Patent Application
20070091295
In many applications, it is useful or necessary to measure a distance between two locations. Towards that end, there are numerous known techniques for measuring distance. In some cases, distance is measured mechanically. For example, a wheel, ball or other rolling member can be rolled across a surface. Indexed counters coupled to the rolling member can then be used to determine the distance traveled. Such mechanical measuring systems have a number of disadvantages, however. Notably, such systems typically require physical contact with an object to which (or over which) distance is being measured. In some cases, such contact is not practical. Even when physical contact may not be a problem, the mechanical components of such a system may be relatively expensive and/or the source of other problems (e.g., dirt accumulation).
Distance can also be measured by reflection of energy (electromagnetic or sound) from an object. Such techniques avoid many of the problems with mechanical measuring systems, and offer numerous other advantages. In many of these techniques, a laser is used. Laser range-finding systems can be very accurate. However, known laser range finding systems have their own set of limitations.
One class of laser range finders includes “time-of-flight” (TOF) systems. In TOF systems, light from a laser is reflected from a target and received in a receptor. By measuring the time needed for light to travel from the laser to the target and then back to the receptor, the distance between the laser and the target can be calculated. TOF systems are commonly used for measuring relatively long distances (tens of meters or more). At closer ranges, the travel time for the light is extremely short (tens of picoseconds), and accurate measurement can be quite difficult without the use of expensive detection circuitry.
Another type of TOF system uses the round trip delay time of the laser light to form part of a variable frequency oscillator circuit. The oscillation frequency is then correlated to the distance. Still another TOF system uses a modulated beam and calculates time of flight indirectly by comparing the output beam with the reflected beam. These techniques suffer from limited measurement range, and temperature drift or calibration issues.
Another group of laser range finders includes triangulation-based systems. In these types of systems, light from a laser is reflected from a target and received by a receptor positioned a known distance from the laser emitter. Based on that known distance and the angle of the reflected light, the distance to the target can be trigonometrically calculated. Triangulation-based systems are commonly used for shorter ranges. As the measurement distances increase, the variation in the angle of reflected light becomes quite small. Accurately detecting such small angles can require expensive optics and detection circuitry.
Yet another type of laser range finder utilizes the self-mixing effect. In particular, a portion of light reflected from a target returns to an emitting laser and enters the emitting cavity. The reflected light mixes with light being generated in the cavity and affects the power output of the laser. The power output variations relate to the distance traveled by the light to the target and back. By measuring changes in the laser power output, distance can be determined. Self-mixing-based systems offer significant advantages over other types of laser range finding. Because the emitting laser is also used as a receptor, fewer components are needed. Self-mixing-based systems can also be very accurate. However, self-mixing-based systems also present a number of challenges. The signal generated by self-mixing can be quite noisy, and accurate measurement of the self-mixing effects on laser power output can require relatively complex and expensive circuits. For at least these reasons, such systems have generally not been used in many applications.
This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.
In at least some embodiments, a laser range finder includes a laser and a photosensitive element. The laser is biased with a modulated bias current, and projects a beam onto a target surface in order to determine a distance between the laser and the target surface. A backscattered portion of the laser beam returns to the laser from the target surface and enters the laser emitting cavity. Because of the self-mixing effect, the output power of the laser beam varies in a periodic manner. The variations in output power are detected by the photosensitive element, which provides a “beat” signal to a frequency detection circuit. The frequency detection circuit includes a difference frequency analog phase locked loop (DFAPLL) having a voltage controlled oscillator (VCO) with a center frequency that is substantially greater than expected beat signal frequencies. The DFAPLL mixes the VCO output with a reference frequency to generate a difference frequency signal, with the difference frequency signal providing a purified form of the beat signal. The frequency of the purified beat signal is determined and used to calculate distance to the target surface.
The present invention is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar elements and in which:
Backscattered light 11 from target surface 2 that enters the emitting cavity of VCSEL 5 mixes with the light being newly generated. Because of the self-mixing effect, the power output by VCSEL 5 in outgoing light beam 10 is thereby affected. Self-mixing per se is known in the art, and thus a detailed description of the phenomenon is not included herein. In general, however, the power of beam 10 will fluctuate (or “beat”) in a periodic manner as a result of interference between the outgoing light being generated in the VCSEL emitting cavity and the backscattered light from a target. Although
The periodic fluctuation of the output beam power corresponds to mode hops in VCSEL 5 occurring at target displacements in multiples of λ/2, where λ is the wavelength of VCSEL 5. The wavelength of VCSEL 5 is periodically changed by modulating the bias current with a triangular wave function, so that phase shifts of 2π occur at every resonant mode created by the laser and the target. As is known in the art, the output wavelength of a VCSEL will increase as the bias current for that VCSEL increases. By modulating the bias current for VCSEL 5 with a triangular wave, the frequency of beam 10 will fall during the upslope of the modulating wave and will rise during the downslope of that modulating wave. The distance D can then be determined by finding the frequency spacing between the adjacent modes.
The distance D from VCSEL 5 to target surface 2 (i.e., the external cavity length for VCSEL 5) can be determined from Equation 1.
In Equation 1, c is the speed of light (m/sec). The frequency modulation coefficient Ω, expressed in GHz/mA, provides the frequency of beam 10 for a given bias current over the operating range of VCSEL 5. The coefficient m is the slope of the triangular bias current modulating wave, expressed in mA/sec. The quantity pavg is the average period (e.g., time between peaks) of the beat signal. Because pavg is equal to the inverse of the average frequency (favg) of the beat signal, distance D can alternatively be determined from Equation 2.
Although Equations 1 and 2 are relatively straightforward, various factors can hinder an accurate determination of the distance D. Some of these factors are described by reference to
Although it is relatively simple to determine how many phase reversals to expect in a particular sampling window for a given modulation wave, it is difficult to precisely calculate how (or if) each of those phase reversal will affect determination of favg or pavg. In particular, the beat signal waveform will not always be in the same phase (i.e., the same part of its cycle) at each modulation wave inflection. Thus, the instantaneous phase reversals accompanying the modulation inflections will occur at different phases of the beat signal. In turn, this will affect how many beat signal peaks (or troughs) are counted in a given sampling window. One example of this is seen by comparing
It is possible, however, to reduce the impact of beat signal phase reversals. As shown in Equation 3, the average beat frequency favg is a function of the number of beat signal waveforms (NTm) during a period (Tm) of the modulating wave.
Increasing NTm can increasefavg without increasing Tm. The ratio of beat signal peaks per phase reversals will thus increase, and the phase changes will thus have a smaller cumulative effect on the overall determination of favg.
Unfortunately, NTm increase achieved with a high optical frequency excursion can severely disturb the beat signal envelope. This can make beat frequency detection more difficult. Numerous other factors can also complicate frequency determination. For example, the beat signal frequently has a low signal-to-noise ratio (SNR). This low SNR is caused by, e.g., speckling from the target surface. The strength of the backscattered light received by the VCSEL may also be greatly diminished, particularly for longer distances and/or absorptive target surfaces. The shape of the beat signal waveform also complicates determination of favg. In
VCSEL 5 of sensor 1 is driven by a triangularly-modulated bias current from current generator 25. The beat signal output by PD 6 of sensor 1 is initially fed to pre-amplifier 20. This increases the strength of the relatively low power signal output by PD 6. The output of preamplifier 20 is then fed to triangle wave band eliminator 21. Because VCSEL 5 of sensor 1 is being driven with a triangle-wave modulated bias current, the beat signal will include a harmonic having the triangular wave frequency. Accordingly, triangle wave eliminator 21 subtracts that triangular wave frequency from the beat signal. The output of triangle wave eliminator 21 is then input to band pass filter 22 to remove frequencies outside a predetermined pass range (e.g., 50-500 KHz). This range corresponds to expected values for favg over a range of values for distance D measurable by sensor 1 and detection circuitry 15. The output from band pass filter 22 is then input to a difference frequency analog phase locked loop (DFAPLL) 28 for additional noise reduction. It should be appreciated by those skilled in the art that wider distance measurement ranges are possible with this technique and the 50-500 kHz range is only an example. In practical applications, the measurement range is limited only by the ability of the optics to gather sufficient light from the measurement target to cause the self-mixing effect to occur in the laser diode.
Because analog phased locked loops (PLLs) have good noise rejection and amplitude modulation rejection qualities, they can be used to regenerate a less-noisy version of a noisy input signal. In particular, an analog PLL can be used to enhance the accuracy with which beat signal frequency (and thus, distance) is measured. However, typical analog PLLs have a limited “lock” range of approximately ±20% of the center frequency of the voltage controlled oscillator (VCO) in the PLL. In other words, such a PLL would only be able to reproduce input frequencies that are within 20% of the VCO center frequency. If such an analog PLL were used in detection circuitry 15, the system would be limited to measuring displacements within 20% of some reference displacement value. If for example, a conventional analog PLL was selected based on a reference displacement of 1 m, the system would only be able to measure values of D between 0.8 m and 1.2 m.
In the embodiment of
DFAPLL 28 of detection circuitry 15 includes a phase detector 29, a loop filter 30, a VCO 31, a reference frequency oscillator 32 and a frequency mixer 33. The signal from photosensitive detector 6, after pre-conditioning by pre-amplifier 20, modulation eliminator 21 and band pass filter 22, is input to phase detector 29. Phase detector 29 measures the difference in phase between the beat signal frequency and the output from frequency mixer 33, which is discussed below. A phase difference signal output from phase detector 29 is then filtered by loop filter 30 and fed to VCO 31. Similar to other PLLs, VCO 31 then adjusts its output frequency based on the phase difference signal. Specifically, if the beat signal frequency is lower than the other frequency input to phase detector 29 (i.e., the input received from mixer 33), VCO 31 decreases its output frequency. If the beat signal frequency is higher than the other frequency input to phase detector 29, VCO 31 increases its output frequency.
The output of VCO 31 is fed to mixer 33. Also fed to mixer 33 is a reference frequency generated by reference frequency oscillator 32. In mixer 33, the frequency of the signal output by VCO 31 is reduced (or “downconverted”) by the reference frequency from oscillator 32. The downconverted output from mixer 33 (the difference frequency output signal) is then fed to phase detector 29. As previously indicated, phase detector 29 compares the beat signal with the difference frequency output signal from mixer 33 to generate the phase difference signal. Because VCO 31 continually adjusts its output so as to reduce the phase difference signal, and because the VCO output is frequency downconverted in mixer 33 so as to be within the range of the beat signal frequency, the difference frequency output signal from mixer 33 will match the beat signal frequency once DFAPLL 28 reaches equilibrium. However, the difference frequency output signal from mixer 33 is a purified form of the signal received from band pass filter 22. In particular, processing by DFAPLL 28 removes noise and other anomalies in the beat signal previously described.
The following example further illustrates the operation of DFAPLL 28. Suppose that the beat signal frequency will, for the range of velocities to be measured, vary between 50 KHz and 500 KHz. An analog PLL without a frequency downconversion stage would be unable to reproduce frequencies over that entire range. For example, a VCO with a center frequency of 275 KHz (halfway between 50 KHz and 500 KHz) would only be able to output frequencies between approximately 220 KHz and 330 KHz. If VCO 31 has a 10 MHz center frequency, however, it would be able to output a sufficiently wide range of frequencies. Even if such a VCO has a dynamic range of only 5%, it could output frequencies from approximately 9.5 MHz to approximately 10.5 MHz, a spread of approximately 1000 KHz. Because the output of that VCO would always be higher than the beat signal frequency, however, the VCO output is reduced by a reference frequency (from oscillator 32) of 9.5 MHz. If the beat signal input to phase detector 29 has a frequency of 50 kHz, VCO 31 will (after DFAPLL 28 reaches equilibrium) output a 9.55 MHz signal. After downconversion by the 9.5 MHz reference frequency, a 0.05 MHz (50 KHz) difference frequency output signal is output by mixer 33 and DFAPLL 28. Similarly, a 500 KHz beat signal input to phase detector 29 would cause VCO 31 to ultimately output a 10 MHz signal. After downconversion by the 9.5 MHz reference frequency, a 0.50 MHz (500 KHz) difference frequency output signal is output by mixer 33 and DFAPLL 28.
The difference frequency output signal from DFAPLL 28 is provided to frequency counter 35. Frequency counter 35 then determines the beat signal frequency favg and outputs a value of favg to distance calculator 36. Distance calculator 36 (which may be, e.g., a microprocessor or other logic circuitry) then determines the distance D based on Equation 2. In at least some embodiments, detection circuitry 15 (including distance calculator 36) and/or sensor 1 and/or current generator 25 are contained on a signal integrated circuit chip.
In certain embodiments, a temperature controller 37 is also included. Temperature controller 37 may be, e.g., a thermostatically-controlled thermoelectric device in thermal communication with sensor 1. The response of many VCSELs to bias current modulation is also dependent upon temperature, and maintaining the temperature of sensor 1 at a known level can further increase measurement accuracy. It is to be noted, however, that a temperature controller is not required. Notably, and as discussed below, a prototype of the system shown in
In the embodiments described thus far, a VCSEL was used in sensor 1. Although VCSELs offer numerous advantages over other types of laser devices (e.g., circular beam shape, lower threshold current), other types of laser devices could be used. For example, an edge emitting laser diode (EELD) could alternatively be employed, as shown in
A prototype of the system shown in
In the prototype, the DFAPLL phase detector had a gain Kd of 1.0V/rad, the VCO had a gain KO of 1.08 fo/V, the VCO center frequency fo was 10 MHz, and the reference frequency was 9.5 MHz. The input signal to the DFAPLL had an amplitude of 20-300 mV., and the AM rejection was better than 40 dB. The lock-in range of the DFAPLL was 5 KHz to 1 MHz. Distances (to a sandblasted metal target) between 50 mm and 500 mm were measured. When laser temperature was maintained between 0° and 55° C., resolution was 1 mm and ranging accuracy was within 4 mm. When laser temperature was maintained at 25+/−2° C., resolution was 1 mm and ranging accuracy was within 2 mm.
As can be appreciated from the foregoing description, various embodiments provide a self-mixing laser range finder offering improved ranging accuracy and dynamic range, as well as simple configuration and low cost. Such a range sensor can be advantageously utilized in a number of applications. For example, various computer input devices can employ a range sensor instead of electrical contacts, potentiometers and other conventional mechanisms for detecting movement of a keyboard key, a slider or other type of control piece. Examples of input devices employing range measurement are described in commonly-owned U.S. patent application Ser. No. 10/953,107, titled “Keyboard or Other Input Device Using Ranging for Detection of Control Piece Movement” and filed Sep. 30, 2004. As but another example, a distance sensor according to embodiments of the invention could be employed to determine the distance between a user and a computer display. That distance could then be used to adjust the display contents to accommodate users who may be too far from the screen to, e.g., read normal sized text. As yet a further example, a distance sensor according to embodiments of the invention could be used to track a user's hand as part of an input command (e.g., the user moves his hand away from the computer to zoom in a display). Beyond computer input, an accurate and inexpensive distance sensor has numerous other uses (e.g., in a camera autofocus, in numerous computer vision applications, etc.).
Although examples of carrying out the invention have been described, those skilled in the art will appreciate that there are numerous variations and permutations of the above described devices and methods that fall within the spirit and scope of the invention as set forth in the appended claims. It is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
