GB2306825A

GB2306825A - Laser ranging using time correlated single photon counting

Info

Publication number: GB2306825A
Application number: GB9521306A
Authority: GB
Inventors: Andrew M Wallace; John S Massa; Gerald S Buller; Andrew C Walker
Original assignee: Heriot Watt University
Current assignee: Heriot Watt University
Priority date: 1995-10-18
Filing date: 1995-10-18
Publication date: 1997-05-07
Anticipated expiration: 2015-10-18
Also published as: GB9521306D0; GB2306825B

Abstract

A pulse laser rangefinder, adapted for very accurate measurement of short distances, uses time correlated single photon counting to measure the time of flight of the pulses. The return light is attenuated so that the probability of a reflected photon reaching the detector is <5% for each pulse. The laser is pulsed repeatedly at a high repetition rate, and accurate measurements are made of the time of arrival of the photons, from which measurements a histogram of the backscatter signal is obtained. The overall timing of the histogram (80fs) is more accurate than the individual measurements (40ps), giving an accuracy of 30 microns in practice. The results are enhanced by also measuring the distance to a reference surface.

Description

1 Description 1.1 Background Laser ranging systems can be divided into two broad classes, triangulation and time-of-flight, reviewed extensively by Besl [1].

In the former case, a spot, stripe or other laser pattern is projected on to a surface.

The image of the pattern is detected by an imaging sensor from a different viewpoint and simple geometric calculation leads to the determination of the (z, y, z) coordinates. Typical systems can provide sub mm. accuracy at stand-off distances of the order of 0.5m. On a smaller scale, optical heads can be placed on coordinate measuring machines in place of a contact probe to provide local accuracy to at least 0.01 mum. However, triangulation systems require very accurate calibration, and are subject to occlusion. They also have problems to disambiguate the true from spurious reflections.

In a time-of-flight (TOF) system, a laser signal is projected towards a target, reflected and received by a coaxial (usually) detector. The depth is proportional to the measured time delay between the target and a reference signal. Potentially, these are attractive since they can focus on narrow fields of view, are not subject to occlusion and do not depend on complex geometric calibration. TOF systems can be divided into pulsed, e.g. [2], Amplitude Modulated (AM), e.g. [3], and Frequency Modulated (FM) systems, e.g. [4]. In each case, the critical limiting factor is the measurement of the go and return path of the laser beam, which is dependent on the time-response of the transmit and receive devices and associated electronics. In a pulsed system, a range resolution of f2mm at ranges up to 30 metres is typical.The limiting factor is the rise time in the detected pulse. In an AM system, the phase change between a continuously modulated laser reference and target signal is measured, resulting in a typical accuracy of Almm or slightly better over similar range. In a typical FM system, a laser diode's optical frequency is modulated and the range is proportional to the measured phase change between the reference and target signal during a round trip [4].

This appears to give the most accurate results to date, although acquisition times can be long, for example a resolution of +0.lem at a time of 30s for each point [5].

1.2 A new method of range measurement based on the time of flight principle We present a new approach to laser distance measurement, based on a short-pulse TOF technique and employing time-correlated single photon counting (TCSPC). This is a statistical sampling technique with single photon detection sensitivity, capable of picosecond timing resolution. It has been used to good advantage in time-resolved fluorescence and photoluminescence experiments in which optical decay time-constants of tens of picoseconds can be resolved e.g. [6], as well as optical time-domain reflectometry with optical fibres [7].

Here, we present the results of a short experimental study to demonstrate the feasibility of the technique to precise surface measurement, and show good agreement between the results of a simulation and actual distance measurements obtained from a simple plantar reflecting target at short range.

Optical ranging using the TCSPC technique relies on the ability to measure single photon events with a timing accuracy of the order of the interrogating pulse duration. The timing accuracy can then be improved by repeating the measurement 10i - 106 times and then averaging to achieve the desired precision. The object under study (target) is irradiated using a high repetition rate pulsed laser source and the scattered signal arriving back at the detector is attenuated such that the probability of detecting one or more photons is < 5% per pulse.

Under these conditions, if the timing process is repeated over many laser pulses then the time distribution of the single photon events being recorded gives an accurate measurement of the distance to the scattering surface. This process is facilitated by feeding timing pulses from the laser and detector to the start and stop inputs of a time-to-amplitude converter (TAC). The TAC generates an analogue output pulse with an amplitude proportional to the timing difference between the start and stop inputs and can have a resolution of < 5ps. This signal is then digitised by an analogue-to-digital converter (ADC) and used to add a single count to the corresponding channel of a multi-channel analyser (MCA). After many laser pulses the MCA data correspond to a histogram of the back-scattered optical signal versus time.In practice the accuracy of an absolute measurement of the distance will be degraded by drift in the timing electronics, however this can be overcome by introducing a second reflecting surface (reference) and making relative measurements between this and the target surface.

If the error on a single timing measurement is t 40ps and the number of counts for each of the target and reference measurements is t 5 * 105, then a timing accuracy of t Sofas can be achieved, which corresponds to a positional accuracy of ms 12um. In practice, we have demonstrated both experimentally and by simulation the use of a time-of-flight ranging system for the measurement of distances to an accuracy of t 30um. Because the system has single photon detection sensitivity it is particularly suited for studying weakly reflecting or distant surfaces.

References [1] P.J. Besl. Active, optical range imaging sensors. Machine Vision and Appilcations, 1(2):127-152, 1988.

[2] M. Koskinen, J. Typpo, and J. Kostamovaara. A fast time to amplitude convertor for pulsed time of flight laser range finding. In Laser radar VII, pages 128-136, 1992.

[3] R. Grabowski, W. Schweizer, J. Molnar, and L. Unger. Three dimensional pictures of industrial scenes applying an optical radar. Optics and Lasers in Engineering, 10:205 226, 1989.

[4] Glenn Beheim and Klaus Fritsch. Range finding using frequency modulated laser diode.

Applied Optics, 25(9): 1439-1442, 1986.

[5] K. Seta and T. Ohishi. Distance meter utilising the inter beat mode of a he-ne laser.

Applied Optics, 29(3):354-3, 1990.

[6] G.S. Buller, J.S. Massa, and A.C. Walker. All solid-state microscope based system for ps time resolved photoluminescence measurements on II-VI semiconductors. Review of Scientific Instruments, 63(5):2994-2998, 1992.

[7] G Ripamonti, F. Zappa, and S. Cova. Effects of trap levels in single photon optical time domain reflectometry. Journal of Lighlwave Technology, 10(10):1398-1402, 1992.

Claims

2 Claims

1. We have developed a new method of range measurement based on time-correlated single photon counting, within the context of the established principle of time of flight of a laser pulse.

2. We employ a source capable of transmitting pulses of short duration (picoseconds) at high repetition frequencies.

3. We use high speed photon-counting detectors to detect the first received photon re sponse from the target or reference on each pulse. This is capable of fine time (and hence depth) resolution, and measurable response from both highly and lowly reflecting surfaces.

4. We use interference filters to increase the measured response from the reference and target laser reflections with respect to ambient illumination.

5. We use an electronic receiver to process the received photon responses. This includes a time-to-amplitude convertor, an analogue to digital convertor and a multi-channel analyser to form a digitised histogram of the number of received pulses as a function of time.

6. We process the measured histogram by software based on either correlation or pulse modelling methods to obtain an accurate measurement of the separation between the reference and target surfaces.

7. The system is capable of point measurement, but can also be scanned to produce range images of a viewed scene.

8. Method as claimed in claim 1, wherein the system is designed for measurement of range from lem to 100m at resolution of the order of 30um, but which can be employed at greater distances provided an adequate laser return is available.

8. The system is designed for measurement of range from lcm to 100m, but could be employed at greater distances provided an adequate laser return is available.

Amendments to the claims have been filed as follows 2 Claims 1. A method of range measurement based on time-correlated single photon counting, including a laser source, a single photon detector, photon counting electronics, a multi channel analyser and pulse processing software.

2. Method as claimed in claim 1, wherein we employ a source capable of transmitting pulses of short duration (picoseconds) at high repetition frequencies.

3. Method as claimed in claim 1, wherein we use high speed photon-counting detectors to detect the first received photon response from the target or reference on each pulse.

This is capable of fine time (and hence depth) resolution, and measurable response from both highly and lowly reflecting surfaces.

4. Method as claimed in claim 1, wherein we use interference filters to increase the mea sured response from the reference and target laser reflections with respect to ambient illumination.

5. Method as claimed in claim 1, wherein we use an electronic receiver to process the received photon responses. This includes a time-tamplitude convertor, an analogue to digital convertor and a multi-channel analyser to form a digitised histogram of the number of received pulses as a function of time.

6. Method as claimed in claim 1, wherein we process the measured histogram by soft ware based on either correlation or pulse modelling methods to obtain an accurate measurement of the separation between the reference and target surfaces.

7. Method as claimed in claim 1, wherein the system is capable of point measurement, but can also be scanned to produce range images of a viewed scene.