FR2747199A1 - DEVICE FOR LOCATING A MOBILE OBJECT - Google Patents

Abstract

The present invention relates to a device for locating a mobile object comprising at least one optical system (2, 3), defining a frame of reference by its shooting axis and by a perpendicular plane containing a photosensitive detector (4, 4 ' ). The photosensitive detector (4, 4 ') comprises paired photosensitive sensors arranged in line, each sensor being connected alternately to an even and odd output. Each of the outputs is connected to a sensor management unit (5) so as to process by image subtraction the detector signals received from the photosensitive sensors.

Description

1 2747199

  The present invention relates to a device for time localization

  real, relative to a three-dimensional frame of reference, of a moving object.

  From patent FR-A-2,583,523, a system is known for locating a movable object relative to an axis (OX), comprising a photosensitive detector arranged at a fixed station and associated with an optic observing the environment of said axis in which the mobile is capable of moving, as well as an optical transmitter linked to the mobile and generating light flashes, remarkable in that: - said photosensitive detector comprises on the one hand, a fixed matrix 10 of controllable photosensitive elements , and, on the other hand, electronic means for controlling photosensitive elements; and - in that synchronization means are provided so that the electronic control means deliver control pulses

  triggering pictures taken synchronized with the flashes of light.

  Thus, the position of the mobile relative to the axis is given by the position of the photosensitive element or elements of the matrix, excited by the image of the light flashes. Such a known location system therefore has the disadvantage that it is necessary to equip the mobile with a transmitter in order to be able to locate it by the photosensitive elements of the matrix and does not make it possible to know the

coordinated along the OX axis.

  In addition, in order to know the position in Z, it is necessary to perform a

  line-to-line calculation not allowing work in real time.

  The object of the invention is to propose a passive device making it possible to determine the coordinates, in real time, of a mobile object, and which does not have

  not the disadvantages of the known device.

  To this end, a device for locating a mobile, according to the invention, comprises at least one fixed optical system, defining a frame of reference by its shooting axis and by a perpendicular plane containing a photosensitive detector, and is characterized in that said photosensitive detector comprises paired photosensitive sensors arranged in line, each sensor being connected alternately to an even and odd output, said outputs being connected to the sensor management unit so as to process by

  subtraction of images the detector signals received from the photosensitive sensors.

  According to a characteristic of the invention, the data transmitted to the computer define address half-words so as to allow the computer to read in a pre-calculated table of results of the position of the mobile.

2 2747199

  According to a characteristic of the invention, said photosensitive sensors are of the charge transfer type. It is advantageous for the photosensitive elements to be of the charge transfer type, generally designated by the abbreviation CCD (Charge Coupled Devices), the sampled output signal of which is particularly suitable for digital processing. According to another characteristic of the invention, said optical sensors have the shape of spade-like triangles so as to allow, by the ratio of the amplitude of the signals delivered on the even / odd sensors, to know the geometric position of the object compared to the benchmark set on the system

l0 optics.

  Preferably, the photosensitive sensors have the shape of triangles

isosceles head spades.

  According to another characteristic of the invention, the device comprises two

  optics each associated with a photosensitive detector.

  According to another characteristic of the invention, said sensors

  photosensitive have the shape of adjacent elongated rectangles.

  According to another characteristic of the invention, said optical sensors have the form of rectangles, the even rectangles being aligned with each other, the sensors

  odd being aligned with each other and offset from even rectangles.

  The features of the invention mentioned above, as well as others

  will appear more clearly on reading the following description of an example

  of realization, this description being made in relation to the attached drawings

  among which: Fig. 1 is a schematic view of the locating device according to the invention, FIG. 2 is the representation of the angular sectors according to the invention, FIG. 3 is the schematic representation of the mesh of the space observed by the location device of the invention, FIG. 4 is the schematic representation of a video signal after subtraction obtained by the photosensitive sensors, FIG. 5 is the schematic representation of the processing of the detection signal, FIG. 6 is a schematic view of the photosensitive sensors according to a preferred embodiment, and FIG. 7 is a graph showing the linear function of the photosensitive sensors of FIG. 6, and Fig. 8 is a schematic view of the photosensitive sensors according to a

another embodiment.

3 2747199

  The device for locating a mobile 1, according to the invention, and shown in FIG. 1, essentially consists of two optical systems 2 and 3, arranged at a fixed station, each associated with a photosensitive detector 4 and 4 ′, a sensor management unit 5, a computer 6, a controller 7 as well as a decision station 8. Such a device makes it possible to identify and know in real time the coordinates of a mobile 1, in relation to a reference frame defined by the optical axis of

  shooting and by a perpendicular plane containing the photosensitive detector.

  It will be noted that the two optical systems must be synchronous and that the focal length of the objective must be adapted to the dimension of the mobile object and to

the maximum detection distance.

  Likewise the integration time of the light energy by the sensors

  photosensitive CCD must be adapted to the speed of the mobile.

  According to a first embodiment, the photosensitive detector 4 or 4 ′ is an electronic circuit consisting of a quartz window behind which are

  arranged with photosensitive sensors in the form of very elongated rectangles.

  The photosensitive sensors are arranged in a strip, that is to say, in line,

  each sensor being connected alternately to an even and odd output.

  The detector 4 and 4 'is placed behind the optical system 2 and 3. Each optical system 2 and 3 thus has a representation of the landscape formed by a

  set of angular sectors (Fig. 2).

  Each angular sector corresponds to one of these photosensitive sensors.

  This gives a certain height Zmax proportional to the optical system used and the distance d between the optical system and the object. All of the

  angular sectors of an optical system form a network.

  In the photosensitive detectors used, the electronic processing means are arranged on each side of the photosensitive sensors for reasons of space to process the signals. The even photosensitive sensors are processed on one side and the odd photosensitive sensors are processed on the other. side. We have two outputs p and i, which are also called measurement channels: an even output p and an odd output i. Electronics, well known to those skilled in the art, can switch the two outputs on the same wire or on two

separated sons.

  The two measurement channels are then processed in the management box 3 5 of common sensors 5 in order to allow the supervision of the whole operation, the parallel processing of the signals put in random access memory as well as the subtraction of images so eliminate electrical signals

  from fixed parts of the observed landscape.

4 2747199

  It will be noted that the sensor management unit 5 includes a subtractor

  image for each optical system.

  The data from the box is then transmitted to the computer 6 through an interface 9 in order to interpret the detection signals and calculate the coordinates of the mobile. The coordinates of the calculated mobile are transmitted to the decision station 8. It will be noted that the decision station 8 can either be a screen giving the information on the mobile to an operator, or an action triggered

  l0 automatically following the approach of the mobile.

  The controller 7 of the location device is connected to the computer 6 and to the sensor management unit 5 in order to allow the synchronization of the

optical systems 2 and 3.

  It is advantageous for the photosensitive sensors to be of the charge transfer type, generally designated by the abbreviation CCD, the output signal of which

  sampled is particularly suitable for digital processing.

  The CCDs currently used are arrays of 512 photodiodes of 12 * 2500 pm2 format arranged in parallel and spaced 12 pm apart. Their projection through the optical system of shooting defines a set of

512 contiguous angular sectors.

  The assembly of the two optical systems 2 and 3, the optical axes of which intersect at a point determined in advance, thus forms a mesh of the common optical field, as shown in FIG. 3. The global device thus gives a representation of the space in closed cells whose real coordinates

can be pre-calculated.

  Each closed cell, formed by the overlap of all the angular sectors observed, can be identified individually by the rank occupied by their two photosensitive sensors in their own network.

formed the cell.

  The object is highlighted by its displacement.

  For each cell, the real (spatial) coordinates linked to a frame of reference fixed on the observation post can be calculated in advance and entered into a table of results. This result table is stored in a

computer memory.

  The operation of the device will now be explained.

  The mobile is detected by subtracting images.

  The images of the same landscape obtained from two distant observation posts are not identical. From their differences we can deduce the distance

2747199

  visible objects in the field. It's the stereoscopic effect. On the other hand, and for the same observation position, if one superimposes a positive image and a negative image of the same landscape, all the elements of these images disappear. We thus achieved a subtraction of images where everything that remained constant between the two moments of shooting will be visible as twice the object, in its starting position in positive and in its arriving position in negative. An image io read at time to is kept on a delay line or in a digital memory. At the moment to + dt a new image i1, in principle little

  o10 different from io, is read and subtracted point to point from io.

  The difference signal obtained contains: the noise specific to each image, the local variations in brightness and position, but above all the object imprint, i.e. the distance traveled L, the length I and the orientation of the object, as shown in FIG. 4. The "fixed landscape" has disappeared as during the

  superposition of a negative and its positive.

  A real-time processing of this difference signal leaves only the signatures of the new objects transformed into logical markers like

shown in Fig. 5.

  Thus, when the photosensitive sensors have detected the presence of an object in a mesh, "logical markers" from each of the optical systems 2 and 3 are transmitted to the computer at different times tl, t2.

  , the rank (value of the clock counter) of each photosensitive detector of the first network is used as a half-word of address or of the table of real coordinates. The second network will give the other address half words ni '. The spatial position of the entire mobile is then read without calculation in the results table 10, which advantageously allows work in real time, giving the cell / XY coordinate correspondence relative to a reference point of the workstation. observation. These coordinates are then transmitted..DTD: to the decision-making post.

  The Z coordinate is not precisely defined, we know the direction, that is to say the position of the mobile in a vertical plane which thickens more or less

depending on the optics used.

  According to another preferred embodiment, the localization device is identical to the previous embodiment, with the difference that the photosensitive sensors of the detectors have the shape of triangular pixels, in particular isosceles as shown in FIG. 6. Such a characteristic makes it possible to know precisely the Z coordinate, that is to say the height of the mobile. Advantageously, the photosensitive sensors have the shape of triangles

6 2747199

  isosceles and are arranged upside down, that is to say that one finds on a line all the even sensors which have the point downwards and will come out on a wire and all

  the odd sensors that have the tips up and will come out on another wire.

  Such a pixel shape has the advantage of having a linear function, which makes it possible to plot the linear function of the even and odd sensors as

  shown in Fig. 7, the sum of the two signals being constant.

  As previously described, all of the networks of the two optical systems make it possible to know the distance of the mobile and the maximum height of the angular sector where the object is located. At time t1, the 0o pixel number of the first network serves as an address half-word for reading a memory, the pixel number of the second network representing the other address half-word: at this address we find the coordinates in X and in Y. The coordinate in X makes it possible to read in a result table the correspondence of the maximum height Zmax, in instantaneous reading; we get a response in a time of

  Around 20 nanoseconds after the request.

  The quantities X, Y and Zmax thus processed are stored in order to carry out in the computer the calculation of the Z coordinate by the following equation: Z Zmax = A / (A + B) o A = Amplitude of the even sensor signal and , B = Amplitude of the odd sensor signal Thus, for example if the image of a mobile, being rather close to Zmax, is projected on this device, there will be a weak signal amplitude A on the even sensors and an amplitude strong signal B on odd-numbered detectors. And conversely if the object is at a very low height, a strong signal amplitude A on the even sensors and a weak signal amplitude B on the odd sensors. When the mobile is in a position close to the middle of Zmax, the signals will have the same amplitude A = B. The table of results 10 has been written, definitively, in a memory and takes account of the focal length, of the dimensions of the network, of the spacing between

two optical systems 2 and 3.

  With the triangle network the detection signal oscillates between two A values

and B.

  According to another variant of the detector, shown in FIG. 8, the photosensitive sensors can have the shape of rectangles, the even rectangles being aligned with one another, the odd sensors (Pi) being aligned with each other and offset by

7 2747199

  compared to even rectangles (Pp). Such a form of sensor makes it possible to delimit three zones, a high zone, a low zone and an intermediate zone. The operation is the same as previously described, thus determining the

  position of an object in one of these three zones.

  It will be noted that the device of the invention works equally well in the

visible only in infrared.

  Without departing from the scope of the invention, it will be readily understood that the location device can comprise a single optical system. In this type of embodiment, we will then use either rectangular pixels which will allow o10 only to detect an object in an angular sector without knowing the coordinates, or triangular pixels which will make it possible to detect the object in an angular sector as well as its coordinate in Z. This type of embodiment can be advantageous, for example, to locate the object on a conveyor belt.

8 2747199

Claims (9)

  1.- Device for locating a mobile object comprising at least one optical system, defining a frame of reference by its shooting axis and by a perpendicular plane containing a photosensitive detector (4, 4 ′) characterized in that said detector photosensitive (4, 4 ') comprises paired photosensitive sensors arranged in line, each sensor being connected alternately to an even and odd output, said outputs being connected to a sensor management unit (5) so as to be processed by subtraction of images the detector signals received from photosensitive sensors.   2.- Device according to claim 1, characterized in that the data processed by the sensor management unit (5) are transmitted to a computer (6) and define address half-words so as to allow the computer ( 6) the   reading in a table of results (10) pre calculated of the position of the object (1).   3.- Device according to claim 1, characterized in that said sensors   optics are of the charge transfer type.   4.- Device according to claim 1, characterized in that it comprises two   optics each associated with a photosensitive detector.   5.- Device according to any one of the preceding claims,   characterized in that said photosensitive sensors have the shape of spade-like triangles so as to allow, by the ratio of the amplitude of the signals delivered to the even / odd sensors, to know the geometric position of the object by   compared to a frame of reference fixed on said optical system.   6.- Device according to one of claims 1 to 4, characterized in that said   optical sensors have the shape of isosceles triangles arranged head to tail.   7.- Device according to one of claims 1 to 4, characterized in that said   photosensitive sensors have the shape of elongated rectangles.   8.- Device according to one of claims 1 to 4, characterized in that said   optical sensors have the shape of rectangles, the even rectangles (Pp) being aligned 9 2747199   between them, the odd sensors (Pi) being aligned with one another and offset with respect to to even rectangles.   9.- Device according to claim 2, characterized in that the computer (6) is connected to a decision station (8).