GB2280260A - Indicating whether or not an object occupies a reference position - Google Patents

Description

INDICATING WHETHER OR NOT AN OBJECT OCCUPIES A REFERENCE POSITION 2280260

This invention concerns apparatus for indicating whether or not an object occupies a reference position.

In particular, although not exclusively, the invention concerns apparatus for indicating whether or not a moving part of a robot occupies a reference position when commended to that position by a control system. Such a moving part may, for example, be an arm or end effector carrying a tool, probe or measuring device.

A moving part of a computer-controlled robot system performing repetitive tasks is often susceptible to drift from pre-programmed positions. There are a number of potential causes for this drift, for example mechanical failures such as slipping couplings; loose belts; motor, encoder and gearbox wear and breakage; and electronic failure of encoders and components; also malicious and accidental damage. The likely result is inaccurate working, non-repeatability of movements and, in extreme instances, damage to the robot or a workpiece and a potential safety hazard for operators. Accordingly it is essential that a moving part of a robot moves at all times with precision to those points in space determined by the computer program controlling the robot.

It is therefore highly desirable to monitor the spatial position and orientation of a moving part of a robot to detect whether drift is taking place and to detect accidental collision of the moving part with the robot or other objects.

2 - Since industrial robots usually perform a predetermined cycle of instructions it is often convenient to monitor the spatial position of a particular moving part of the robot when the robot is at a predefined point in its cycle (usually the start of the cycle).

A prior art system for monitoring the spatial position of an article such as a robot end effector is shown in the accompanying diagrammatic Figures 1(a) and 1(b), in which Figure 1(a) is a side view showing a laser beam emitter 10 mounted on a fixed pedestal 12, an end effector 14 mounted on a robot arm, a photocell 16 mounted on the pedestal, and a meter 18 for indicating a signal generated by the photocell, and Figure 1(b) shows detail of the robot arm end effector.

The robot arm end effector 14 includes an effector block 20 in which there is provided a collimator tunnel 22 arranged to direct incident light onto a mirror 24 within the block, and an exit tunnel 26 to allow incident light reflected from the mirror to exit from the block.

In operation, when the robot is at a particular point in its operational cycle the laser emitter 10emits a laser beam 28 towards the effector 14 in a predefined direction. The collimator tunnel 22, the mirror 24 and the exit tunnel 26 are arranged so that only when the end effector 14 carried by the robot arm is in a predefined spatial position and orientation will the laser beam 28 from the emitter 10 be able to pass down the collimator tunnel 22 onto the mirror 24 and be reflected as beam 28A through the exit tunnel 26 onto the photocell 16, thereby to generate a signal which will be shown via appropriate circuitry on the meter 18. The signal at the meter shows the output from the photocell 16 and allows the equipment to be set up. Further circuitry allows this signal to be compared with a user definable reference signal to allow logic level "good" or "bad" position output signals. The signal will'therefore indicate that the end effector 14 is in the required position and orientation at the particular point in the robot cycle. Conversely, the absence of the signal when the laser beam is emitted at that point in th e robot cycle will indicate that the end effector 14 and hence the robot arm is not in its required position or orientation and that remedial action should be taken.

Although the system described above is satisfactory for most purposes, a theoretical analysis of possible errors in the optical path of the system indicates that in some circumstances it is possible for rotational and positional errors of the end effector to combine to generate an unacceptably large overall error. One way of reducing this overall error is to increase substantially the length of the collimator tunnel 22. However, this results in a bulky system which may impede operation of the robot.

However, this system is insensitive to rotational and -in-line positional errors.

It is an object of the present invention to provide an anDaratus for monitoring the spatial position and orientation of an object such as an end effector of a robot arm, the apparatus having a substantially reduced combined positional and rotational error and a reduced ratio of sensitivities to rotational and positional errors, compared with the system described above.

In its most general terms the present invention monitors the spatial position and orientation of an object by using two laser beams which are reflected by two different reflective surfaces on the object and fixed relative to one another, and both reflected laser beams must hit respective photocells to indicate that the object is in its correct spatial position and orientation.

According to the present invention there is provided an apparatus for indicating whether or not an object occupies a reference position comprises a light source, means for producing first and second parallel and closely spaced, collimated beams of light, first and second detector means fixed relative to the source and spaced apart to receive light beams travelling in different directions, and reflector means carried by the object, said means having first and second angled, reflective surfaces which are arranged, when the object is located at the designated spatial location, to reflect the first and second collimated beams of light towards the first and second detector means respectively.

According to a first embodiment of the present invention, the means to enable the article to produce at least two non-parallel directed beams of electromagnetic radiation preferably comprises laser light source means fixedly mounted independently of the article and arranged to provide at least two directed beams of laser light, and reflector mounted on the article.

The beams of laser light are preferably substantially parallel and the reflector means comprises at least two angled reflective surfaces.

Preferably, the two angled reflective surfaces are angled relative to each other such that when the article is in the reference position at least one laser beam will be reflected by a said reflective surface through substantially an angle of 90 0 to its respective detector, and the other laser beam will be reflected back towards its respective detector in the immediate vicinity of the laser light source.

Alternatively, the two angled reflective surfaces may be positioned and angled relative to each other so that both laser beams are reflected through an angle of about 45 0 in opposite directions.

The means for providing said at least two parallel beams of laser light preferably comprises a single laser source in conjunction with an aperture plate.

According to a second embodiment of the present invention, the means to enable the article to produce at least two non-parallel directed beams of electromagnetic radiation may comprise a source of electromagnetic radiation mounted on the article in cooperation with at least two collimators also mounted on the article.

The object in either embodiment of the invention is preferably an end effector of a robot arm.

The invention will now be described by way of example only with reference to Figures 2-5 of the accompanying diagrammatic non-scale drawings, in which, Figure 1(a) is a side view of the prior art system described above,

6 Figure 1(b) is a detail view of part of the system shown in Figure 1(a), Figure 2 is a side view of a first embodiment of an apparatus according to the invention for indicating whether or not a robot hand occupies a reference position, Figure 3 is a side view of a second embodiment of the invention employing a semiconductor laser diode source and aperture plate beam splitter, Figure 4 is a side view, partly in section, of a third embodiment of the invention, Figure 5 is block diagram of electronic circuitry associated with the invention, and Figure 6 illustrates the effect of errors on the beam positions relative to the detectors in the embodiments.

Referring to the embodiment of Figure 2 there is shown a rigid support member 30 which may for example be a pedestal 31 fixed to the floor, or which may be a fixed part of a rigid framework or a wall.

Fixed to the support member 30 is a laser diode 32 provided with a power supply 34. The laser diode 32 produces a laser beam 48 which is directed at a spatial location indicated generally at 36 which is on a programmed locus of an object such as an end effector 38 of a computer operated robot arm 40.

Fixed to the support member 30 immediately adjacent.the laser diode 32 is a first photocell 42 which is directed at spatial location 36, so that the line of sight from photocell 42 to said location is very nearly parallel to the line of sight from the laser diode to the location.

Similarly, fixed to the support member 30 but spaced from the laser diode 32 is a second photocell 44 which is also directed at spatial location 36. The spacing of photocell 44 from the laser diode 32 is such that the line of sight from photocell 44 to location 36 is substantially at right angles to the line of sight from the laser diode to the location.

There is provided, fixed relative to the laser diode 32 and in its line of sight to the location 36, an aperture plate 46 which is arranged to split the laser beam 48 from the laser diode into first and second parallel laser beams 50,52.

Mounted on the end effector 38 is a single reflector having facets (or reflective surfaces) 54,56 angled relative to one another. When, and only when, the end effector 18 is at location 36 and in its correct spatial orientation, will mirror 54 reflect the first split laser beam 50 to the first photocell 42 as beam 50A, and mirror 56 will simultaneously reflect the second split laser beam 52 to the second photocell 44 as beam 52A.

The output signals from the photocells 42,44 are sent to a logic circuit, indicated schematically at 58 (described in further detail below with reference to Figure 5), by respective electrical signal lines 60,62. The logic circuit 58 is so arranged to accept only simultaneous receipt of output signals from the photocells 42,44 via lines 60,62. Then it generates a pass output signal 59 indicating that the end effector is both at location 36 and in its correct spatial orientation.

The geometry of the optical system described above ensures that the laser beams 50A,52A reflected from the mirrors 54,56 are reflected to the photocells 42,44 at 0 an included angle as close as possible to 90 This ensures that movements of the end effector 38 along the axis of the laser beam 48 are easily detected. However, under some circumstances it may be difficult to set up the system in that photocell 42 has to be located very close to the laser diode in order for the reflected laser beam 50A from mirror 54 to pass back through the aperture plate 46.

A preferred optical system according to the invention is illustrated schematically in Figure 3, wherein parts in common with the arrangement of Figure 2 carry like references. In this arrangement on the rigid support member 30 the laser beam detecting photocells 42,44 are mounted equidistantly on opposite sides of the laser diode source 32. The laser source comprises a I milliwatt semiconductor laser diode and emits a beam 48 which emerges through a window in the device itself has a substantially elliptical bright cross-section. Immediately in front of the device window is positioned a beam-shaping plate 33 which contains a slot aperture. When the apparatus is aligned initially the laser diode is rotated so that the major axis of the bright ellipse of the laser diode is aligned with the longer dimension of the slot aperture.

The distal end of the end effector 38 carries a reflector block 37 having two angled, polished facets which constitute the reflective surfaces 56, 56. These facets for an apex, with and included angle of 135, pointing towards the laser source 32. A short distance in front of this reflective block 37 there is positioned a beam-splitter 39. This comprises a plate in which two circular apertures are formed the diameter and spacing of which are compatible with the dimensions of the shot aperture in beam-shaping plate 33. In the example being described the beam-shaping aperture is a slot 5 mm x 1.5 mm with half-circular ends, while the beam-splitter plate is formed with two circular apertures of 1. 5 mm diameter spaced apart at 3.5 mm centres. During initial setting up the beam-splitting apertures are aligned with beam-shaping slot aperture, the two "split" laser beams 50,52 fall symmetrically on to the reflective surfaces 54,56 respectively, and reflected back towards the photo- detectors 42,44.

The laser diode 32, beam-shaper 33 and photo-detectors 42,44 are all fixed relative to each other, and that the reflective surfaces 54,56 and beam-splitter 39 are likewise fixed relative to each other, but are movable with respect to the laser diode and beam shaper. Bearing this in mind changes in the trajectory of the reflected beams 54,56 caused by errors in the position of the end effector 38 will be readily appreciated. In practice the arrangement is sensitive to translational and rotational errors El-E7 about all six mutually perpendicular axes. Essentially when the end effector 38 is at its reference position, or within a small error distance of it, the reflected beams 50A and 52A will fall onto their respective photo-detectors 42,44 and illuminate them. As the error distance increases progressively less of the reflected beams will fall on to the detectors or the split beams 50,53 will be occluded by misalignment of the beam-shaper 33 and beam-splitter 39. The range of possible error effects is illustrated in Figure 6.

The sensitivity of the system of the system, in terms of its optical alignment, is determined by the relative sizes of the beam-shaper and beam-splitter apertures, and also by the cross-section of the laser beams 50A,52A relative to the light receiving area of the detectors 42,44. Another influence is the optical lever effect inherent in the length of the light paths, in particular the distance between the laser and-the reflector block which we prefer to limit to about 50 mm to 70 mm.

An alternative optical system according to the invention is illustrated schematically in Figure 3, wherein features identical to those in Figure 2 are given the same numbering. The difference between Figure 3 and Figure 2 is that the photocells 42,44 fixed to the rigid support member 30 are both spaced equal distances either side of the laser diode 32 such that the incident laser beams 50,52 are both reflected back to the photocells as beams 50A,52A each at an angle of 45 0 to the axis of the laser diode. Although this arrangement is less sensitive to detection of movements of the end effector 38 along the axis of the incident laser beam 48 than is the arrangement of Figure 2, it is easier to set up.

A third optical system according to the invention is illustrated schematically in Figure 4, in which features identical to those in Figures 2 and 3 are given the same numbering. In this instance, however, the laser source is not mounted on the support member 30 but is instead incorporated in a collimator block 64 mounted on the end effector 38, and is indicated schematically by feature 66. The collimator block 64 is provided with first and second collimators 68,70 at right angles to each other (although other angles may be chosen within the scope of the invention), and the laser source 66 is mounted at or adjacent the intersection of the axes of the collimators 68,70.

The laser source 66 may be a single laser diode and means such as a prism or mirrors 74 may be provided to split the primary laser beam into two beams 76,78 directed respectively along each collimator 68,70. Alternatively, the laser source 66 may be provided by two laser diodes, each directed along a respective collimator 68,70, or it may be provided by one or more laser beams directed along fibre optics from a primary laser source separate from the collimator block 64.

The collimators 68,70 in the collimator block 64, and the photocells 42, 44, are angled and positioned with respect to each other so that when and only when the end effector is at its predetermined spatial location 36 and is in its correct orientation will the photocells receive simultaneous signals via beams 76,78 from the laser source 66 within the collimator block.

Although the embodiment of Figure 4 may seem to have the apparent disadvantage of requiring a laser source to be mounted at the end of a robot arm together with the attendant disadvantage of power lines or fibre optics routed through or attached to the robot arm, there may be occasions when this is preferable to a laser source mounted externally of the robot.

The photocells 42,44 in any of the above embodiments according to the invention are preferably mounted on the support member 30 in such a manner that their positions and angles relative to each other may be easily adjusted in two orthogonal directions and clamped. This enables each direction to be alternately "tuned" for best reception of the incoming laser beam until a maximum is obtained in both directions simultaneously.

The "black box" logic circuitry 58 as shown in Figure 5 utilises continuous voltage readings for each photocell 42,44. These voltage readings and an user-variable reference level are indicated by three analogue meters. If either photocell voltage falls below the reference level a "fail" condition is generated.

The "black box" 58 includes electronic circuitry which discriminates the two photocell outputs and provides a digital pass or fail signal.

The "black box" 58 and laser diode 32 are powered by a regulated 9 V dc mains power supply unit which also provides a 12 V dc supply used to power a flashing xenon beacon 72.

The voltage from each photocell 42,44 is amplified by an inverting amplifier and discriminated separately, because adding the two voltages and discriminating their sum could cause a problem if the reference voltage is set at a low level, ie it would be possible for one photocell to be completely illuminated while the other is in darkness and the total illumination will still be sufficient for a pass condition.

The photocells 42,44 have the same sensitivity control which also acts as a potential divider for the supply voltage. A boost control is provided to increase the gain of the photocell amplifiers. This is useful if small beams are being used or if higher sensitivity is required.

To eliminate the possibility of a noisy or fluctuating signal from a robot positioned on the borderline of its operational envelope, Schmitt Triggers are used as discriminators. These have two preset levels, both of which have to be crossed by the input voltage to effect a change of state.

The outputs from the two Schmitt Triggers are connected to a NOR gate, which only returns a TRUE condition if both inputs are FALSE. This is used as the pass signal, since the inverting photocell amplifiers give a low voltage in full illumination and a high voltage when in darkness. The pass signal is also inverted (using a NOR gate with one input FALSE), to give a fail signal. The pass and fail signals are used to illuminate green and red light emitting diodes, respectively. The pass signal also operates a relay, via a power transistor, which breaks an "Output Fail" connection and makes an "Output Pass" connection. This allows the "black box" to be interfaced to circuitry using any voltage supply.

There are two possible inputs to the circuit, both 24 V dc, which are used to operate relays to convert the signals to 5V. One input turns on the xenon beacon 72 (via another relay and power transistor) and the other turns on the laser diode and xenon beacon.

For high accuracy position checking an intense, narrow low divergence beacon of light is preferred and laser radiation is considered an ideal and readily available source. Form a safety aspect a Class I or Class II laser, as defined by British Standards BS 7192 (1989), have been found technically serviceable, safe and easy to use.

- 14 In particular, a semiconductor laser diode was found to be compact, and to have advantageous beam characteristics. The distributed nature of the output light beam was found especially advantageous. A laser unit of this type which measured 11 mm diameter and 40 mm length emitted a beam of approximately 6 mm diameter. In practice, the beam comprised a bright central bar of about 2 mm width between parallel-fringes so the beam cross-sectional profile could be considered roughly elliptical. This proved ideal in conjunction with the aperture plate beam splitter, for example in the embodiment illustrated in Figure 3 to produce the desired parallel, low divergence beams 50A, 52A.

Considering the photo-sensitive aspects of the detectors 42,44 the photocell should have a small receiving region, roughly the same size as the incident laser beam, for accurate error detection and responsive to the incident beam without being unduly affected by ambient light. The upper portion of Figure 6 illustrates (schematically for clarity) a rectangular profile laser beam incident on a circular photo-cell. It will be appreciated from studying the layout of the embodiments of Figures 2, 3 and 4 that the photocells are quite widely spaced apart, but for the sake of compactness in Figure 6 they are shown closer together in pairs linked by dashed lines.

The uppermost illustration of Figure 6 shows both incident beams falling with perfect alignment onto their respective photocells. This case would occur only when the reflector 38, and therefore by inference the robot arm, occupied the reference position. As has been mentioned above, however, the robot arm may be displaced from the reference position and it is necessary to determine whether or not any error is insignificant or 15 not. Essentially there is a circle of error around each photocell which can be tolerated. This corresponds to a threshold of illumination above which the error is insignificant, but below which the robot must be suspended from operation until corrective action has been taken. The threshold may be varied electronically by altering a discrimination level within the electronics 58. Clearly the threshold level range must begin at an illumination level well above ambient light levels to avoid false triggers. A typical illumination level would equivalent to about 50 of the photocell area illuminated by the laser beam with a range of variation, say, from 25 to 75%. In Figure 3 the resolved components of possible errors are indicated by arrows E2-E7 and the effect of these errors on the relative position of the laser beam and detector photocell at a 50 illumination level are shown in Figure 6 under sub-headings Error E2 Error E7. It will be appreciated that gross errors which result in the laser beam 48 completely missing reflector means 38 fail to illuminate the photocells 42,44 and, of course, are consequently detected.

In operation, the robot's normal control system periodically commands the robot arm to return to the reference position. The normal control feedback circuits will indicate when that position is apparently achieved. The laser position checker is then activated as an independent check that the reference position has been attained within the error threshold limits. Thus by raising the illumination threshold level it will be seen that the positional error allowed is reduced, and reducing the illumination threshold level increases the positional error allowed.

If the detected positional error falls below the set threshold level or fails completely the laser detection electronics output signal 59 is latched to its corresponding state. Preferably this signal is used to apply an inhibit to the robot control system to prevent further operation until the positional error has been corrected and the system reset. Simultaneously the signal 59 is also applied to whatever arrangement is in use to summon operator assistance for remedial action.

In other embodiments of a system of this kind photocells capable of providing information regarding the nature of a detected error were employed. For example, four-quadrant photocells can be used to indicate the direction of shift of the laser beam, as can be readily appreciated from Figure 6. Two-ring (or more) photocells are capable of indicating the amplitude of an error but not its direction. It follows therefore that more sophisticated combinations of photocell arrangements can yield more accurate and comprehensive error information. This information could be used to automatically steer the robot to its zero reference position. This would require, however, a control interface with the robot control system. The system built in accordance with the illustration of Figure 3 required only minimum interface with the normal robot control system, ie in order to give practical effect to the inhibit signal 59. The arrangements described can therefore be carried into practice as entirely separate and virtually independent systems.

One application of the invention has it installed in a manufacturing cell in which high levels of light, such as would be produced by welding arcs, may be present. Such extraneous light sources, probably unsynchronised with the apparatus being described, could saturate the 17 photo-detectors of the invention thereby producing false indications. one solution to this drawback employed modulation of the laser source in combination with demodulation of the photo-detector output signals, in addition a narrow-band optical filter may be located over the photodetector apertures.

Claims (1)

1. Apparatus for indicating whether or not an object occupies a reference position comprises a light source, means for producing first and second parallel and closely spaced, collimated beams of light, first and second detector means fixed relative to the source and spaced apart to receive light beams travelling in different directions, and reflector means carried by the object, said means having first and second angled, reflective surfaces which are arranged, when the object Is located at the designated spatial location, to reflect the first and second collimated beams of light towards the first and second detector means respectively.

   2 Apparatus as claimed in claim 1 wherein the light source comprises a semicircular laser In combination with means for shaping the light beam.

   3 Apparatus as claimed in claim 2 wherein the means for shaping the light beam is fixed relative to light source.

   4 Apparatus as claimed in claim 2 or claim 3 wherein the means for shaping the light beam comprises a plate having a beam-shaping slot aperture formed therein.

   Apparatus as claimed in any preceding claim wherein the means for producing the first and second collimated beams in fixed relative to the reflector means.

   A 6 Apparatus as claimed in claim 5 wherein the means for producing the first and second collimated beams comprises a plate formed with two apertures therethrough spaced apart by a distance equal to a required spacing between the first and second collimated beams.

   7 Apparatus as claimed in claim 6 wherein the apertures formed in the aperture plate are circular.

   8 Apparatus as claimed in any preceding claim wherein the first and second reflective surfaces of the reflector means comprise polished faces of a metal block carried by the object.

   9 Apparatus as claimed in claim 6 wherein the reflective surfaces subtend an included angle of substantially 1350.

   Apparatus as claimed in by preceding claim wherein the first and second parallel beam are reflected by the reflector means through substantially equal angles towards the detector means when the object carrying the reflector means occupies the designated reference position.

   11 Apparatus as claimed in any preceding claim wherein the first and second detector means comprise photo-sensitive detecting means each of which has at least one light sensitive region.

   12 Apparatus as claimed in claim 11 wherein said light sensitive regions have an exposed light receiving surface the area of which is substantially equal to - cross-section area of the first and second light beams.

   13 Apparatus as claimed in claim 11 or claim 12 wherein the photo-sensitive detecting means each comprises a plurality of light sensitive regions disposed in a regular pattern about a nominal position whereby to indicate a displacement error of an incident light beam relative to said nominal position.

   14 Apparatus as claimed in any preceding claim wherein the light source and detector means are attached to relatively fixed parts of a robot apparatus and the reflector means is carried by a relatively movable robot arm or hand.

   Apparatus for indicating whether or not an object occupies a reference position substantially as hereinbefore, described with reference to Figures 2 to 5 of the accompanying drawings.

   16 Apparatus for indicating whether or not an object occupies a reference position substantially as hereinbefore described with reference to Figure of the accompanying drawings.

   i