FI121441B - A method for indoor navigation and positioning - Google Patents

Description

INDOOR NAVIGATION AND LOCATION METHOD

The invention relates to indoor navigation and positioning, in particular to a method for locating an object moving indoors.

5

BACKGROUND OF THE INVENTION

Locating a moving object indoors very accurately is a challenging task. Current technologies are mainly divided into three groups: (1) Global Navigation Satellite 10 System (GNSS) indoor positioning technologies, (2) cellular network-based positioning technologies, and (3) short-range RF (Radio Frequency) positioning technologies.

GNSS indoor positioning techniques include highly sensitive GPS (Global Positioning System) receiving technology and pseudolite technology (Lachapelle, 2005. GNSS Indoor 15 Location Technologies. Journal of Global Positioning Systems. Vol. 3, No. 1-2, pp. 2-11). . The GPS signal is a relatively weak signal and is greatly attenuated after passing through obstacles such as walls. Therefore, GPS usually operates outdoors in the sky, as the receiver needs a direct line of sight to the GPS satellites. However, a very sensitive GPS receiver can receive and monitor very weak GPS signals indoors. However, in indoor positioning, the positioning accuracy decreases from 5 to 10 meters for outdoor positioning, depending on the material of obstacles such as walls and ceilings. Positioning accuracy is greatly reduced because the remaining signal after passing obstacles is too weak. One way to increase signal strength is to install pseudolites indoors where navigation is needed. A pseudolite is a device that transmits 25 GPS-like signals. The GPS receiver tracks the GPS-like signal transmitted by the pseudolite and uses distance measurement between the pseudolite and the user's receiver to determine the antenna location of the user's receiver. The positioning principle here is the same as for GPS positioning. The only difference is that the pseudolite is located in a fixed location (relative to the user), while the GPS satellite 30 moves in space. The distance between the pseudolite and the user's receiver varies from a few meters to a few kilometers, while the distance between the GPS receiver and the GPS satellite is about 20,000 kilometers. When using pseudolite, the positioning accuracy is about 5 to 10 meters, which is about the same as outdoor 2 GPS positioning. The largest source of error that prevents better positioning accuracy is due to signals reflected from walls and other objects.

The most common measurements of cellular location based techniques are Time of Arrival (TOA), Time Difference of Arrival (TDOA), Angle of Arrival (AOA), Cell ID (Cell ID) and Signal Levels (Chen, 2002). Integration of Satellite and Cellular Positioning Technologies, Proceedings of the ION National Technical Meeting, January 28-30, 2002, San Diego, CA, pp. 312-322). The location of the user may be determined by the term "within this cell", when the wireless access point is known, or by TOA, TDOA, or AOA measurements. The accuracy of cellular network-based positioning accuracy from tens of meters to tens of kilometers depends on the application of positioning technology.

Short range RF positioning technologies include wireless local area network (Wi-15 Fi / WLAN), Bluetooth, RFID (Radio Frequency Identification) and non-standardized RF system. The RADAR system developed by Microsoft Research utilizes an existing RF network (Bahl and Padmanabhan, 2000, RADAR: An In-Building RF-based User Location and Tracking System. In Proc. IEEE INFOCOM, pp. 775-784, Tel-Aviv , Israel, March 2000). It uses 20 RF signals as an indicator of the distance between the transmitter and the receiver. This distance information is then used to locate the user through triangulation. The MIT (Massachusetts Institute of Technology) has implemented a system called the Cricket Location System for Mobile Device Monitoring (Smith, Balakrishnan, Goraczko, and Priyantha, 2004, Tracking Moving Devices with a Cricket Location System). and Services (MOBISYS), pp. 190-202, Boston, MA, USA; or htto: //cricket.csail.mit.edu). The system comprises positioning transmitters (beacons) mounted on the ceiling of the building and receivers, called listeners, which are attached to the devices to be located. Each directional transmitter periodically transmits its position information as an RF message. Simultaneously, transmitter 30 also transmits an ultrasonic pulse. Listeners listen for directional transmitter messages and measure distances to nearby transmitters and use these distances to calculate their own positions.

3

The use of WLAN access points is also common in locating a mobile device. When the base stations are known, the mobile device locates when it contacts the base station in its environment. The method corresponds to a cell-ID solution for cellular networks, but with shorter base station distances, the positioning result is also more accurate.

5

A similar approach can be applied to RFID, Bluetooth and Infrared technologies, eg Active Badge. The power level of the RF signal can also be used as an indicator of the distance between a mobile device and a reference station (e.g., a WI_AN base station). The position of the mobile device can be determined by triangulation. Another way is to register the signal strengths in advance in the operating environment and store them in a database; locating the mobile device by comparing current signal strengths with the strengths stored in the database.

Indoor navigation has two positioning practices, active and passive. In active case 15, the device to be located transmits an RF signal and the stationary devices receive the RF signal. In passive practice, the signals are transmitted by stationary transmitters (e.g., pseudolites) or space satellites (e.g., GPS satellites) and the device being located receives RF signals. It will be apparent that all current techniques require a process that involves transmitting and receiving an RF signal 20 at one end.

SUMMARY OF THE INVENTION

The present invention relates to a novel indoor positioning method based on laser 25 range camera technology. In contrast to the prior art, the solution of the present invention is not an RF solution. The system consists of a network of reference targets, a laser rangefinder and a user receiver. The method is capable of automatically locating a user within a network of reference sites with a distance camera. The positioning process is characterized in that it 1) uses a rangefinder to take an image and measure distances to all points in the image and also to the reference points of the reference objects; 2) analyze the resulting image and identify the reference objects, and finally 3) determine the location of the moving object based on the coordinates of the reference points and corresponding distance measurements.

4 When constructing the system, the reference object network is pre-installed eg on the ceiling or on a wall in the interior where the moving object is to be monitored. In this case, the reference is not at all an electronic device, but quite simply custom-made characters printed on paper or other materials. In this new way, no RF signal is transmitted.

The invention encompasses methods for identifying reference objects by image analysis; identify the object ID (Identification) using a specially designed bar code system and estimate the location of the mobile device using the Least 10 Squares method. By the method of the present invention, the location of a moving device can be determined by making distance measurements with respect to one or more reference objects.

In the following, the invention is illustrated by the drawings and examples, which are not intended to limit the invention in any way.

PATTERNING

Figure 1. A method for locating a moving object by means of a laser rangefinder 20.

Figure 2. Diagram of the positioning process

Figure 3. Design of reference objects with barcodes

Figure 4. Locating a moving object using a single reference object.

DETAILED DESCRIPTION OF THE INVENTION

The positioning method shown in Figure 1 of the present invention comprises the following steps.

- installing the reference objects at known points, step A in Figure 2.

The reference object is provided with specially designed barcodes used to identify the reference object ID.

acquisition of the image and performing distance measurements with respect to all points in the image, step B in Figure 2. Distance measurements with reference points are used to estimate 5 positions. The reference points may be the center of the reference object or the northern, southern, eastern or western points of the reference object, Figure 3.

- analyzing the obtained image and identifying the reference object using a specially designed bar code system, steps C and D in Figure 2.

determining the position of the moving object (user) using known coordinates of the reference object and corresponding distance measurements, step E in Figure 2.

10 The coordinates of the reference points are predefined before installing a network of reference points. It is desirable that at least three of the reference objects corresponding to the reference space are visible in each image obtained. Therefore, the distance between the reference points in the network depends on the angles of the distance camera.

15 As the user moves with his or her distance cameras around the room, the system continuously acquires images and determines distance measurements with reference points.

Upon receipt of the image, the system analyzes it and determines the identification of reference objects in the image. The reference object has a center filled circle surrounded by a central periphery having bar code sectors, and the outer periphery has north (N), south (S), east (E), and west (W) reference points, Figure 3. The reference is identified by the Fourier correlation method. conversion. It is identified by creating an intensity profile at the center of the reference object and converting it to a barcode. The barcode contains the reference object ID and two check bits.

25

After identifying the reference objects from the resulting image, the system determines the location of the user carrying the distance camera by utilizing distance measurements with reference points and predetermined coordinates of points using the Least Squares method as follows: Δχ = (ΑτΑ) -, Ατ1 where 30 6 Δχ = (x-x y0, z - z0) T and I ί ~ 1 Al _2 _2 λ Λ / ΐ \ Τ * - \ P -Po * P ~ Pq) '") P ~ Pq)? where (x, y, z) T is user (range camera) coordinates to be determined, (x0, yo, Zo) J is the approximate coordinates of the user's location that can be initially set to 5, p 'is the distance measurement from the distance camera to the center of the ith reference object, p'Q is the estimated distance, which is obtained from the formula

Po => / (* '"* o) 2 + (y - yo) 2 + (z1 - -o) 2> where (xl, y', z ') are the coordinates of the center of the i th reference object which were predefined before installing benchmarks indoors. Matrix A is defined as: 10 ax fcj q a2 b2 c2 xQ-x 'x0-x' Xo ~ x 'A =. . . with at = —-, bt = —-, and c, = —-::: Po Po Po _Pn ^ n Cn_ where n (n> 3j is the number of reference items shown in the current image).

15 If only one reference object is visible in the image, the system will attempt to determine distance measurements relative to the north (N), south (S), east (E), west (W), and center (O) points of the reference object (Figure 3), instead of defining them with respect to the centers of the various comparators. With these distance measurements and known coordinates of reference points (predefined), the distance camera position can be determined by the same 20 algorithms mentioned above by placing north, north, south (E), west (W) and center of a single reference object instead of multiple reference points. (O) reference points. The positioning accuracy is lower when using a single benchmark than when using multiple benchmarks (n> 3). 1

In the case where two reference objects are shown in the image, the position of the user carrying the distance camera is determined as the average obtained from the solution of two individual reference objects.

Claims (8)

7 1. A method for locating a moving subject using a distance camera. The method is characterized in that it comprises the following steps: 5 a) installing reference objects indoors b) acquiring an image and evaluating distance measurements c) analyzing the obtained image and identifying reference objects d) determining the position of a moving object by reference coordinates and 10 corresponding distance measurements. Method according to claim 1, characterized in that the reference objects are located at known locations. 15 The method according to claims 1-2, characterized in that the reference objects are specifically designed for indoor positioning. Method according to claims 1 to 3, characterized in that step c) comprises an algorithm which identifies the reference object from the image using a correlation method. 20 Fourier transforms. Method according to claims 1 to 3, characterized in that step c) comprises an algorithm which identifies the reference object from the image using a specially designed barcode system. 25 Method according to claims 1 to 5, characterized in that it locates a moving object when more than two reference objects are visible in a single image. 30 7) A method according to claims 1-5, characterized in that it locates a moving object when two reference objects are visible in a single image. Method according to one of claims 1 to 5, characterized in that it locates a moving object when a single reference object is visible in a single image. 8