FR2898825A1 - Parasite target e.g. object, detecting system for e.g. non-polluted variable geometric working environment, has units calculating relation connecting quantity, where relation verification depends on integration of target with paint sensors - Google Patents

Abstract

The system has cylindrical active and passive capacitive sensors (F1, F2) made of paint, associated to active and passive conditioners (2, 4), where the sensor (F1) is an aluminum tube. The conditioners communicate with a digital signal processing unit (P) and a storing unit (M) e.g. RAM, by a communication unit (5) e.g. analog/digital converter. The conditioners produce output quantity representing load variation in the sensors. The units (P, M) calculate relation connecting the quantity, where relation`s verification depends on integration of a parasite target e.g. person, with the sensors. An independent claim is also included for a method of detecting a parasite target.

Description

Capacitive variation detection system The invention relates to the

  domain of presence detectors using capacitive sensors for spatial detection. The invention relates in particular to the analysis and control systems applied, for example, to protection in hazardous environments, such as the working and moving space of a robot or a press. Presence detection systems using capacitive influence coefficients are known for determining whether a target, for example an object or a person, is disturbing the capacitive device. This type of sensor is adapted in particular to a protected space delimited by fixed elements. In order to adapt to environments with variable geometry or articulated mechanical systems, existing systems are based on the analysis of capacitive influence coefficients relative to a signature. The signatures are representative of the capacitive influence coefficients in all the programmed positions, in the variable geometry environment, for example of a robot or a press. A signature of a determined position, established in a non-polluted environment, is compared with the analysis of the capacitive influence coefficients corresponding to this determined position. Each new position, for example of the robot or the press, requires a comparative analysis with the signature associated with this position. The disadvantage of this type of system is that it works in all or nothing, compared to a determined security threshold. The all-or-nothing operation does not make it possible, for example, to determine the position of the detected target or the target volume of the detected target. Moreover a modification of the displacements in the working environment requires the computation of a set of new signatures. In general, this type of signature analysis system is not very precise and does not make it possible to dynamically configure a machine with variable geometry. In particular, the most mobile elements are often the most dangerous elements and require precise identification. The object of the present invention is to overcome several disadvantages of the prior art by creating a detection system in a variable geometry working environment, using the couplings and the influence coefficients of a set of capacitive sensors, making it possible to realize a dynamic analysis of capacitive sensors placed on moving elements, the dynamic analysis allowing a flexible adaptation in a variable geometry environment, with a determination of the position and the size of the detected targets. This objective is achieved by means of a detection system in a variable geometry working environment comprising storage means, processing means, communication means, characterized in that it comprises an active capacitive sensor associated with an excited conditioner and at least one passive capacitive sensor associated with a non-excited conditioner, at least one of the capacitive sensors is disposed on a moving part of the working environment, the excited conditioner is electrically connected to the active capacitive sensor to provide an electrical power of excitation and comprises means for measuring the load variation of the active capacitive sensor, each non-excited conditioner is electrically connected to its associated capacitive sensor and comprises means for measuring the load variation of its associated capacitive sensor, the conditioners communicate with the processing means and the memory means p ar the communication means, the conditioners each produce at least one output quantity representative of the charge variation in their associated capacitive sensor, the storage means and the processing means are arranged to calculate at least one binding relationship at least an output quantity of each conditioner, the relation being verified if no target interacts with the capacitive sensors, the relation being no longer satisfied if a target interacts with at least one capacitive sensor According to another particularity, an alarm signal is sent to a control system of the variable geometry environment, via communication means, when the relation is no longer verified. According to another particularity, each conditioner comprises an operational amplifier operating in linear mode, the inverting input of the amplifier being electrically connected to the associated capacitive sensor, the non-inverting input being electrically connected to a means of producing a voltage, a resistance being electrically connected between the inverting input and the output of the operational amplifier, the output quantity of the conditioner corresponding to the amplitude of the output voltage of the operational amplifier. According to another feature, the resistance of each conditioner to the same determined value, the means for producing a voltage are controlled via the communication means, by the processing means associated with the storage means, the means for producing a voltage supply a voltage zero for non-excited conditioners and a sinusoidal amplitude and pulse determined for the excited conditioner. According to another feature, the output voltages of the operational amplifiers are electrically connected to the communication means which comprise conversion means transforming the analog values of the output voltages into representative digital values so as to be processed by the processing means associated with the means. memorisation. According to another particularity, the processing means associated with the storage means control a permutation of the active sensor among the capacitive sensors by controlling, the means for producing a voltage to force the voltage of the previously excited conditioner to zero and the voltage of a conditioner previously not excited at a sinusoidal voltage of determined amplitude and pulsation.

  According to another feature, the processing means associated with the storage means control the means for producing a voltage whose amplitude is clivalised in a determined ratio, so as not to influence the other capacitive sensors.

  According to another particularity, the processing means associated with the storage means are arranged so as to calculate an initial capacity for each capacitive sensor, in a determined initialization position. A second objective is to provide capacitive sensors whose size and weight are adapted to highly mobile elements, the manufacture of these sensors remaining cheap. According to this objective, the capacitive sensor comprises a first layer of conductive paint on which is superimposed a second layer of dielectric paint, on which is superimposed a third layer of conductive paint, the first and the third layer being electrically insulated one of the other by the second layer, the first layer being electrically connected to the ground of the operational amplifier of the conditioner with which the sensor is associated, the third layer being electrically connected to the inverting input of the operational amplifier of the conditioner. According to another feature, the conductive tab of the operational amplifier corresponding to the inverting input and respectively to ground is directly soldered to the third and respectively the first paint layer of the sensor. According to another feature, a coaxial shielded cable comprising a central electrical wire surrounded by a conductive sheath, the central son electrically connects the third layer of paint of the sensor to the inverting input of the amplifier of the conditioner and the conductive sheath connects the first coat of paint from the sensor to the conditioner amplifier's ground. Another objective is to position the capacitive sensors to optimize the analysis by the system according to the invention.

  According to this objective, the system according to the invention comprises at least two capacitive sensors at a determined distance from one another, intended to be activated successively, to achieve a variation of the distance of the active capacitive sensor to the target.

  According to another feature, the system comprises at least a plurality of capacitive sensors arranged on a part, around its axis of rotation, to be activated successively, to determine the orientation of the target relative to the workpiece. Another object of the invention is to provide a detection method by a system according to the invention, allowing the detection of a target. This object is achieved by a method for detecting a target in a variable geometry work environment by a system according to the invention characterized in that it comprises at least: a step of measuring and memorizing the output quantities conditioners, in an initialization position in which the environment is not disturbed by any target, - a step of measuring and storing the output quantities of the conditioners in a verification position, - a step of comparing a an experimental calculation result depending on the quantities measured with a result of a provisional calculation depending at least on an output quantity of a conditioner and a characteristic value of a conditioner, for determining in case of equality target is detected or in case of inequality no target is detected.

  According to another particularity, the step of measuring and storing the output quantities of the conditioners, in an initialization position is followed by a step of calculating and storing the initial capacitances of the sensors in the initial position. According to another particularity, the initial capacity of the active sensor and the initial capacity of a passive sensor are determined by the following calculations: C1s = (Vlinit2 - VE2) 1 / 2I (R.W.VE) CJS = VJinit / (VE.R.W)

  in which: - Cis is the initial capacity of the active sensor in the initial position, expressed in farad, - CJS is the initial capacitance of the passive sensor numbered J, in the initial position, expressed in farad, -Vlinit is the amplitude of the voltage at the output of the active conditioner in the initial position, expressed in volts, 10 - VJinit is the amplitude of the voltage at the output of a passive conditioner associated with a sensor numbered J, in the initial position, expressed in volts, - VE is the amplitude of the excitation voltage of the active sensor, expressed in volts, - W is the pulsation of the excitation voltage of the active sensor, expressed in radians per second, - R is the resistance in the conditioners, disposed between the inverting input and the output of the operational amplifier, expressed in ohm. According to another particularity, the step of measuring and memorizing the output quantities of the conditioners in the verification position is followed by a step of calculating and memorizing the experimental calculation, which is followed by a calculation step and memorizing the provisional calculation. According to another particularity, the result of the experimental calculation and the result of the provisional calculation are determined by the following calculations: N N

  R3 = V12 - Vlinit2 '+ Vlinit2 - ~ VJ2 J = 2 J = 2 N N

  R4 = 2. (VE.R.W) 2. Where R 3 is the result of the experimental calculation, expressed in square volts, R 4 is the result of the experimental calculation, expressed in square volts, R 4 is the result of provisional calculation, expressed in square volts, - VI is the amplitude of the output voltage of the active conditioner in the verification position, expressed in volts, - Vj is the amplitude of the output voltage of a passive conditioner associated with a sensor numbered J, in the verification position, expressed in volts, According to another particularity, the step of comparing the experimental calculation result with the result of the provisional calculation, in case of equality at a margin of error determined by , is followed by a jump to the step of measuring and storing the output quantities of the conditioners in another verification position. Another objective is to achieve a location of the target 15 in the variable geometry environment. According to this objective, the step of comparing the experimental calculation result with the result of the provisional calculation, in case of no equality according to the error margin, is followed by at least: a step of deactivating the active sensor and activating the passive sensors in the active state by a minimized excitation so that the influence between sensors is negligible, - a step of measuring and storing the output quantities of the activated conditioners. According to another particularity, the method is followed by at least: a step of calculating and memorizing the capacitive influence coefficient of the target on each activated capacitive sensor; a step of calculating and memorizing the coefficient of Capacitive influence of the target on the deactivated sensor. In another feature, the capacitive influence coefficient of the target on the activated sensor numbered J and the capacitive influence coefficient of the target on the deactivated sensor are determined by the following calculations: NNN ECJ, = 1I (VE. R. W). ..VJ - E CJs J = 2 J = 2 J = 2

  CDJ = I (VJ2-VE'2) 112I (VE ', R. W) ~ - CJs N N N

  CD1 = [C1s + ECJ1] ù [(C1s + ECJ1 + R3 / (VE.R.W) 2 1112 - E CDJ J = 2 J = 2 J = 2

  in which: - CDJ is the capacitive influence coefficient of the target on the activated sensor numbered J, expressed in farad,

  CD1 is the capacitive influence coefficient of the target on the deactivated sensor, expressed in farad,

  - VE 'is the amplitude of the new excitation voltage of an activated sensor, expressed in volts,

  CJ1 is the capacitive influence coefficient of the activated sensor numbered J on the deactivated sensor, expressed in farad,

  According to another particularity, the method is followed by at least:

  a step of measuring and storing a displacement of the deactivated capacitive sensor, the value of the displacement being determined by displacement measuring means in communication with the processing and debugging means,

  a step of activating the deactivated capacitive sensor and deactivating the activated capacitive sensors; a step of measuring and memorizing the output quantities of the conditioners;

  a step of deactivating the active sensor and activating the passive sensors in the active state by a minimized excitation so that the influence between sensors is negligible; a step of measuring and memorizing the output quantities of the sensors; activated conditioners, 8 - a step of calculating and storing the capacitive influence coefficient of the target on each activated capacitive sensor, - a step of calculating and storing the capacitive influence coefficient of the target on the deactivated sensor.

  According to another particularity, the method is followed by at least: a distance evaluation step comprising calculating and memorizing the variation of the capacitive influence coefficient of the target on the active capacitive sensor, and the calculation; and memorizing the ratio of the capacitive influence coefficient and this variation.

  According to another feature, the variation of the capacitive influence coefficient of the target on the active capacitive sensor and the distance between the active capacitive sensor and the target are determined by the following calculations: S01 = CDI. d01 / E PVar = (CD ,,, û CDI) / dl 15 d01 = - CDI / PVar in which: - S01 represents the surface facing the capacitive sensor active with the target, expressed in square meters, - CD1 the coefficient of d capacitive influence of the target on the active sensor, expressed in farad, - d01 represents the distance between the target and the active sensor, expressed in meters, - E represents the permittivity of the medium between the target and the active sensor, expressed in farad per meter, 25 - PVar represents the variation of the capacitive influence coefficient of the target on the active sensor, expressed in farad per meter, CD1n represents the new coefficient of capacitive influence of the target on the active sensor after the displacement, expressed in farad, CD1 represents the capacitive influence coefficient of the target on the active sensor before displacement, expressed in farad, - dl represents the displacement of the target relative to the active sensor, expressed in meters.

  According to another particularity, the step of evaluating the distance according to the method is followed by at least: a step of sending an emergency stop message to the environment control machine with variable geometry. According to one variant, the step of evaluating the distance according to the method is followed by at least: a jump to the step of measuring and memorizing the output quantities of the conditioners in a verification position. According to another variant, the step of evaluating the distance according to the method is followed by at least: a step of configuring a new active conditioner, a jump to the step of measuring and memorizing the output quantities of the conditioners in a verification position. The invention, its characteristics and its advantages will appear more clearly on reading the description given with reference to FIGS. 20 referenced below: FIG. 1 represents an example of a system with two sensors according to the invention; FIG. 2 represents an example of sensor arrangement according to the invention in a variable geometry environment; FIG. 3 represents an exemplary system according to the invention with N sensors; FIG. 4 represents a comparison between an experimental calculation and a provisional calculation according to the invention, in the configuration of FIG. 1; FIG. 5 represents an example of a sensor according to the invention; FIG. 6 is a graphical representation illustrating the calculation method, according to the invention, of the distance between a mobile sensor and a target. FIG. 7 represents the steps of a detection method according to the invention; - Figure 8 shows a sectional view of an example of two capacitive sensors, according to the invention, offset by a predetermined distance, arranged on a part of the environment variable geometry; FIG. 9 represents a perspective view of the offset capacitive sensors represented in FIG. 8; FIG. 10 represents a perspective view of sensors arranged along an angular sector on a part of the variable geometry environment. The invention will now be described with reference to the figures mentioned above. FIG. 1 represents two cylindrical sensors (F1a, F2a) located at a determined distance (d) from each other. The sensors are made of a conductive material, such as aluminum. Without limitation, in the example of Figure 1, the axes of the cylindrical sensors are taken parallel. The cylindrical sensors each have the same height (H2) and the same diameter (H1). This configuration is used in particular for the experimental validation of the method according to the invention. On the other hand, a sensor of this type could for example be placed around a mobile robot arm or on a fixed barrier defining a work area. In the system according to the invention, a first active sensor (F1 a), for example an aluminum tube, is connected by an electrical link (3a) to an excitation and analysis device, also called an active conditioner ( 2). In general, an electronic conditioner is a device which, by means of incorporated electronic devices, serves to supply the sensors and / or to process the signals delivered by these sensors. In the present invention an active conditioner provides electrical excitation power to the active capacitive sensor (F la). On the other hand, the active conditioner comprises means for analyzing the variation of the charge in the capacitive sensor (Fia). A second passive sensor (F2a) is connected by an electrical link (3b) to an analysis device also called a passive conditioner (4). The passive conditioner does not provide electrical excitation to the capacitive sensor. On the other hand, the passive conditioner comprises means for analyzing the load variation of the associated capacitive sensor (F2a). The conditioners shown in FIG. 1 are electrically connected each to a capacitive sensor. In a nonlimiting manner, each conditioner produces an analog output voltage whose value is representative of the load variation. The active conditioner (2), connected to the active capacitive sensor (Fia) via an electrical link (3a), applies a sinusoidal excitation voltage of determined amplitude and frequency to this electrical link (3a) and measures the current in this link (3a) electric by means of measuring the current. The capacitive sensor (Fia) active is equivalent to a capacitor having an electrode connected to the conditioner, the impedance of the equivalent capacitor being dependent on its own capacitance and potential capacitive influence coefficients of other capacitive sensors. The capacitive influence coefficient is for example calculated according to the calculation referenced L21, in appendix 3, in which the coefficient (CDK) of capacitive influence of a target on a capacitive sensor is equal to the permittivity of the medium (E) multiplied by the facing surface (S) and divided by the distance (dk). The medium is for example the air separating two plates of the same surface (S), arranged parallel to each other at a distance (d). In another example, the calculation (L21) of the capacitive influence coefficient is applied to elementary surfaces. In addition, the parasitic capacitances of the assembly are also taken into account.

  Without limitation, the active sensor (Fia) is connected to the inverting input of an operational amplifier (A02) of the active conditioner (2).

  This inverting input is electrically connected to a terminal of a resistor (R) of a determined value. On the other hand the other terminal of this resistor (R) is electrically connected to the output of the operational amplifier (AO2). The inverting input of the operational amplifier (AO2) is thus looped back on its output. The non-inverting input of the operational amplifier (AO2) is electrically connected to a generator (G2) of sinusoidal voltage of determined amplitude and frequency. On the other hand, this generator is connected to ground, that is to say, to the null potential of reference. The output voltage of the operational amplifier produces the output voltage of the conditioner. The operational amplifier (AO2) is used in linear mode. The voltage applied to the sensor is therefore identical to the voltage (VE) at the terminals of the generator. The voltage (R.11) across the resistor follows Ohm's law and is therefore the image of the current (11) in the electrical connection with the sensor.

  The output voltage (U1) of the amplifier added to the voltage (R.11) across the resistor is therefore equal to the voltage (UE) at the terminals of the generator, as illustrated by the relation referenced LI, in the appendix. 1. This relation (LI) between quantities varying with time, is also valid in harmonic regime, as illustrated by the relation L2, of appendix 1. In relation referenced L2, V1C is the complex amplitude, in harmonic regime, the output voltage of the active conditioner; 11C is the complex amplitude of the current and VEC is the complex amplitude of the excitation voltage applied to the non-inverting input of the operational amplifier (AO2). On the other hand, in the harmonic regime, the current (11C) multiplied by the impedance (ZC1) of the equivalent capacitor (C1) connected to the inverting input of the operational amplifier, is equal to the excitation excitation voltage ( VEC) as illustrated by the relationship L3 of Annex 1. The passive conditioner (4) electrically connected to the second passive capacitive sensor (F2) is similar to the active conditioner (2) except that the non-inverting input of the Operational amplifier (AO4) is electrically connected to ground, i.e. to the reference null potential. The amplifier operates in linear mode. In harmonic regime, the complex amplitude of the output voltage (V2C) of the passive conditioner is equal to the inverse of the complex amplitude of the current (I2C) multiplied by the resistance (R), as indicated by the relation L4 , in annex 1. The voltage applied to the electrical connection (3b) between the passive conditioner (4) and the associated passive sensor (F2a) is zero. The load variation in the passive capacitive sensor (F2a) depends on the capacitive influence coefficient of the active sensor (F1 a) on the passive sensor (F2a). In addition, the internal parasitic capacitances are also taken into account. The complex amplitude of the current (I2C) in the passive capacitive sensor is therefore a function of the equivalent capacitor (C2) connected to the inverting input of the amplifier, and of the complex amplitude (VEC) of the excitation voltage, as indicated by the relation L5, in appendix 1. In known manner, the complex impedance (ZC2) of the electrically equivalent capacitor (C2) connected to the passive conditioner (4) is a function of the capacity of the capacitor (C2) equivalent and the pulsation (W) of the applied voltage. The parasitic capacitances, such as the internal capacitance of the inverting input, influence the measured current, the measurement of the current being equivalent to the measurement of the variation of charge in the capacitive sensor. The conditioners (2, 4) are electrically connected communication means (5) for transmitting one or more output quantities and possibly receive one or more commands. The communication means (5), for example analog / digital converters or multimeters, communicate the amplitude of the output voltage of an operational amplifier in the form of a representative digital data item to a processing system. The conditioners are thus in communication with the processing system which comprises processing means (P) associated with storage means (M). The processing means include, in a nonlimiting manner, a processor or a dedicated integrated circuit of the ASIC type or a programmed logic circuit of the FPGA type.

  The storage means include, but are not limited to, volatile memory of the RAM type or of the non-volatile memory such as a hard disk or a CDROM. The processing means read / write access to the storage means (M) for example by a bus (B). The bus (B) is also connected to the communication means (5) or to other interfaces. In an exemplary embodiment described below, the communication means also comprise digital / analog converters converting a code or a computer command produced by the processing system into a voltage or an analog current received for example by a conditioner. In a nonlimiting manner, the processing system is electrically connected to a device (6) for controlling the machine on which the capacitive sensors (F1, F2) are arranged via an associated communication interface (16). The processing system communicates, for example to an apparatus (6) for controlling a robot, an emergency stop signal or a file containing information on the detection operations performed. In a nonlimiting manner, the processing system communicates in the same way, with a device for analyzing or displaying information. A verification program, resident in a memory space (MP) of the memory (M) and executed by the processing means (P), follows the verification method shown in FIG. 7. This method is, for example, applied to a system with two capacitive sensors (Fia, F2a), as shown in FIG. 1. The capacitive sensors and the conditioners are activated when a general power supply, not shown, is started. The active conditioner (2) then produces a sinusoidal excitation voltage of determined pulsation and determined amplitude at which the active capacitive sensor (F1) is subjected. A data representative of the amplitude and respectively the pulsation of this excitation voltage is stored in a memory space (MVE and respectively MW) of the processing system. The information concerning the pulsation and the amplitude of the excitation voltage are also accessible by the processing means (P) to perform different types of calculations.

  In another embodiment described below, the processing system (M, P) controls the amplitude and the pulsation of the excitation voltage of a conditioner. This information is for example read by the processing means (P) in order to generate an excitation voltage command to an active conditioner (2). The resistance (R) in each of the conditioners (2, 4) has the same determined value, this resistance being stored by the memory (M) in a storage space (MR), in the form of a representative digital code.

  In order to perform steady-state measurements, the verification program includes, for example, timing steps that are not shown. The verification program begins with a step (ETPOI) initialization measurement initial. This step (ETPO1) consists of measuring the output quantities (V1 init, V2init) of the conditioners, in an initial position. In its initial position, the machine (6), for example the robot arm, is not brought into contact with or near a parasitic target (DO). The environment of the machine in the initial position, is not disturbed by a target (DO). The communication means (5) for example measure the rms value of the output voltage and provide the processing system (M, P) with a digital signal, for example coded on a byte, representative of the rms value. The processing means (P) then calculate the amplitude of the output voltage by multiplying the rms value by the square root of two. In an alternative embodiment, the analog value of the output voltage (U1 and U2 respectively) for each of the conditioners (FIa, F2a) is translated in digital form into a representative signal comprising for example 256 amplitude levels. The signals representative of the amplitude over time are for example stored in memory (M) and analyzed by the processing means (P) to determine the modulus of the complex amplitude of the sinusoidal signal at the output of the operational amplifier.

  The amplitudes of the initial voltages (V1 ends and V2init respectively) at the output of the conditioners, measured and possibly calculated by the processing means, are stored in memory spaces (MV1 ends and respectively MV2init).

  When the initial output quantities are stored (cond01), the verification program executes the calculation step (ETP02) and stores the initial capacities. During this step (ETP02) the processing means access memory spaces (MR, MW, MVE, MV1 init, MV2init) to read in particular the resistance (R), the pulsation (W) and the amplitude (VE) of the excitation voltage and amplitudes (V1 ends and V2init) initially measured voltages, for two initial capacitance calculations for (C1S, C2S) the two sensors (F1 a, F2a). The calculations carried out correspond to the calculations referenced L6 and L8, in Appendix 1, in which V1init and V2init are the modules of the complex voltages, in harmonic regime, at the output of the active and passive conditioners. Similarly, VE represents the modulus of the excitation complex voltage. Each initial capacity (C1S and C2S respectively) calculated is stored in a memory space (MC1S and respectively MC2S) by the processing means (P). The computation referenced L6 is equivalent to the computation referenced L7 which expresses the modulus of the complex tension at the output of the active conditioner, in the initial position. The expression L7 is obtained in a known manner by the application of the law of meshes under harmonic regime. The computation referenced L8 is equivalent to the computation referenced L9 which expresses the modulus of the complex tension at the output of the passive conditioner, in the initial position.

  The expression L9 is obtained in a known manner by the application of the law of meshes under harmonic regime. When the initial capacitances (C1S, C2S) are memorized (cond02), the verification program executes the step (ETP03) for measuring and storing the output quantities (V1, V2) of the conditioners (2, 4), in a position of verification of the machine. The verification position according to the invention is any position of the machine equipped with the detection system according to the invention. The processing means (P) perform a measurement of the output signals (U1, U2) of the conditioners and calculate their amplitude (V1, V2). The codes representative of the amplitudes (V1 and respectively V2) of the outputs of the conditioners associated with the sensors (Fia and respectively F2a) are placed by the processing means in memory spaces (MV1 and respectively MV2). After storing the output quantities (V1, V2) in the verification position (cond03), the verification program performs an experimental verification calculation step (ETPO4), relating to directly measured values. The processing means (P) accesses the memory spaces (MV1 ends, MV1, MV2init, MV2) containing information representative of the amplitudes of the output voltages in the initial position and in the verification position, for the conditioners of the system. The processing means (P) carry out the experimental verification calculation, referenced L10 in Appendix 1, in which V1init and V1, respectively, is the modulus of the complex voltage measured at the output of the active conditioner, in the initialization and verification position respectively. . V2init and respectively V2 is the modulus of the complex voltage measured at the output of the passive conditioner, in the initialization and verification position respectively. The result of this calculation is placed in a memory space (MR1) by the processing means (P). After storing (cond04) the result (RI) of the experimental verification calculation (L10), the verification program performs a forecasting step (ETP05), relating to the output quantity of one or more conditioners, measured directly. , characteristics of the conditioners and initially calculated capacitance values. The processing means (P) thus access the memory spaces (MV2, MVE, MR, MW, MC1S, MC2S) concerning the various stored representative data necessary for the predictive calculation referenced L11, in Appendix 1. The result (R2) of this calculation (Li i) is placed in a memory space (MR2) by the processing means (P). The calculation referenced L11 is derived in a known manner from the expression (L1 1a) expressing the modulus of the complex voltage (V1) as a function of the modulus of the excitation complex voltage (VE), the resistance (R), the the pulsation (W) and the equivalent capacitance (Cl) at the input of the conditioner. The capacitance (Cl) equivalent at the input of the conditioner is taken equal to the capacity (C1S) initially calculated, added to the coefficient (C21) of capacitive influence of the passive sensor (F2a) on the active sensor (FI a). The computation referenced L11b is also used to obtain the computation referenced L11. Calculation L1 1b expresses the modulus of the complex voltage at the output of the passive conditioner (4). In known manner, the capacitive influence coefficient (C12) of the active sensor on the passive sensor has a value equal to the capacitive influence coefficient (C21) of the passive sensor on the active sensor. After storing (cond05) the result (R2) of the predicted calculation (L11), the verification program performs a step (ETP06) of comparing the result (R1) of the experimental verification calculation (L10) with the result (R2) of the provisional calculation (L11). The equality (condO61) between the two results (R1, R2) is considered to be verified within a margin of error. Figure 4 shows a series (SE) of experimental verification calculations compared with a series (SP) of predictive calculations, in an environment disturbed by no target (DO). The series (SE, SP) of calculations are each represented in the form of a curve. The curves are mapped, a point of the abscissa axis representing a determined measurement configuration and the ordinate axis representing the corresponding result (R1 and R2 respectively) for each of the calculations (CE and CP respectively). In this experimental verification, two cylindrical sensors as shown in FIG. 1 have been separated from each other by a distance referenced d, the distance of distance (d) being the variable of the abscissa axis. The results of the calculations were carried out in a non polluted environment, that is to say without target coming to parasitize the environment with variable geometry. The tolerated margin of error, according to which the results of the calculations are considered equal to each other, is either fixed or variable and depends, for example, on the value of the experimental or provisional calculation. In a nonlimiting manner, the margin is, for example, inversely proportional to the value of the experimental calculation. In this way, the comparison made by the verification program makes it possible to determine whether a parasitic target (DO) is present in the variable geometry environment because in the case where a target (DO) is present the equality of the results (RI , R2) is no longer verified.

  If the comparison is verified (cond061), the verification program executes, for example an unpolluted environment step (ETP07). In a nonlimiting manner, this step (ETP07) comprises for example a transmission operation. At the end (cond07) of the transmission, the verification program jumps to step (ETP03) for measuring and storing the output quantities (V1, V2) of the conditioners (2, 4), in a position of checking the machine. Without limitation, the information previously stored and representative of a measurement or a calculation, are either overwritten and lost, stored in an archive storage space or transmitted to the machine (6) driving the variable geometry environment. If the comparison is not verified (condO62), the verification program performs a polluted environment step (ETP08). The verification program starts in a nonlimiting manner, a sequence of steps for locating and identifying the target according to the invention, or directly causes an emergency procedure, such as for example the immediate stopping of the robot arm or of the press. Even when triggering an emergency procedure on an all-or-nothing basis, as soon as a target is detected, the detection system of the present invention has the advantage of being operational without requiring a signature or variable representative of the position at the time of measurement. The verification relation of R1 = R2 is indeed a valid verification for all the positions of the variable geometry environment, the non-equality implying that a target is detected. A further step of location and identification, starting at the polluted environment step (ETP08), will be described later. A location and identification of the target (DO) are advantageously carried out in a system according to the invention comprising an active sensor (F1) interacting with a plurality of passive sensors (F2 to FN).

  In an alternative embodiment the variation of charge in a capacitive sensor (FI to FN) is measured by a conditioner comprising a toric sensor. The capacitive sensor is electrically connected to a device imposing a voltage being either a zero voltage or a sinusoidal alternating voltage. The toroidal sensor comprises a magnetic core disposed around the electrical connecting wire between the capacitive sensor and the device imposing a voltage. A son of measurement surrounded around the torus, according to a determined number of turns, produces at its terminals a voltage representative of the variation of the current through the torus. This voltage representing the variation of the current is then read by the processing system (P, M) via the communication means (5). The invention is not limited to two sensors. FIG. 2 shows an example of arrangement of a plurality of sensors (FI to FN) according to the invention. An active sensor (FI) is connected to an active conditioner (2). The other passive sensors (F2 to FN) are each connected to a passive conditioner (4). Passive sensors electrically connected to a null potential have a capacitive effect on the active sensor (F1). The active (FI) sensor produces an electric field to the passive sensors (F2 to FN). The active sensor (FI) therefore has a coefficient of influence (C12, C13, ... and respectively C1 N) on each of the passive sensors (F2, F3,

  . and respectively FN). The capacitive influence coefficient (C: 1J) of the sensor (F1) active on a passive sensor (FJ) is equal to the capacitive influence coefficient (CJ1) of the passive sensor (FJ) on this active sensor (F1). Advantageously, the passive capacitive sensors are connected to a constant potential, which is zero in the present invention ... DTD: Thus the passive sensors (F2 to FN) do not create an electric field and do not interact with each other. A target (DO) is for example symbolized by a toolbox. Depending on its distance from the capacitive sensors, the target (DO) may or may not have a capacitive influence coefficient on the sensor. Thus a target is detected by the system in a position close to an active or passive sensor. FIG. 3 also shows the coefficient (CDS) of the capacitive influence of the target (DO) on the passive ground sensor (FS) and the coefficient (CSD) of the capacitive influence of the passive ground sensor (FS). on the target (DO). The coefficient (CD1) of capacitive influence of the target (DO) on the active sensor (F1) and the capacitive influence coefficient (C1D) of the sensor (FI) active on the target (DO) are also represented.

  Figure 3 is a schematic representation of the detection system according to the invention. The active (2) and respectively passive (4) conditioner is for example a circuit similar to the active (2) and respectively passive (4) conditioner of FIG. 1. Without limitation, another electronic assembly, performing a function of the type non-inverting differentiator is usable in an alternative embodiment. The conditioners (2, 4) shown in FIG. 3 comprise an output (mes) to which is electrically connected at least one electrical link (3) with an electrode of a capacitive sensor, as well as in FIG. conditioners (2, 4) include an output (out) delivering for example a voltage representative of the variation of the load in the capacitive sensor, in the same way as in Figure 1. On the other hand each conditioner (2, 4) comprises a first input (cmdl) and respectively a second input (cmd2) for controlling the amplitude and respectively the pulsation of the excitation voltage applied to the electrical link (3).

  The control inputs of the passive conditioners (4) are controlled null by the processing system, in order to impose a zero potential on the passive capacitive sensors (F2 to FN). The first and respectively: the second control input of the active conditioner (2) is, for example, controlled at a voltage whose value is representative of the amplitude and respectively of the pulsation of the sinusoidal voltage applied to the active capacitive sensor (F1 ). In another variant embodiment, the conditioner associated with the capacitive sensor provides an output quantity whose frequency is representative of the coefficient of capacitive influence or the variation of the load in the sensor. A verification process followed by locating steps and determining the nature of the target will now be described. The conditioner comprises, as previously described, processing means associated with storage means, receiving signals from the communication means and representative of the output quantities produced by the conditioners. The verification program is stored (MP) in memory (M) and is executed by the processing means (P) as previously described. In the verification method, represented in FIG. 7, followed by the verification program, the step (ETPO1) of initialization initialization measurement comprises a first control phase of the conditioners. The processing means access the memory spaces (MW, MVE) comprising the pulsation and the amplitude of the excitation voltage in order to generate a control of the active conditioner (2). This digital control is translated by the communication means (5) and transmitted to the inputs (cmdl, cmd2) of the active conditioner (2). The inputs (cmdl, cmd2) of the passive conditioners (4) are controlled at zero by a command from the processing system. A difference from the previous description of this step is that the output quantities of the N conditioners are measured. The amplitudes of the voltages in the initial position are denoted by V1 init for the active conditioner and VJinit for passive conditioners, J varying from 2 to N in the case of a system with N conditioners. When the initial output quantities (V1 init, V2init to VNinit) have been stored in memory spaces (MV1 init, MV2init to MVNinit) by the processing means, the condition (cond01) is satisfied to start the step (ETP02) next. The relations referenced L12 and L13, give the expression of the complex voltage module (V1 to VN) at the output of the operational amplifier of the conditioners (2, 4), for each sensor (F1 to FN), in the initial position . VE represents the modulus of the excitation complex voltage. C1S represents the capacitance measured in the initial position for the active conditioner. CJS represents the capacitance measured in the initial position for the passive conditioner J.

  The verification program performs the calculation of the initial capacities, according to the calculations referenced L12a and L13a. As previously described, the processing system comprises memory spaces (MW, MR, MVE, MV1 init, MV2init MVNinit) comprising data representative of the pulse (W) applied to the active conditioner, the common resistance (R) to all conditioners, the amplitude (VE) of the excitation voltage applied to the active conditioner and the initial quantities (V1 ends, V2init to VNinit). When the results (C1S, C2S to cl N) of these computations are stored (cond02) in memory spaces (MC1S, MC2S to MC1N), the verification program then executes a step (ETP03) for measuring and storing output quantities of the conditioners, in a verification position. The amplitudes of the output voltages (VI, V2 to VN) are measured and / or calculated by the processing means associated with the storage means, via the communication means (5), as previously described. When the amplitudes are stored (cond03) by the processing means in memory spaces (MV1, MV2 to MVN), the verification program then executes an experimental verification calculation step (ETPO4), relating to measured values directly. The calculation carried out by the processing means (P) associated with the storage means is referenced L14 in Appendix 2. In this calculation V1 and VJ, respectively, represents the module of the complex voltage at the output of the active and respectively passive conditioner numbered J. The conditioners Passives are numbered, generally, from 2 to N. When the result (R3) of this calculation is stored (cond04) in memory (MR3), by the processing means (P), the verification program executes a step ( ETP05) for calculating the output quantity of one or more conditioners, measured directly, characteristics of the conditioners and capacitance values initially calculated. The calculation carried out by the processing means (P) associated with the storage means is referenced L15 in Appendix 2. The calculation referenced L15 is derived from the two calculations referenced L15a and L15b. L15b expresses the sum of the sum of the modules of the complex voltages at the output of the operational amplifiers of the passive conditioners. The sum (C1J + CJS) represents the capacitive influence coefficient of the active sensor on the sensor numbered J passive added to the capacity of the sensor J initially calculated. In the relation L15a, CJ1 represents the capacitive influence coefficient of the passive sensor numbered J on the active capacitive sensor. When the result (R4) of this calculation is stored (cond05) in memory (MR4), the verification program performs a step (ETP06) comparing the result of the calculation (R3) of experimental verification with the result of the calculation (R4) forecast. As previously described, the equality of these two calculations, with a determined tolerance, implies that no target is detected and that the variable geometry environment is not disturbed.

  During the execution, by the verification program, of the comparison step (ETP06), the result is valid whatever the position of, for example, the robot or the press and whatever the number of sensors ranging from 2 to N sensors. The previously described steps for N conditioners therefore remain valid for a two-packer system, as well as the set of steps for locating and determining the target (DO). In a nonlimiting manner, a sensor (FS) is produced at ground level, in the variable geometry environment, as shown in FIG. 2. The realization of this planar sensor with a layer or with three layers of different paints, will be described later.

  When the equality is not verified (cond062) the verification program, for example executes a step (ETP08) disturbed or polluted environment, which starts a sequence of steps of location and identification of the target. This step (ETP08) is a step of activating the previously passive sensors, in the active state. During this step, the processing system (M, P) controls all the previously passive conditioners in the active state and controls the previously active conditioner in the passive state. The previously active sensor (F1) is no longer energized. Each of the previously passive sensors (F2 to FN) is excited. The excitation is minimized so that the influence between the capacitive sensors (F2 to FN) is negligible. The processing system (M, P) communicates an excitation voltage amplitude control (VE ') lower than the voltage applied during the preceding steps (VE), in a determined ratio, for example of ten, such that: 10.VE '= VE The new excitation quantity (VE') is stored in a memory space (MVE ').

  After a timer (cond08) for establishing a steady state, the verification program performs a step (ETP81) of measuring the output quantities of the conditioners associated with the previously passive sensors (F2-FN). In a nonlimiting manner, these measurements are carried out simultaneously or successively. The processing system (M, P) measures, via the communication means, the new output quantities (V2 to VN) of the conditioners associated with the newly excited sensors (F2 to FN). When these new quantities (V2 to VN) are stored (cond8l) in memory spaces (MV2n to MVNn), the verification program performs a step (ETP09) of evaluating the capacitive influence of the target on the previously passive sensors. The coefficients (CD2 to CDN) of capacitive influence of the target (DO) on these sensors (F2 to FN) are calculated by the processing means (P), according to the calculation referenced L16 in Appendix 3. In the calculation L16, VE 'represents the modulus of the excitation complex voltage. CDJ represents the capacitive influence coefficient of the target on a sensor numbered J, J varying from 2 to N. CJS represents the initial capacitance of the sensor numbered J, these previously measured capacitances being stored in memory spaces (MC1S, MC2S to MCNS ).

  When the capacitive influence coefficients (CD2 to CDN) of the target (DO) on the sensors (F2 to FN) are stored (cond09) in a memory space (MCD2 to MCDN), the verification program performs a step (ETPIO ) for calculating and storing the capacitive influence coefficient (CD1) of the target (DO) on the previously active deactivated sensor (FI). The sum of the capacitive influence coefficients (CJ1) of a passive sensor (FJ) numbered J on the active sensor (F1) is firstly calculated according to the relation L15c of appendix 2. When this result is stored in a memory space (MSOMCJI), the processing means (P) associated with the storage means (M) calculate the coefficient (CD1) of the capacitive influence of the target (DO) on the previously active deactivated sensor (FI) according to the calculation referenced L17 in Appendix 3. In another embodiment, the verification program performs a step of evaluating the capacitive influence coefficient (CDS) of the target (DO) on the ground sensor (FS). The target (DO) is then supposed to be placed on the ground. This coefficient of capacitive influence is calculated according to the calculation referenced (L16a) in Appendix 3. In this calculation, VS represents the module of the output voltage of the conditioner associated with the ground sensor and CSS represents its capacity initially calculated. These quantities VS and respectively CSS are identical to a module VJ and respectively a coefficient CJS, J having a value between 2 and N. The index of the sensor (FS) sol can be taken from 2 to N, without limitation. When this coefficient (CDS) of capacitive influence is stored in a memory space (MCDS), the verification program calculates the capacitive influence coefficient of the target (DO) on the sensor (F1) deactivated according to the calculation referenced L17a, in appendix 3. The computation referenced L17a corresponds to the computation L17, in which the capacitive influence coefficients of the target on the previously passive activated sensors are taken to zero or negligible, with the exception of the influence of the target (DO ) on the ground (FS). When the capacitive influence coefficient (CD1) of the target (DO) on the previously active sensor (F1) is stored (cond10) in a memory space (MCD1), the verification program performs a measurement step (ETP11) of 'a relocation. During this step, the treatment system (M, P) communicates with the variable geometry environment control machine (6) so as to measure the displacement of the active sensor (FI) arranged, for example on a robot or a press . The control machine (6) communicates information (dl) representative of the displacement of the active sensor (F1) to the treatment system (M, P). In order to carry out this step, the active sensor (F1) must be on a moving part of the environment (VRA) with variable geometry. In an alternative embodiment, as shown in Figures 8 and 9, a first and a second successively active sensors are placed close to each other at a distance (d1) determined. To achieve a variation of the relative distance between the active sensor and the target (DO) these two sensors are activated successively. The second sensor is for example set back a distance (dl) determined with respect to the first sensor. The calculation of the capacitive influence coefficient of the target on the active sensor, with the first active sensor, followed by the same calculation with the second active capacitive sensor is equivalent to a relative displacement of the target (DO) with respect to the active capacitive sensor . When the first sensor is active, the second sensor is passive. When the second sensor is active, the first is passive. In the examples of FIGS. 8 and 9, the two electrodes (12a, 12b) of the two sensors are interlaced and arranged on an insulating layer (11). The insulating layer is disposed on a conductive layer (10) connected to ground. The electrical connection (3) connecting a conditioner (2, 4) to a sensor is connected to an electrode of a capacitive sensor. In a nonlimiting manner, the two electrodes (12a, 12b) are offset so as not to be vis-à-vis.

  When the displacement representative information (d1) is stored (cond11) in a memory space (Md1), the verification program performs a step (ETP111) for activating the conditioner associated with the previously active deactivated sensor (F1) and for deactivation of the conditioners associated with the sensors (F2 to FN) activated previously passive. After a timer (cond111) for establishing a steady state, the verification program performs a step (ETP112) during which the output quantities of the conditioners are again measured and stored. After memorizing (cond112) the output quantities, the verification program executes the step (ETP08) of deactivation of the active conditioner and activation of the conditioners associated with the previously passive sensors (F2 to FN) and then the step (ETP81). ) of measuring the output quantities of the sensors F2 to Fn, then the step (ETP09) of calculating the influence coefficients of the target (DO) on the sensors F2 to FN and finally the step of calculating the new coefficient ( CD1 ne) capacitive influence of the target (DO) on the IF sensor. A difference, compared to the step previously described, is that after the calculation of the new coefficient (CDIne) of capacitive influence of the target (DO) on the sensor F1, the result is stored in another memory space (MCD1ne ). When this new coefficient (CDI ne) of capacitive influence is memorized (condlOn) the verification program performs a step (ETP12) of evaluation of the distance (d01) to the target (DO) and of the surface (SO1) in look of the target (DO). Figure 6 illustrates the calculation modes of the distance (d01) and the surface (SO1) opposite. The coefficient (CD1) of capacitive influence is calculated, for example, according to the calculations referenced L21 or L22, in appendix 3. The capacitive influence coefficient is taken equal to the permittivity (E) of the medium multiplied by the surface (S ) opposite and divided by the distance (dl, dJ). FIG. 6 represents the coefficient (CD1) of capacitive influence as a function of distance (d), according to two point curves (CD1a, CD1b) each representing a surface of the target facing the different sensor. For the same measured value (CD1m) of the capacitive influence coefficient, the slopes (Pa and respectively Pb) are different, and each corresponds to a different distance (da and respectively db). An acceptable warning range (DANG) is for example chosen from 3 cm to 20 cm.

  The verification program calculates the variation (PVar) according to the referenced calculation (L18) in appendix 3, the result being stored in a memory space (MPVar). The verification program then calculates the distance (d01) according to the calculation referenced (L19) in appendix 3, the result being stored in a memory space (MdO1). Then the verification program calculates the area (SO1) next to the sensor (FI), the result being stored in memory (MSOI). In a nonlimiting manner, the distance (d01) and the surface (SO1) or an alert information are transmitted to the machine (6) for controlling the robot or the press. If (condl2a) the calculated distance is less than a minimum acceptable distance, the control machine, for example a robot arm, immobilizes by an emergency control (ETP13), the apparatus of the variable geometry environment. In the case (condl2b) where the distance corresponds to an authorized distance, the verification program performs a triangulation step (ETP14). Thus, the verification program evaluates the distance of the target (DO) with respect to several sensors.

  During the triangulation step (ETP14), the processing means associated with the storage means control a conditioner associated with a sensor, as an active conditioner, the other conditioners being controlled as passive conditioners. In a nonlimiting manner, the active controlled conditioner during this step (ETPI4) is different or identical to the previously active conditioner. A programming mode provides for example to keep the same active conditioner if (condl4a) the calculated distance (d01) does not belong to a determined risk range and stored by the processing system. In the case (condl4b) where the distance belongs to the risk range, the triangulation consists of configuring (ETP15) a new active conditioner and performing a delay for setting (condl5) a transient regime. The verification program then executes the verification process again from the step (ETP03) for measuring and storing the output quantities of the conditioners in a verification position. In a nonlimiting manner, the measurements in the initial conditions are all carried out at the beginning of each cycle, at the beginning of each cycle. The environment is then verified as being unpolluted by a target. In a nonlimiting manner, the measurements are performed with each of the capacitive sensors activated successively. In an alternative embodiment, the target is located by the immobile capacitive sensors in the variable geometry environment. The treatment system excites all the conditioners associated with fixed sensors so that the sensors do not exert influence on each other. The processing means then evaluate the capacitive coefficient of the target on each of the stationary capacitive sensors, according to the calculations referenced L21 and L22, in Appendix 3. The processing means calculate, for example, a ratio of these two coefficients (CDK, CDJ ) of influence. A relation binding the two distances (dK, dJ) as the relation L23 is for example used to obtain a relation of the type L24. This relation is exploited according to various known algorithms. In one variant, a target is localized by its capacitive influence on three fixed sensors linked together by a relation allowing the processing means to determine the position of the target by executing determined algorithms. Figure 10 is a non-limiting example of positioned next to each other. In a nonlimiting manner, electrodes (12) of the capacitive sensors are positioned on an angular sector or on the entire periphery of a part (BRA) of the variable geometry environment. The electrodes (12) are positioned on an insulating layer (11), the insulating layer being positioned on a conductive layer (10) connected to ground. In this embodiment, the same grounded conductive layer (10) and the same insulating layer are used to make a plurality of capacitive sensors. Each capacitive sensor is connected to an active or passive conditioner. The distribution of the sensors, for example, around an axis (dBRA) of rotation of the part (BRA), allows analysis according to the invention, taking into account the orientation of the part (BRA). Indeed the capacitive influence coefficient of the target on each of the capacitive sensors, will have a variable value depending on the distance of the capacitive sensor relative to the target (DO). The highest capacitive influence coefficient determines, for example, the closest capacitive sensor. By combining the various means of localization, according to known algorithms, the location of different parts of a target makes it possible to determine the position of the different constituent elements of the target (DO). The location of the constituent elements of the target thus makes it possible to determine the shape of the target and thus by analyzing the shape, according to known algorithms, of determining the nature of the target introduced into the variable geometry environment.

  In a preferred embodiment the sensors are made with paint having outstanding electrical characteristics. FIG. 5 represents an exemplary embodiment of a sensor according to the invention. The sensor is made, without limitation, by three layers (10, 11, 12) of paint. A lower first layer (10) is produced by an electrically conductive paint. A second layer (11) is made by a dielectric paint. A third layer (12) is made by an electrically conductive paint. The first layer (10) is electrically connected to the non-inverting input of the operational amplifier of either an active conditioner (2) or a passive conditioner (4). This first layer carries out a grounding of the element, such as for example a robot arm (BRA), on which the sensor according to the invention is made. This grounding makes it possible to avoid disturbances coming from the element on which the sensor is placed or conversely the grounding avoids a disturbance of the sensor on this element. The third layer produces the sensor according to the invention. This third layer is electrically connected to the inverting input of the operational amplifier belonging to an active conditioner (2) or to a passive conditioner (4). In another embodiment, the sensor is made by a layer disposed on an insulating material, such as on a concrete floor. The advantages of sensors made with paint, are that the sensors are easily achievable from a technical point of view and their cost is low on the one hand thanks to the materials used and on the other hand because a sensor does not require a particular installation on the machine. In addition its negligible weight allows to have this type of sensor on any part of a machine without requiring a lightening, beforehand, the load of the machine. A robot arm, for example, at a load limited by its available power. Loads displaced by the end of the arm (BRA) of the robot require in particular a lot of power, because of the large lever arms. An increase in the load at the end therefore requires a significant gain in power. On the other hand, the end of a robotic arm (BRA) performs the movements of greater amplitude and is usually the fastest part and the most dangerous part of the robot. The sensor according to the invention, thanks to its negligible weight, has the advantage of being integrated on fast elements and therefore dangerous in a machine, without penalizing the performance of this machine. It should be obvious to those skilled in the art that the present invention allows embodiments in many other specific forms without departing from the scope of the invention as claimed. Therefore, the present embodiments should be considered by way of illustration, but may be modified within the scope defined by the scope of the appended claims, and the invention should not be limited to the details given above.

  ANNEX 1 Time values: (L1) U, (t) + R. Il (t) = UE (t) Complex quantities: (L2) V, C + R.I, c = VE_c ZC1 • 11c = VEC (L3) V2C = - R. 12C (L4) VEC = ZC2. 12C (L5) Scalar quantities: (L6) Case = (Vlinit2 ù VE2) 1'2 / (R.W.VE) These = V2init / (VE.R. W) (L8) Vlinit = VE. (1 + (R.W.C1s) 2) v (L7) V2init = VE.R. WC2S (L9) R1 = V12 -% / init2 + V22 - V2init2 (L10) R2 = 2.V2. [VE.R. W. (C1s ù Ces)] ù 2.VE2.R2. ## EQU1 ## V2 = VE.RW (C12 + C2s) (L11b) APPENDIX 2 Vlinit = VE (1 + (RWC1s) 2)% (L12)

  C1s = (Vrnit2 - VE2) 112 / (R.W.VE) (L12a) VJinit = VE.R. W.CJS (L13) CJs = VJinit I (VE.RW) (L13a) NN R3 = V12 - Vlinit2 + E VJinit2 - EVJ2 (L14) J = 2 J = 2 NN R4 = 2. (VE.R, W) 2. E (VJI (VE.RW) - CJs) • (C1s - E CJs) (L15) J = 2 J = 2 N V1 = VE (1 + R2, W2, [C1S + E CJ1]) 112 (L15a) J = 2 NN 25 EVJ = VE.R. W .. E (C1J + CJS) (L15b) J = 2 J = 2 N N N ECJ, = 1I (VE.R, W). EVJ - E CJs (L15c) 30 J = 2 J = 2 J = 215 APPENDIX 3 CDJ = [(VJ2 ù VE'2) 1/2 I (VE ', R. W)] - CJS (L16) CDS = [ (Vs2 ù VE'2) 1/2 I (VE ', R. W)] û Css (L16a) NNN CD1 = [C1s + LCJ1] û [(C1s + ECJ1 + R3 / (VE.RW) 2] 1 / 2 E CDJ (L17) J = 2 J = 2 J = 2 NN 10 CD1 = [C1s + L.CJ1] û [(C1s + ECJ1 + R3 / (VE.RW) 2] 1/2 - CDS (L17a J = 2 J = 2 PVar = (CD1n ù CD1) / d1 (L18) dO1 = - CD1 / PVar (L19) SO1 = CD1 dO1 / E (L20) CDK = ES / dK (L21) CDJ = ES / dJ (L22) dK = f (du) (L23) CDK / CDJ = dJ / f (dJ) (L24) 15

Claims (14)

  1. A detection system in a variable geometry working environment comprising memory means (M), processing means (P), communication means (5, B), characterized in that it comprises a sensor ( FI) associated with an excited conditioner (2) and at least one passive capacitive sensor (F2 to FN) associated with a non-excited conditioner (4), at least one of the capacitive sensors is arranged on a moving part (BRA) of the working environment, the energized conditioner (2) is electrically connected to the active capacitive sensor (FI) to provide it with electrical excitation power (VE, 11) and comprises means (A02, R) for measuring the variation for charging the active capacitive sensor (FI), each non-excited conditioner (4) is electrically connected to its associated capacitive sensor (F2 to FN) and comprises means (A04, R) for measuring the load variation of its sensor ( F2 to FN) capacitive associated, the conditioners (2, 4) communicate with the processing means (P) and the storage means (M) by the communication means (5, B), the conditioners (2, 4) each produce at least one magnitude (U1, U2 ) representative of the load variation in their associated capacitive sensor, the storage means (M) and the processing means (P) are arranged to calculate (ETP06) at least one relationship linking at least one output quantity of each conditioner, the relation being verified if no target (DO) interacts with the capacitive sensors (F1 to FN), the relation being no longer satisfied if a target (DO) interacts with at least one capacitive sensor.   2. Detection system according to claim 1, characterized in that an alarm signal is sent to a system (6) for controlling the variable geometry environment, via communication means (16), when the relation n is no longer verified.   3. Detection system according to one of claims 1 and 2, characterized in that each conditioner (2,   4) comprises an operational amplifier (AO2, AO4) operating in linear mode, the inverting input of the amplifier being electrically connected to the associated capacitive sensor (F1 to FN), the non-inverting input being electrically connected to a means (G2 ) to produce a voltage, a resistor (R) being electrically connected between the inverting input and the output of the operational amplifier, the output quantity of the conditioner corresponding to the amplitude of the output voltage of the operational amplifier. 4. Detection system according to claim 3, characterized in that the resistance (R) of each conditioner to the same determined value, the means (G2) to produce a voltage are controlled (cmdl, cmd2) via the means (5) communication means, by the processing means (P) associated with the storage means (MVE, MW), the means (G2) for producing a voltage provide a zero voltage for the non-excited conditioners and a sinusoidal amplitude voltage (VE) and pulse (W) determined for the excited conditioner (2).   5. Detection system according to claim 4, characterized in that the output voltages of the operational amplifiers are electrically connected to the communication means (5) which comprise conversion means transforming the analog values of the output voltages into representative digital values so that to be processed by the processing means (P) associated with the storage means (M).   6. Detection system according to claim 5, characterized in that the processing means (P) associated with the storage means (MVE, MW) control a permutation of the active sensor among the capacitive sensors by controlling, to the means (cmdl, cmd2) to produce a voltage, to force the conditioner voltage previously excited to zero, and the voltage of a previously un-energized conditioner to a determined amplitude (VE) and pulse (W) sine.   7. Detection system according to one of claims 5 to 6, characterized in that the processing means (P) associated with the storage means (MVE ') control the means for producing a voltage whose amplitude (VE') is divided according to a determined ratio, so as not to influence the other capacitive sensors.   8. Detection system according to one of claims 5 to 7, characterized in that the processing means (P) associated with the means (MC1S to MCNS respectively) of storage are arranged to calculate an initial capacity (C1S to respectively CNS) for each capacitive sensor, in a determined initialization position.   9. Detection system according to one of claims 3 to 8, characterized in that the capacitive sensor comprises a first layer (10) of conductive paint on which is superimposed a second layer (11) of dielectric paint, on which is superimposed a third layer (12) of conductive paint, the first and third layers being electrically insulated from one another by the second layer (11), the first layer (10) being electrically connected to the ground of the amplifier (A02 , A04) of the conditioner to which the sensor is associated, the third layer (12) being electrically connected to the inverting input of the operational amplifier of the conditioner.   10. Detection system according to claim 9, characterized in that the conductive tab of the amplifier (A02, A04) operational corresponding to the inverting input and respectively to the ground is directly soldered to the third and respectively the first layer of paint the sensor.   11. Detection system according to claim 9, characterized in that a coaxial shielded cable comprising a central electrical wire surrounded by a conductive sheath, the central son electrically connects the third layer (12) of paint from the sensor to the inverting input. of the conditioner amplifier and the conductive sheath connects the first layer (10) of paint of the sensor to the ground of the amplifier of the conditioner. 12 detection system according to claim 10 or 11, characterized in that it comprises at least two capacitive sensors at a distance (dl) determined from one another, intended to be activated successively, to achieve a variation of the distance from the active capacitive sensor to the target (DO). 13. Detection system according to claim 10 or 11, characterized in that it comprises at least a plurality of capacitive sensors arranged on a part (BRA), about its axis (dBRA) of rotation, intended to be activated successively, to determine the orientation of the target (DO) relative to the part (BRA). 14. A method for detecting a target in a variable geometry work environment by a system according to one of claims 9 to 13, characterized in that it comprises at least: a measurement and storage step (ETPO1) magnitudes (V1 init to VNinit) of outputs of the conditioners (2, 4), in an initialization position in which the environment is not disturbed by any target (DO), - a step (ETP03) of measurement and for storing the quantities (V1 to VN) of the conditioners (2, 4) output in a verification position, - a step (ETP06) for comparing an experimental calculation result (R1, R3) (L10, L14) dependent on quantities (V1 init to VNinit, V1 to VN) measured with a result (R2, R4) of a predicted calculation (L11, L15) dependent on at least one output quantity (V2, V2init) of a conditioner and a characteristic value (R, VE) of a conditioner, to determine in case of inequality that a target (DO) e detected or in the case of a tie no target is detected. Detection method according to Claim 14, characterized in that the step (ETPO1) for measuring and memorizing the quantities (V1init at VNinit) of the outputs of the conditioners (2, 4) in an initialization position is followed by a step (ETP02) for calculating and memorizing the initial capacitances (C1S to ci N) of the sensors (FI to FN) in the initial position. 16. The detection method as claimed in claim 15, characterized in that the initial capacitance (C1S) of the active sensor and the initial capacitance (CJS) of a passive sensor are determined by the following calculations: CIS - (Vlinit2 -VE2) ii21 (RWVE) CJS = VJinit I (VE • R • W) in which: CAS is the initial capacitance of the active sensor in the initial position, expressed in farad, - CJS is the initial capacitance of the passive sensor numbered J, in the position initial, expressed in farad, - V1; nit is the amplitude of the output voltage of the active conditioner in the initial position, expressed in volts, - VJ; nit is the amplitude of the output voltage of an associated passive conditioner to a sensor numbered J, in the initial position, expressed in volts, - VE is the amplitude of the excitation voltage of the active sensor, expressed in volts, - W is the pulsation of the excitation voltage of the active sensor, expressed in radians per second, - R is the resistance in condi controllers, arranged between the inverting input and the output of the operational amplifier, expressed in ohm. 17. Detection method according to claim 16, characterized in that the step (ETP03) for measuring and memorizing the quantities (VI to VN) output conditioners (2, 4) in the verification position is followed by a step (ETPO4) of calculating and memorizing the experimental calculation which is followed by a step (ETP05) of calculating and storing the provisional calculation. 18. The detection method as claimed in claim 17, characterized in that the result (R3) of the experimental calculation and the result (R4) of the provisional calculation are determined by the following calculations: NN R3 = V'2 - Vlinit2 + E VJinit2 - FV J2 J = 2 J = 2 NN R4 = 2. (VE.RW) 2. E (VJl (VE.RW) CJs) • (C1s - E Cjs) J = 2 J = 2 in which: 15 - R3 is the result of the experimental calculation, expressed in square volts, - R4 is the result of the provisional calculation, expressed in square volts, - VI is the amplitude of the voltage at the output of the active conditioner in the verification position, expressed in volts, - Vj is the amplitude of the voltage at the output of a passive conditioner 20 associated with a sensor numbered J, in the verification position, expressed in volts, 19. A detection method according to claim 18, characterized in that the step (ETP06) of comparing the experimental result (R1, R3) (L10, L14) with the result (R2, R4) of the predicted calculation (L11, L15), 25 in case of equality at a margin of error determined by, is followed by a jump to the measurement and storage step (ETP03) quantities (V1 to VN) output conditioners (2, 4) in another verification position. 20. Detection method according to claim 19, characterized in that the step (ETP06) for comparing the experimental result (R1, R3) (L10, L14) with the result (R2, R4) of the calculation (L11). , L15), in case of no equality according to the error margin, is followed by at least: 10- a step (ETP08) of deactivation of the active sensor and activation of passive sensors in the active state by a minimized excitation so that the influence between sensors is negligible, - a step (ETP81) for measuring and memorizing the output quantities of the activated conditioners. 21. Detection method according to claim 20, characterized in that it is followed by at least: a step (ETP09) of calculation (L16) and storage of the capacitive influence coefficient (CDJ) of the target ( DO) on each capacitive sensor (FJ) activated, - a step (ETP1O) of calculation (L17) and storage of the capacitive influence coefficient (CD1) of the target (DO) on the sensor (F1) deactivated., 22 Detection method according to Claim 21, characterized in that the coefficient (CDJ) of the capacitive influence of the target (DO) on the activated sensor (FJ) numbered J and the coefficient (CD1) of the capacitive influence of the target. (DO) on the deactivated sensor (F1) are determined by the following calculations: NNN ECji = 1i (VE.R.W). E CJs J = 2 J = 2 J = 2 CDJ = I (see J2 ù VE'2) 1/2 I (VE ', R. W)] - CJs NNN CD1 = [C1s + ECJ1] where [// (C1s + ECJ1 + R3I (VE.RW) 2] 1/2 - E CDJ J = 2 J = 2 J = 2 in which: - CD, / is the capacitive influence coefficient of the target on the activated numbered sensor J, expressed in farad, - CD1 is the capacitive influence coefficient of the target on the deactivated sensor, expressed in farad, - VE 'is the amplitude of the new excitation voltage of an activated sensor, expressed in volts , - Cil is the coefficient of capacitive influence of the activated sensor numbered J on the deactivated sensor, expressed in farad, 23. Detection method according to claim 22, characterized in that it is followed by at least: a step (ETP11) measuring and storing a displacement (dl) of the capacitive sensor (F1) deactivated, the value of the displacement being determined by displacement measuring means in communication with the processing (P) and storage means ( M), - a step (ETP111) of activation of the capacitive sensor (F1) deactivated and deactivation of the capacitive sensors (F2 to FN) activated, - a step (ETP112) of measurement and storage of the output quantities of the conditioners, - a step (ETP08 ) of deactivation of the active sensor and activation of the passive sensors in the active state by a minimized excitation so that the influence between sensors is negligible, - a step (ETP81) for measuring and memorizing the output quantities activated conditioners, -a step (ETP09) of calculation (L16) and storage of the capacitive influence coefficient (CDJ) of the target (DO) on each capacitive sensor (FJ) activated, - a calculation step (ETP10) (L17) and storage of the capacitive influence coefficient (CD1) of the target (DO) on the sensor (F1) deactivated. 24. The detection method as claimed in claim 23, characterized in that it is followed by at least: a distance evaluation step (ETP12) comprising the calculation (L18) and the storage of the variation (PVar) the capacitive influence coefficient (CD1) of the target (DO) on the active capacitive sensor (FI), and the calculation (L19) and the storage of the ratio of the capacitive influence coefficient and this variation (PVar). 25. The detection method as claimed in claim 24, characterized in that the variation (PVar) of the capacitive influence coefficient of the target (DO) on the active capacitive sensor (F1) and the distance (d01) between the sensor (F1). The active capacitive and the target (OD) are determined by the following calculations: SO1 = CDI.dOl / E PVar = (CD1a-CD,)! dl dOl = - CDI / PVar in which: - S01 represents the surface facing the capacitive sensor active with the target, expressed in square meters, - CD1 the capacitive influence coefficient of the target on the active sensor, expressed in farad, - dOl represents the distance between the target and the active sensor, expressed in meters, - E represents the permittivity of the medium between the target and the active sensor, expressed in farad per meter, - PVar represents the variation of the capacitive influence coefficient of the target (DO) on the active sensor (F1), expressed in farad per meter, - CD1n represents the new capacitive influence coefficient of the target on the active sensor after the displacement, expressed in farad, - CD1 represents the coefficient of capacitive influence of the target on the active sensor before the displacement, expressed in farad, - d1 represents the displacement of the target relative to the active sensor (FI), expressed in meters. 26. Detection method according to claim 25, characterized in that it is followed by at least one step of sending an emergency stop message to the variable geometry environment control machine. . 27. The detection method according to claim 25, characterized in that it is followed by at least one step in step (ETP03) for measuring and memorizing the output quantities (VI to VN) of the conditioners ( 2, 4) in a verification position. 28. The detection method as claimed in claim 25, characterized in that it is followed by at least one step (ETP15) for configuring a new active conditioner, a jump to the measurement step (ETP03). and storing the output quantities (V1 to VN) of the conditioners (2, 4) in a verification position.