EP1696300A1 - Optical joystick - Google Patents

Abstract

Optical joystick has a radiation device including a radiation source (2) attached to a lever (1), and a number of radiation sensors (3). The radiation device produces a cone of rays (7), which are moveable by the movement of the lever relative to the radiation sensors. The radiation sensors are used for deriving binary signals, which indicate the position of the lever with respect to the radiation sensors. The radiation source and the radiation sensors are supplied with only one power supply. An independent claim is also included for a method of determining a position of a lever with respect to radiation sensors in a joystick.

Description

The invention relates to an optical joystick.

In the automotive industry, especially on special vehicles, such as on forklifts, joysticks are used for ergonomically optimized control of driving and especially for hydraulic controls. A joystick or joystick includes a base and a lever that is movable with respect to the base. Here, the position of the lever relative to the base is converted into electrical signals. The position of the lever is also understood to mean the deflection of the lever with respect to the zero point position. The focus of the joystick applications lies in the two-dimensional XY position determination. Two independent electrical X and Y signals are generated, which are traditionally generated via resistance taps each of a sliding or rotational resistance with linear characteristic. However, there are also one-dimensional controls.

A disadvantage of this realization with resistors is that the mechanical structure is complicated and expensive on the one hand and on the other always subject to mechanical wear. This wear can lead to malfunction or even total failure of the joystick.

From GB 2 334 573 A, a joystick is known, which determines the position of the joystick on the basis of light sensors, wherein the measured light quantity of the light sensors is used to determine the position.

Disadvantage of this solution is a bad linear behavior of the joystick and a low precision.

The object of the present invention is therefore to make available an optical joystick or a method for determining the position which at least partially avoids the disadvantages mentioned. This object is achieved by the joystick according to claim 1 or by a method according to claim 27. Advantageous developments of the invention are defined in the subclaims.

According to the invention there is thus provided a joystick comprising a lever, a radiation device and a number of radiation sensors, wherein the radiation device generates a cone of rays and the cone of rays is movable by moving the lever relative to the radiation sensors, characterized in that binary signals are emitted by the radiation sensors. Signals are derived, which indicate the position of the lever with respect to the radiation sensors. Due to this configuration, the position of the beam cone on the radiation sensors and thus the position of the lever is determined wear-free and also precise. The deriving of binary signals also allows the digital further processing of the information about the position of the radiation source with respect to the radiation sensors, and ultimately the position of the lever relative to the joystick base.

In a preferred embodiment, the radiation device comprises a radiation source, wherein the radiation source is attached to the lever. In a further embodiment, the radiation device comprises a radiation source and a radiation-reflecting device, wherein the radiation-reflecting device is attached to the lever and the radiation-reflecting device can be illuminated by the radiation source.

Preferably, the radiation sensors are arranged as a matrix, in particular as a cross-matrix or as a solid matrix. If the lever of the joystick can only be deflected in two directions, a cross matrix is preferably used. Is the lever of the Joysticks deflected in all directions, preferably a full matrix is used.

In a further preferred embodiment, the radiation sensors are mounted on a preferably flat surface.

Preferably, the flat surface is aligned horizontally with the lever in a zero position. This ensures that the lever always has the same distance to the surface at a predetermined deflection angle of the lever regardless of the deflection direction.

The radiation source comprises at least one light-emitting diode (LED), preferably an infrared light-emitting diode. The radiation sensors comprise light sensors, preferably infrared light sensors. The use of an infrared light-emitting diode and infrared light sensors makes the joystick insusceptible to daylight and artificial light or other extraneous light sources.

The radiation device is preferably constructed in such a way that the cone of rays can simultaneously illuminate a plurality of radiation sensors. By such a structure selected ensures that in each position of the lever always at least one radiation sensor is illuminated. Thus, a position determination can be carried out in any position of the lever.

The radiation sensors can be connected to an electronic circuit for evaluating the position of the cone of rays relative to the infrared light sensors.

The transmitter generates from the derived from the radiation sensors binary signals two analog signals, which represent the position of the beam cone with respect to the infrared light sensors. The analogue signals can then be further processed analogously outside the joystick, for example.

In a preferred embodiment, the sensor matrix comprises a plurality of rows and columns, wherein the radiation sensors of the matrix are connected to each other line by line and column by column. The columns of the matrix are selected in the transmitter in time-division multiplex mode via a line decoder. By connecting the radiation sensors in rows and columns, it is possible to determine the state of all sensors in columns in time-division multiplex mode.

It has proven to be advantageous to connect the rows of the matrix in the evaluation electronics each with a transistor. By means of a transistor, the switching through a line can be controlled.

A transistor of one row will turn on precisely when a predetermined intensity of radiation is incident on an infrared light sensor of that row and if that infrared light sensor is on the column selected by the line decoder. A transistor of one row blocks below the predetermined radiation intensity, this switching point being set via a resistor at the base of the transistor and a capacitor in parallel with this resistor. This ensures that an entire column of the matrix is available at the transistors of the rows for readout. By adjusting the switching point of the transistors, it is possible to determine how much radiation intensity must be incident on an infrared light sensor for the corresponding transistor to switch. Furthermore, a different switching point can be defined for each line.

Via the capacitor, which is parallel to the resistor connected to the base of the transistor, glitches are suppressed, e.g. short, on the infrared light sensors incident infrared light pulses.

For example, a switched transistor represents a binary 1 signal and a latched transistor represents a binary 0 signal.

The evaluation electronics preferably have analog multiplexers for generating analog signals from the binary signals.

In a preferred embodiment, the evaluation electronics have low-pass filters for determining the average values from the analog signals. The low passes allow averaging over the time of the detected analog signals of the rows or columns.

The radiation sensors are preferably arranged symmetrically. Such an arrangement makes it possible that the position of the beam cone with respect to the radiation sensors can be evaluated in all directions with the same accuracy.

Preferably, the binary signals derived from the radiation sensors represent Cartesian coordinates for the lever position.

According to the invention, there is further provided a method for determining the position of a lever with respect to a number of radiation sensors, wherein the beam cone is moved by moving the lever relative to the radiation sensors, characterized in that binary signals are derived from the radiation sensors, which Specify the position of the cone of rays with respect to the radiation sensors.

The beam cone generated by the radiation device is incident on the radiation sensors, wherein the beam cone simultaneously illuminates a plurality of radiation sensors. This avoids that no radiation sensor is illuminated when moving the lever.

Preferably, the radiation sensors are connected to an electronic circuit, wherein the electronic circuit derives from the radiation sensors a number of binary signals and converts them into two analog signals.

As a result, analog signals are made available for further processing, which represent the position of the cone of rays with respect to the radiation sensors.

The binary signals on the rows are derived column by column from the radiation sensors, wherein the selection of the columns takes place periodically in time division multiplex mode and wherein the transistors connected to the rows generate binary 0 and 1 signals by their switching. The binary signals at the columns are generated by a logical AND of all rows with the respective selected column, with all rows being logically ORed together. By this method, the binary states of all radiation sensors can be determined.

From the derived binary signals, analog signals are generated per row and column, the binary 1 signals of each row and each column being represented by analog signals of different voltage. A radiation sensor is thus assigned two - in pairs unique - voltage values with regard to its position in the matrix.

Preferably, all analog signals of the rows and all analog signals of the columns are connected together to form an analog signal for the rows or to an analog signal for the columns. These two analog signals - for rows and columns - represent the voltage value for the current column and row.

From all line analog signals or from all column analog signals, an analog signal is generated, which corresponds in each case to the arithmetic mean value over time. As a result, at least one resolution is achieved, as is achieved with joysticks with potentiometers.

Fig. 1

den Grundaufbau des optischen Joysticks,

Fig. 2

eine Sensormatrix in Form einer Vollmatrix,

Fig. 3

eine erste Ausführungsform der Auswerteelektronik, und

Fig. 4

eine zweite Ausführungsform der Auswerteelektronik.

The invention will be described in more detail with reference to the drawing and two embodiments. In the drawing show:

Fig. 1

the basic structure of the optical joystick,

Fig. 2

a sensor matrix in the form of a full matrix,

Fig. 3

a first embodiment of the evaluation, and

Fig. 4

a second embodiment of the transmitter.

1 shows the basic structure of the optical joystick with lever 1, radiation source 2, radiation sensors 3, evaluation electronics 4, signal lines 5 and beam cone 7, radiation source 2, radiation sensors 3 and evaluation electronics 4 being accommodated in a housing 6 and the signal lines 5 being accommodated in the housing 6 are led to the outside.

This joystick is characterized mainly by a very simple and almost wear-free functioning basic structure.

The radiation source 2 is firmly connected to the lever 1, in such a way that the emitted rays of the radiation source 2 can be incident on the radiation sensors 3. Two flexible cables supply the radiation source 2 permanently with working current. By moving the lever 1, the position of the radiation source 2 thus also moves relative to the radiation sensors 3.

The distance between the radiation sensors 3 to each other and the radiation source 2 is selected such that the beam cone 7 generated by the radiation source 2 from each position of the radiation source 2 impinges at least one radiation sensor 3 at any time. As a result, the risk is met that under certain circumstances no radiation sensor 3 is illuminated by the beam cone 7. By this arrangement, more output voltages representing the position of the radiation source 2 with respect to the radiation sensors 3 can be generated than there are radiation sensors.

The radiation sensors 3 are arranged as a matrix, two special forms being the full matrix and the cross matrix. Other forms of the matrix, such as a triangular matrix, are also possible. Which form of matrix is used essentially depends on the possible deflections of the lever 1, for example. If the lever 1 can be deflected only in two directions, a cross-matrix is preferably used, the radiation sensors 3 being arranged along the directions of movement of the radiation source 2. If the lever 1 is fully deflectable in all directions, a full matrix will preferably be used. In a solid matrix, the entire deflection surface of the radiation source 2 is equipped with radiation sensors 3, wherein the radiation sensors 3 are preferably arranged symmetrically.

The radiation sensors 3 are mounted on a preferably flat surface, wherein the surface is aligned horizontally to the lever 1 in the zero position. To increase the accuracy towards the edges of the surface, the surface may also be formed as a curved surface, thereby ensuring that the illuminated by the beam cone 7 surface always remains constant.

The radiation source 2 is here an infrared light emitting diode. The radiation sensors 3 are designed here as infrared light sensors. It is also possible to use CMOS sensors, such as those used in digital cameras, as radiation sensors 3.

FIG. 2 shows a sensor matrix 9 as a solid matrix with 225 radiation sensors 3. The sensors d1 to d13 here show thirteen of these 225 radiation sensors 3. With full deflection of the lever 1, the surface 19 is illuminated by the beam cone 7. In zero position 1, the surface 20 is illuminated by the beam cone 7.

The matrix of the radiation sensors 3 is connected to an evaluation unit 4. The two embodiments of the evaluation electronics 4 described below use infrared light sensors 3, these being arranged as a full matrix. However, the basic principle of the transmitter 4 is applicable to any form of matrix. For connecting the radiation sensors 3 to the transmitter 4 can be used as a multipolar cable.

The transmitter 4 is used to evaluate the position of the radiation source 2 relative to the radiation sensors 3. The transmitter 4 generates two analog signals from a plurality of binary signals. In this case, in the time-division multiplex operation, all output values read out via the evaluation electronics 4 are detected and processed further for each row and for each column when interrogating the matrix. All values are used to calculate the arithmetic mean over time and thus serve to determine the analog output signals. Due to the structure of the transmitter 4 described below - despite digital detection of the lever position - no approach of a reference point when switching the joystick required. A one-time zero adjustment takes place during the production of the joystick by a calibration. With the transmitter 4 can be almost any output voltages, represented by the two analog signals, between Uout _min and Uout _max realize. U-out _min represents the smallest possible output voltage, Uout _max represents the largest possible output voltage.

The infrared light sensors 3 are connected to one another line by line and column by column and thus form a matrix. In this case, the collectors of the infrared light sensors 3 of a column are connected to one another and the emitters of the infrared light sensors 3 of a row are connected to one another.

FIGS. 3 and 4 each illustrate an embodiment of the evaluation electronics 4. The columns of the sensor matrix 9, to which the voltage, the input signal, is applied cyclically, are used as the lines of the sensor matrix 9 to serve as an exit. In time-division multiplexing, the columns are cyclically set to high level, so that exactly one column is always set to high level. A high-level column represents an "active" column. The control of the columns via a line decoder 8 with downstream pnp transistors in emitter circuit. These transistors thus apply a high level to all collectors of the infrared light sensors 3 of the active column. The clock for the cyclical selection of columns receives the line decoder 8 from a clock 18th

Each row of the sensor matrix 9 is connected to exactly one npn transistor in emitter circuit with high current gain. These transistors operate in switch mode. If an infrared light sensor 3 of the active column strikes radiation of the radiation source 2, this results in a base current at the transistor of the corresponding row. Such a through-connected transistor represents a binary 1 signal with respect to the infrared light sensor 3 of the active column and the line of the transistor, while a non-switched transistor represents a binary 0 signal. Collector resistances against + 5V ensure a trouble-free further processing with TTL level.

In order to effectively suppress diffused infrared light from the environment and very weak infrared light radiation from the radiation source 2, at the base of each line transistor connected to all the emitters of the infrared light sensors 3, there is a resistance to ground and a capacitor in parallel to. The resistor adjusts the switching point and ensures that the line transistors quickly block with very small infrared light radiation. Above all, the capacitor suppresses very short interference pulses, which would lead to a short-term switching of the infrared light sensors 3. Without these measures, very small infrared radiation and very short interfering pulses could negatively affect the evaluation of the measurement signal.

In order to obtain the binary values of the columns, all rows of the sensor matrix 9 are logically linked to the columns of the sensor matrix 9. For this purpose, all outputs of the lines are first of all OR-linked. The resulting new signal is ANDed with the input signal of the active column. So if there is at least one infrared light sensor 3 on the active column, which receives enough infrared light radiation from the radiation source 2 so that the line transistor can switch through the corresponding row, then the input signal of the active column represents a binary 1 signal. the active column, otherwise a binary 0 signal.

The thus obtained binary signals of the columns and rows are converted with analog multiplexer 11 into analog signals with the smallest possible voltage jumps. This achieves an accuracy which at least corresponds to the accuracy of classical joysticks with potentiometers.

The first exemplary embodiment of the evaluation electronics 4 shown in FIG. 3 has a dual analogue multiplexer. The rows of the sensor matrix 9 are each connected to Schmitt trigger inverter 14. The ORing of the rows is realized using Schottky diodes. The anodes of the Schottky diodes are each connected to the outputs of the Schmitt trigger inverter 14. The cathodes of the Schottky diodes are connected together to form the OR-linked signal. The principle described above for determining the binary signals of the columns is realized in this embodiment with the aid of line drivers 13. In this case, each column of the sensor matrix 9 is linked to a line driver 13. The above-described OR-linked lines are linked to the enable inputs of all line drivers and thus constitute an AND connection of the OR-linked lines with the respective column.

In this embodiment, each row and each column of the sensor array 9 is represented by a voltage. In each case one enters for columns and lines loaded low-impedance voltage divider 10, which is each capacitively buffered, each with as many voltage gradations, as columns and rows are available to use. The voltage range between U _min and U _{max of} the voltage divider 10 can be varied. U _min is the smallest on the voltage divider 10 tapped voltage, U _max is the largest voltage divider 10 tapable voltage. The voltage divider 10 are each constructed of exactly one resistor less than columns or rows are present. In this embodiment, the voltage divider 10 consists of the same size resistors. The voltage divider 10 can also be constructed from resistors of different sizes. With in each case different sized resistors, a voltage divider 10 can be constructed with a non-linear rise of the voltages that can be tapped off. A voltage divider constructed in this way can be used, for example, when a positive or negative acceleration is to be controlled by a joystick. With voltage divider 10, there are as many voltage sources as there are rows or columns. Each voltage source represents by its voltage the binary 1-signal of a row or a column. The voltages are provided individually by analog multiplexer 11 for further processing. Each voltage source is connected to an analog multiplexer 11, the voltage source being used as the input signal of the analog multiplexer 11. Each of these analog multiplexers 11 is connected to a row via a Schmitt trigger inverter 14 or to a column via a line driver 13, the signals of the rows or columns serving as control signals of the analog multiplexers 11. The outputs of the analog multiplexers 11 of the rows and the columns are respectively connected to each other and thus represent the current column or row signal in analog form. If, for example, a binary 1 signal is present on one line, the analog multiplexer connected to this line provides the corresponding voltage of the voltage source at the output.

The interconnected outputs of the analog multiplexers 11 for rows and columns are each connected to a low-pass filter 12 with a small time constant. With these low-pass filters 12, the averaging of the analog signals takes place. Thus, an output voltage corresponding to the arithmetic mean value of the voltages over time is generated for the columns or for the rows.

Fig. 4 shows a second embodiment of the transmitter 4 with an analog multiplexer. In contrast to the first exemplary embodiment according to FIG. 3, a common analog multiplexer is used here for generating the analog signals for rows and columns per row and column. For this reason, within a clock in which one column of the sensor matrix 9 is active, the binary signals of the columns and the binary signals of the rows must be processed. For this purpose, a selection signal is generated with the aid of a D-flip-flop, which operates as a 2: 1 divider 15, which provides the binary signals of the lines for processing within a clock initially and then the binary signals of the columns. The 2: 1 divider 15 has two outputs. One output represents the processing of the rows (row output), the other output represents the processing of the columns (column output). The outputs of the 2: 1 divider 15 thus provide alternately a high signal.

As in the first embodiment, the lines of the sensor matrix 9 are each connected to a Schmitt trigger inverter 14 here. In addition, here the enable input of the Schmitt trigger inverter 14 is connected to the row output of the 2: 1 divider 15. This Schmitt trigger inverter 14 switches as only when the row output of the 2: 1 divider 15 carries a HIGH level. The ORing of the lines is analogous to the first embodiment. In addition, the OR-linked signal is applied to an input of a Schmitt trigger NAND gate. The column output of the 2: 1 divider 15 is applied to the second input of this Schmitt trigger NAND gate. The principle described above for determining the binary signals of the columns is realized in this embodiment with the aid of line drivers 13. Here, too, each column of the sensor matrix 9 is linked to a line driver 13.

However, here the output of the Schmitt trigger NAND gate is tied to the enable inputs of all line drivers 13. The outputs of the line driver 13 for the columns thus provide, while at the column output of the 2: 1 divider 15 is applied to a high level, an AND operation of the OR-linked rows with the respective column.

The outputs of the Schmitt trigger inverters 14 are each connected to a line driver 13. These line drivers 13 are only active when the Schmitt trigger inverter 14 is enabled.

The line drivers 13 for columns and rows are thus active alternately within one cycle.

In this embodiment, one row and one column of the sensor matrix 9 are each represented by a common voltage. The voltage sources are generated analogously to the first exemplary embodiment via a low-resistance loaded voltage divider 10. As a result, one voltage source is provided for each row and one column, which voltage represents the binary 1-signal of a row or a column. The voltages are provided individually by analog multiplexer 11 for further processing. An analog multiplexer 11 is connected to each voltage source. The outputs of the line driver 13 for rows and columns are respectively connected in parallel to the inputs of the analog multiplexers 11 and serve as control signals for the analog multiplexers 11. The outputs of the analog multiplexers 11 are interconnected and thus represent the current column or row signal in analogous form, depending on which output of the 2: 1 divider 15 leads to a HIGH level.

This analog signal is forwarded to an analog demultiplexer 16. At the outputs of the analog demultiplexer 16 each have a low pass 12 is connected. The output signal of a low-pass filter 12 represents the analog signal of the rows, while the output of the other low pass 12 represents the analog signal of the columns. The analog demultiplexer 16 forwards the received signal to a respective low-pass 12 with a small time constant. Which output of the analog demultiplexer 16 must pass through the received signal, this decides on its control signals. The control signals are received by the analog demultiplexer 16 from the 2: 1 divider 15. By means of the low-pass filters 12, the averaging of the analog signals takes place. Thus, an output voltage corresponding to the arithmetic mean value of the voltages over time is generated for the columns or for the rows.

In both embodiments, the outputs of the low-pass filters 12 are each connected to a rail-to-rail analog operational amplifier 17, with which an impedance matching is performed and the output voltage is adjusted according to the required output voltage values. In each case, a further low-pass filter is connected to these rail-to-rail analog operational amplifiers 17 with which the alternating components of the analog signals are filtered out. Another analog operational amplifier ultimately sets the output voltage range.

There are now two "pure" analog signals to the outputs of the transmitter 4.

In order to provide two opposite signals per axis, an electronic circuit for generating a second reference signal, an adjustable inverter for both axes, may be connected downstream.

In the embodiments described here, a commercially available infrared LED (CN106, LD274) is used as the radiation source 2. The full matrix 9 is composed of 15 x 15 sensors 3 in SMD technology "1206". The distance of the radiation source 2 to the sensor matrix 9 is about seven centimeters. Using smaller sensors, this distance can also be reduced. The best results for the evaluation are achieved with four to sixteen illuminated sensors 3 simultaneously. The transmitter 4 is here completely constructed with commercially available discrete standard components in CMOS technology. The transmitter 4 is connected to the sensor matrix 9 via a 34-pin cable, with 32 pins occupied. The radiation source 2, the radiation sensors 3 and the entire evaluation electronics 4 work with a single 5V supply. The power consumption of the entire electronics is less than 100 mA. The line decoder 8 is a 4-to-16 line decoder (74HCT154). As the line transistors, high current gain type BC847C transistors are used. The resistors to ground on the line transistors have a resistance of 22 kΩ, the capacitors in parallel to a capacitance of 27 nF. The Schmitt Trigger Inverter (74HCT14) is used. As line driver 13 are each 2 pieces 74HCT241 used. The OR operation is realized with Shottky type LL103 diodes. The Schmitt trigger NAND gate is of type HCT4093. The voltage divider 10 is constructed with fourteen 100 Ω resistors connected in series with a 10 kΩ resistor applied to each voltage tap. As analog multiplexer 11, the type CD4066 is used. The time constant for the low-pass filter 12 is 100 μs. A time constant of 100 ms is selected for the subsequent low-pass filter.

In a further embodiment, a cross-matrix is used, with seven sensors 3 being used for each deflection direction of the joystick along the four directions of movement. When using a cross-matrix, the transmitter 4 is much smaller and easier than when using a full matrix. The states of the sensors of the cross matrix need not be determined in multiplex mode. This eliminates the clock 18 and the line decoder 8. Furthermore, this eliminates the line driver 13, the Schmitt trigger inverter 14 and the 2: 1 divider 15th

In yet another embodiment of the invention, a one-dimensional matrix is used. In this embodiment, movements or deflections of the lever can be detected along a straight line.

With the structure shown at the outset and the analog signals for rows and columns generated with this evaluation unit 4, the joysticks according to the invention achieve at least the resolution of conventional joysticks with potentiometers.

In a further embodiment, the radiation device comprises, in addition to a radiation source, a radiation-reflecting device. The beam reflecting device is attached to the lever 1 of the joystick. Preferably, a mirror is used. The radiation source 2 is mounted in or on the joystick so that it can irradiate the mirror at any time. Preferably, the radiation source 2 is mounted on the same surface on which the radiation sensors 3 are mounted. The radiation source 2 and the radiation sensors 3 can then be supplied with a single power supply. This further increases the robustness of the joystick, since the lever 1 no power is needed. The mirror deflects the incident radiation of the radiation source 2 from, while simultaneously generating a beam cone 7 that the beam cone 7 is incident on the radiation sensors 3 and thus makes the position of the lever using the beam cone 7 determinable. This embodiment also discloses independent of the digital evaluation of the radiation sensors 3 by the evaluation electronics 4 a separate invention.

Used reference signs

1

lever

2

radiation source

3

radiation sensors

4

evaluation

5

signal lines

6

casing

7

ray cone

8th

Line Decoder

9

sensor matrix

10

voltage divider

11

analog multiplexer

12

lowpass

13

Line Driver

14

Schmitt trigger inverter

15

2: 1-splitter

16

Analog demultiplexer

17

Rail-to-rail analog operational amplifier

18

clock

19

Area of the cone of rays at full deflection

20

Area of the cone of rays in zero position

d1 to d13

Sensors of the sensor matrix

Claims (34)

Joystick comprising a lever (1), a radiation device and a number of radiation sensors (3), the radiation device generating a beam cone (7) and the beam cone (7) movable by moving the lever (1) relative to the radiation sensors (3) is, characterized in that of the radiation sensors (3) binary signals are derived, which indicate the position of the lever (1) with respect to the radiation sensors (3). Joystick according to claim 1, characterized in that the radiation device comprises a radiation source (2). Joystick according to claim 2, characterized in that the radiation source (2) on the lever (1) is mounted. Joystick according to claim 1, characterized in that the radiation device comprises a radiation source (2) and a radiation-reflecting device. Joystick according to claim 4, characterized in that the beam-reflecting device is mounted on the lever (1) and that the beam-reflecting device can be illuminated by the radiation source (2). Joystick according to one of the preceding claims, characterized in that the radiation sensors (3) are arranged as a matrix. Joystick according to one of the preceding claims, characterized in that the radiation sensors (3) are arranged as a cross-matrix. Joystick according to one of claims 1 to 6, characterized in that the radiation sensors (3) are arranged as a solid matrix (9). Joystick according to one of the preceding claims, characterized in that the radiation sensors (3) are mounted on a flat surface. Joystick according to claim 9, characterized in that the flat surface is aligned horizontally to the located in a zero position lever (1). Joystick according to one of the preceding claims, characterized in that the radiation source (2) comprises at least one light-emitting diode and that the radiation sensors (3) comprise light sensors. Joystick according to claim 11, characterized in that the light-emitting diode (2) emits infrared light and that the light sensors (3) are infrared light sensors. Joystick according to one of the preceding claims, characterized in that the radiation device is constructed such that the beam cone (7) can simultaneously illuminate a plurality of radiation sensors (3). Joystick according to claim 13, characterized in that the radiation sensors (3) are connected to an evaluation electronics (4). Joystick according to claim 14, characterized in that the evaluation electronics (4) generates two analog signals from the binary signals, which represent the position of the beam cone (7) with respect to the radiation sensors (3). Joystick according to one of claims 4 to 11, characterized in that the solid matrix (9) consists of a plurality of rows and columns, wherein the radiation sensors (3) of the solid matrix (9) are connected to each other in rows and columns. Joystick according to Claim 16, characterized in that the columns of the full matrix (9) can be selected in time-division multiplexing via a line decoder (8). Joystick according to Claim 17, characterized in that the rows of the full matrix (9) in the evaluation electronics (4) are each connected to a transistor. Joystick according to claim 18, characterized in that a transistor of a row then switches when a predetermined radiation intensity incident on a radiation sensor (3) of this line and when this radiation sensor (3) is located above the column selected by the line decoder (8) , Joystick according to claim 19, characterized in that the transistor blocks a line below a predetermined radiation intensity, wherein a resistor at the base of the transistor and a capacitor parallel to this resistor, the switching point of the transistor is adjustable. Joystick according to claim 20, characterized in that via the capacitor, parallel to the resistor connected to the base of the transistor, glitches are suppressed. Joystick according to claim 21, characterized in that a transistor connected through represents a binary 1-signal and a blocked transistor represents a binary 0-signal. Joystick according to one of claims 14 to 22, characterized in that the transmitter (4) has analog multiplexer (11) for generating analog signals from the binary signals. Joystick according to Claim 23, characterized in that the evaluation electronics (4) have low-pass filters (12) for determining the mean values from the analog signals. Joystick according to one of the preceding claims, characterized in that the radiation sensors (3) are arranged symmetrically. Joystick according to one of the preceding claims, characterized in that the binary signals derived from the radiation sensors (3) represent Cartesian coordinates for the lever position. Method for determining the position of a lever (1) with respect to a number of radiation sensors (3) in a joystick, wherein a beam cone (7) is moved by moving the lever (1) relative to the radiation sensors (3), characterized in that the radiation sensors (3) are derived binary signals indicating the position of the lever (1) with respect to the radiation sensors (3). A method according to claim 27, characterized in that the beam cone (7) incident on the radiation sensors (3), wherein the beam cone (7) a plurality of radiation sensors (3) illuminated simultaneously. A method according to claim 28, characterized in that two analog signals are derived from the binary signals. A method according to claim 29, characterized in that the binary signals at the rows are derived in columns from the radiation sensors (3), wherein the selection of the columns periodically in time multiplex operation and wherein the transistors connected to the rows generate binary 0 and 1 signals by switching them, and the binary signals at the columns are generated by a logical AND of all the rows with the respective selected column, all rows being logical OR - linked. A method according to claim 30, characterized in that generated from the derived binary signals analog signals per row and per column, wherein the binary 1-signals of each row and each column is represented by analog signals of different voltage. A method according to claim 31, characterized in that by interconnecting all the analog signals of the columns and all the analog signals of the lines, an analog signal for the columns and an analog signal for the lines is formed. A method according to claim 32, characterized in that from each column analog signals and from all line analog signals each an analog signal is generated, which corresponds to the arithmetic mean over time. Method according to one of claims 27 to 33, characterized in that the beam cone (7) is deflected at a beam-reflecting device and then incident on the radiation sensors (3).

Images