area (90, 91, 92) of the reference station (2' ) .

3. Method according to claim 2, wherein if the stations (2) reached by the signal and comprised within the same sub-area (90, 91, 92) of the reference station (2') are less than three, three stations (2) reached by the signal are selected on the basis of the distance parameter and the position of the mobile device (1) is calculated on the basis of the distance parameter of each station (2) .

4. Method according to one or more of the preceding claims, wherein the distance parameter is the strength of the received signal .

5. Method according to claim 4 , wherein the values of the strength of the received signal are weighted with the standard deviation of a range of strength values of the signal in relation to the distance, which standard deviation is calculated beforehand.

6. Method according to one or more of the preceding claims wherein the distance parameter is the time of flight of the signal .

7. Method according to one or more of the preceding claims wherein it is possible to add one or more further stations (2) , said further station being used a single time as the transmitting mobile device for calculating the position of the further station, and the further station being used after said calculation of the position as a transreceiver station (2) .

8. System for real-time location of inanimate objects and/or persons (10) and/or animals within an environment (9) , comprising

one or more transmitting mobile devices (1) , each mobile device (1) being associated to a single object or person (10) or animal,

a plurality of transreceiver stations (2) placed in predetermined positions within the environment (9) , a monitoring unit (3) in communication with said stations (2) ,

characterized in that

the mobile device (1) is intended to transmit a signal, the stations (2) reached by the signal are intended to receive the transmitted signal and to calculate for each station (2) a parameter of the distance from the mobile device (1) , which distance parameter is obtained from the reception characteristics of the signal by the station (2) , and the monitoring unit (3) is intended to calculate the position of the transmitting mobile device by carrying out: a selection of a reference station (2') among the stations (2) reached by the signal, on the basis of the distance parameter;

a selection of a set of stations (2) , among the stations (2) reached by the signal, which set comprises the reference station (2') and three or more stations (2) closest to the reference station (2');

the identification of a plurality of clusters corresponding to all the possible combinations of at least three stations (2) of the set of stations;

the calculation for each cluster of the position of the mobile device (1) on the basis of the distance parameter of each station (2) of the cluster;

the calculation of the position of the mobile device (1) by working out the mean of the positions calculated for each cluster.

9. System according to claim 8, wherein the mobile device (1) and the stations (2) are of the Ultra Wide Band (UWB) type.

10. System according to claim 8 or 9, wherein the stations (2) are distributed one every 50 m² of the environment .

11. Method according to one or more of the preceding claims 1 to 7, wherein the real-time localization is used for determining at least the starting position of the user and/or at least also the position of a destination or target to be provided as starting point and/or destination point of a navigation system.

12. Method according to claim 11 wherein there are further provided one or more further steps for determining the position of the user when moving along the path towards the destination or target in different moments of the displacement along a path and/or in different points of said path.

13. Method according to claim 11 or 12, wherein the tracking of the position of the user along the path occurs by a combination of steps detecting the position according to one or more of the claims 1 to 7 and a combination of steps tracking the displacement of the user performed by using a A* algorithm (A-star) in combination with Pedestrian detection and particle filter algorithms.

1 . Method according to one or more of the claims 11 to 13, wherein the following steps are provided:

i) providing a digital map of the site or area where the path is provided, that is the starting point and the destination point, in the form of an array of pixels ;

ii) converting the individual pixels such that they take only values 0 and 1 according to a specific criterion ;

iii) defining at least a starting point and/or at least a point of arrival by the real-time location system (RTLS) according to one or more claims 1 to 7; iv) calculating the shortest path between the starting point and the destination point by a so called A* algorithm (A star algorithm) ;

v) during the displacement along the path, at each moment t, tracking the position of the user along the path by inertial and statistical algorithms;

vi) defining each new position of the user determined at moment t as a new starting position of the path of the user and calculating a new shortest path by the A*algorithm having said new position as the starting point;

15. Method according to claim 14, wherein the conversion of step ii) provides the following steps:

111) the general colour image of the map being defined as S, it is converted into a black and white image I ; .

112) defining as pi (x,y) each pixel of I as the triad of values (x, y, z) , where the pair (x,y) identifies the coordinate of the pixel and z is the RGB value taken by the pixel in the coordinate (x,y) ;

113) defining a threshold value th that identifies the RGB value such that pi(x,y) is considered as an obstacle or an admissible area, in this manner:

a. if pi(x,y) >= th then it is a inadmissible area or obstacle;

b. if pi(x,y) < th then it is admissible area:

114) for each pixel pi(x,y) of I:

a. if pi(x,y) is inadmissible area or obstacle then pi(x,y) = 1;

b. if pi(x,y) is admissible area then pi(x,y) = 0;

ii5) providing the matrix I composed only of 1 and 0 to the A* algorithm for calculating the shortest path .

16. Navigation method, characterized in that it comprises the steps according to one or more of the preceding claims .