This invention relates to the measurement of distance using a light, beam.

There are of course a wide variety of techniques which employ a light beam in the determination of distance, these techniques including interferometric and pulsed time-of-flight measurements. This invention is more particularly concerned with techniques in which a modulated light source is employed and the phase of the reflected light compared with the phase of modulation so as to provide a measure of distance. Apparatus utilising such techniques has already been proposed but there remains a need for apparatus which can provide at reasonable cost, measurement accuracy comparable with, say, steel measuring tapes. It is a difficulty in this regard that the forms of electronic circuitry which are dictated by economic constraints have generally unstable characteristics and offer low signal-to-noise ratios.

It is an object of one aspect of the present invention to provide hand-held apparatus for measuring distance through the use of a light beam, which is capable - at reasonable cost - of measurement accuracy comparable with or better than, measuring tapes.

Laser diodes are now commercially available and it would be convenient to employ these diodes in certain forms of optical distance measurement. Laser diodes are generally robust and relatively inexpensive. Moreover, they offer the fundamental advantage that the amplitude of a light beam can be modulated by simple modulation of the electrical supply to the diode, eliminating the need for a separate amplitude modulating cell in the beam path.

In order to achieve high accuracy, it is necessary to use high frequency modulation; a typical distance measuring device might employ modulation frequencies of 0.5 to 1.0MHz. Difficulties have been encountered using laser diodes modulated at these frequencies. Whilst a repeatable distance measurement can be achieved for a particular target, it is found that small perturbations can produce a disproportionately large shift in the apparent measured distance. Thus, for example, slight movement of the target with respect to the measurement device can produce a difference in the measured distance which exceeds the "real" distance change by an order of magnitude or more. Also, changes in the target surface are found to produce apparent changes in distance.

These difficulties have considerably hampered the use of laser diodes and have held back, it is believed, the development of relatively low cost but accurate laser distance measuring devices. Similar problems may be encountered in other applications of laser diodes and, indeed, whilst the problems have found to be at their most acute with coherent light, related problems may occur with LED's and other light emitting semiconductor devices.

It will be recognised that laser diodes and LED's are available in the IR and UV ranges. Throughout this specification, the term "light" is for convenience used to refer to the emitted radiation and no restriction to the visible spectrum is intended, unless otherwise stated.

Accordingly, the present invention consists, in one aspect, in apparatus for measuring distance, comprising a laser diode arranged to produce a light beam; modulated laser drive means adapted to produce

amplitude modulation in the beam; detector means positioned to detect light reflected by a remote target surface; phase measurement means receiving an output from the detector means and adapted to measure a phase difference with respect to the phase of said modulation and processor means adapted to provide a measure of the distance to the target surface from said phase difference; wherein there is provided scrambling means serving to introduce in the beam a time varying spatial randomisation of amplitude or phase or both to average the effects of a time lag between commencement of light emission by respective different regions of the laser diode.

The present inventors have recognised that the observed discrepancies in measurement using a modulated laser diode are related to the fact that not all parts of the laser diode switch on simultaneously. There is a time lag between commencement of light emission from different portions of the active region of the diode. In the great majority of applications of laser diodes, this time lag (which may typically be of the order of nanoseconds) has no observable effect. However, where the time lag is such that the distance of travel of light in the time lag period is comparable with the desired distance resolution, there are perceptible effects.

Consider the instantaneous injection of current into a laser diode. Different parts of the active region of the diode will begin emitting light at slightly different points in time. For a particular device at a particular temperature, the behaviour may be repeatable, that is to say- different portions of the active region may be switched on in a set order. It will be understood that the wavefront which is formed by

coherent integration of all elemental points according to Huygens' principle has phase and amplitude variations which are both spatially and temporally variant. When this complicated but possibly repeatable wavefront hits a target, the surface structure of the target will form an unpredictable combination of phase variations between scattering centres coupled with a temporal variation which varies from point to point. The summation of the resulting wavelets in a detector system, whilst probably repeatable for an unchanging light path, will be totally dependent on which wavelet happens to be in phase with which other. Thus, small changes in target position or in the proportion of the original beam which is detected, can change dramatically the apparent switch-on time of the diode. Measurements of distance are of course directly related to time measurements and this phenomenon therefore explains the observable large shifts in apparent distance on small changes in target position.

The solution offered by the present invention is to introduce into the beam a time varying spatial randomisation of amplitude or phase which preferably varies over an interval of time, short compared with the period of modulation. In this way, an effective averaging is conducted over the switch-on characteristic of the laser diode.

This invention will now be described by way of example with reference to the accompanying drawings in which:-

Figure 1 is a schematic diagram illustrating apparatus according to the present invention;

Figure 2 is a block diagram of the electronic circuitry forming part of the apparatus according to this invention; and

Figure 3 is a plot illustrating timing measurements made by the circuitry of Figure 2.

Referring initially to Figure 1, the apparatus comprises a laser diode 10 operating at 66θnm to 68θnm provided with a pair of lenses 12, 14 serving to bring the laser beam to an intermediate focus 16. A short distance beyond the focus 16, there is positioned the diffusing element 18 of a diffuser 20. The construction of this diffuser and the functior. it performs will be discussed in detail hereafter. An inclined glass plate 22 serves as a beam splitter such that the major proportion of the laser power is output as a parallel beam 24 through a window 2ό and reflected off mirror 28 towards the target. A minor proportion of the laser output from the beam splitter is directed off mirror 30 in a beam 32 through a window 34 and thus to a further mirror 36.

Light reflected from a remote target surface is collected by a receiving lens 38 and focussed onto one end of an optical fibre 40 serving to transmit light to a detector 42 in the form of an avalanche photo diode. The same lens serves to collect light from the beam _ 2 reflected by mirror 36.

A beam interrupter 44 is continuously rotated by means of a motor 46 and is shaped so as to interrupt the light beams 24 and 32 alternatively. For a calibration interval, light is thus reflected directly to the avalanche photo diode, without passing to the target. This establishes a known path length against which the distance measurement can be calibrated.

Referring now to Figure 2, a modulation oscillator 50 is connected with a laser driver _ 2 so as to provide modulation of the laser output at 450kHz.

The photo diode 42 is connected with a current to voltage converter 54. A local oscillator 56 running at 454.5kHz provides an output to a mixer 8 which also receives an output from the current to voltage converter 54. It will be understood that in this way, the mixer _ provides a low frequency signal at 500Hz in which the phase of the essentially multiplexed reference and target signals is preserved. The mixer output passes through an automatic gain control unit 60 to a zero crossing detector 62 providing multiplexed outputs to separate channels of a counter 64. This counter receives a 100MHz clock signal from clock 66 and provides outputs, as will be described, to a processor 68. This processor may take the form of a commercially available microprocessor. In addition to a keypad 70 and display 72, the processor 68 is connected with a calibration motor detector. This provides, in any convenient fashion, a signal which is synchronous with the rotation of the interpolator motor 46.

Referring now to Figure 3. there are shown the phase displaced outputs corresponding with the target and reference signals respectively. In theory, both signals would have an exact 0: 0 mark space ratio with identical period. That is to say in theory T- = T3 = T5 = T6 and the phase difference at the leading and trailing edges would be equal, that is to say (in terms of timing) T7 = Tδ. However, due to comparative offsets and noise, particularly with the target signal, these equalities do not apply and the relationships will change from one cycle to the next. The simplest approach is to carry out the necessary compensation mathematically so that the time difference between assumed zero crossing points is calculated as:-

T7 + T2 + T3 - T5 + T6 2 2

The phase difference is accordingly given

(T7 + T2 + T3 - T5 ÷ T6) x 2ττ

2 2 T2 + T3

The value for the period (T2 + T3) is taken from the reference signal channel which is likely to be less noisy.

Because of the relatively poor signal to noise ratio expected with a necessarily weak signal, the preferred embodiment of this invention involves the taking of a plurality of measurements. With the nominal frequency of the phase signals of 500Hz, and an acceptable delay of approximately 1 second before a measurement is produced in ideal conditions, it will be seen that 500 readings can be taken. It would be possible to double the number of readings per second by taking calculations also on the falling edges of the two wave forms but since this would not be a wholly independent measurement, the overall improvement in accuracy may not be significant.

The microprocessor 68 is programmed in a manner which will be self evident to the skilled man and requires no further elaboration here, to calculate continuously the mean and standard deviations. In one arrangement, the microprocessor is programmed to cease taking measurements once a preset confidence level (expressed in terms of the standard deviation and mean) has been achieved. In the case of strong signals, this may take only a few hundred milliseconds. With weak signals, it might take up to two seconds. If a reading to the preset confidence level cannot be made within two seconds, the microprocessor will be programmed to display the calculated means together with an

indication of the confidence to be attached to that measurement. For example, the display could include indicator lights associated with different classes of accuracy as defined for measurement tapes. An additional indicator "Estimate" can be provided for measurements not meeting the class standards. Additionally or alternatively, the number of significant figures in the display of the calculated mean measurement could be truncated so as to indicate the variation of measurements. Thus a calculated mean of 30.123 would be displayed as 30.1m indicating an error of ±O.lm.

It will be recognised that the described measurement introduces a technical ambiguity in that phase differences of δ and 2π + δ cannot be distinguished. However, with the described modulation frequency of 450kHz, the distance at which a 2τr phase shift occurs is approximately 300 metres. This will be beyond the normal range of the described apparatus. In other arrangements, the phase ambiguity may be important and the present invention therefor provides a simple manner by which phase differences of δ and 2π + δ can be distinguished. A frequency shift of 0.1i can be introduced on demand, introducing a phase shift of O. %. Clearly, the measured phase shift (on introduction of the modulation frequency shift) will be significantly greater in the case of 2π + δ phase difference than a phase difference of δ. The two distances can therefore be readily distinguished.

The diffuser 20 and its function will now be described.

A glass substrate 30 is provided with a thickness of typically 2mm. To one side, there is positioned a layer of polycarbonate sheet 32 having a microscopically rough surface facing the glass substrate, with the opposite surface smooth. The thickness of the poly carbonate sheet

is typically 0.25mm. Between the glass substrate 30 and the polycarbonate sheet 3 , there is provided a bonding layer 3 of UV setting cement. This has a refractive index of around 1.5 which is much closer to that of the polycarbonate sheet than the refractive index of air. The effect of the bonding layer, which of course has a surface conforming exactly to that of the microscopically rough surface of the polycarbonate sheet, is to dramatically reduce the diffusing effect of that surface, as compared to the effect of a boundary in air. Typically, the deviation of a light ray at the polycarbonate/bonding layer boundary is approximately 85 _/ - less than that of the polycarbonate sheet in air.

The effect of the diffuser on the beam is to introduce phase shifts which vary across the beam width in accordance with the microscopic structure of the diffusing boundary. As the diffuser is rotated these different phase shifts will change with time. The effect is to ensure that the detected light is averaged so as to remove any error arising from the turn-on (or turn-off) characteristic of the laser diode. The period over which the averaging takes place will be short compared with the period of measurement. In one case, a large number of measurements may be taken in a relatively short period with a mean being taken as the measurement result. The scrambling according to this invention can be arranged to provide an average over the total measurement period although it is preferred that there is a degree of averaging within each defined measurement interval within the overall measurement period.

The level of beam dispersion introduced by the dispersing element is determined not only by the structure of the diffusing element but also by its location relative to the focus 16. Moving the diffusing

element away from the focus 16 will have the effect of increasing dispersion but will increase the projected spot size. Care should be taken that the projected spot size is not increased to a point at which the detected image extends beyond the detector active region. It will be apparent that whilst the described diffusing element has the advantage of providing a weak, spatially varying diffusing action in a simple construction, a wide variety of alternatives exist. Similarly, the necessary degree of time variance can be produced in ways other than by revolving a diffusing disc. A vibrating diffusing element could be employed or an alternative construction used in which means other than displacement are used to introduce time variance.

Thus, for example a multi-element LCD array can be envisaged which is analogous in construction to an LCD display device. The array is positioned in the beam path and the individual elements activated in a time varying manner to introduce phase shifts in differing parts of the beam, similarly to the rotating diffuser.

An alternative procedure is to produce scrambling by placing an amplitude mask in the beam path. The mask would have a regular or irregular pattern blocking or attenuating parts of the beam. Time variance can be introduced by rotating the mask as with the described diffusing element.

Still further techniques exist for achieving the desired end of scrambling the beam. In certain cases, for example, it will be feasible to rotate the target. The changes in time of the spatial orientation of scattering centres in the target surface relative to the beam will produce a scrambling effect which, whilst not as efficient as that of

- li ¬

the described diffuser, will overcome some of the problems referred to. It will be understood that relative rotation between the beam and the target surface can also be achieved by effective rotation of the beam either through physical spinning of the laser diode or through an appropriate rotating optical arrangement.

An alternative is to sweep the beam over the surface of the target so as again to change the spatial orientation of scattering centres with respect to time. A rotating, large diameter, narrow angle wedge can be placed in front of both the laser and the receiving lens. The laser spot can then be made to rotate on the target while the image on the photo detector follows it. A variation of this would be an inclined parallel glass disc or prism system rotating in front of the laser and receiver lens to displace the laser beam and its image by an amount, rather than to deflect it through a small angle. This has the advantage of preventing the laser spot from becoming unnecessarily large at longer distances. A combination of the two methods could also be used. A still further alternative would be the use of a vibrating mirror controlled using two actuators in quadrature to provide a circle or elypse of varying radii.

It is proposed that the described apparatus will be provided in a torch shaped body having a display and a simple keypad. Use will be very simple and the fact that the laser will on most surfaces produce a visible dot enables the user to see exactly from which point the measurement is being taken. The apparatus is designed to operate at maximum range on surfaces of 60% reflectivity. For use with less reflective surfaces, separate target reflectors can be provided; these

can be of self adhesive disposable form, magnetic, clip-on or a variety of other forms. The targets are preferably retro reflective, including for example an array of corner cubes. It is expected that the use of a reflector will considerably increase the range of the apparatus. Separate filters can be provided to reduce the amount of ambient light where this is excessive and to attenuate the laser beam in the case of short range measurements or with a highly reflective target. A supplementary lens may also be provided to increase the range of the apparatus in conditions where ambient light levels are low.

In a preferred form of this invention, a multi-element receiving lens is used as described and claimed in copending patent application GB 9020285.4 of the first named applicant.

This invention has been described by way of example only and a variety of modifications as possible without departing from the scope of the invention. For example, while the invention has been described using commercially available laser diodes, other semiconductor devices having an active region adapted to emit light on electronic excitation may be suitable in certain applications. It is necessary for the beam which is produced to have low divergence over the range of interest, taking into account the appropriate maximum size of detector optics.