3.2. Laser rangefinding device

In order to verify the utility of the integrated receiver chips, three prototypes of portable pulsed TOF laser rangefinding devices were constructed. Their smaller size and power consumption relative to the laser rangefinding device constructed using discrete components in electrical parts, shown in Fig. 1a, would also open up new possible applications. The prototypes are presented in papers IV, V and VI (the device described in paper VI is shown in Fig. 1b). The first prototype, presented in paper IV, employs PIN diodes as photodetectors and has the optics integrated into the device, whereas the prototypes presented in papers V and VI make use of APDs and a separate optomechanical measurement head. Typical of all the prototypes was the aim to reduce power consumption, weight and size by means of application-specific integrated circuits (ASICs).

In the first prototype a bare laser diode was used without mixing the signal with optical fibres. This reduces the accuracy, because of the inhomogeneity of the laser beam, but is very simple and easy to construct and is accurate enough for the cm-level accuracy aimed at in that device. In the other prototypes, where the optomechanical measurement head is used, the connecting fibres homogenise the laser beam (Määttä et al. 1990). The length of the fibres, 2 metres, is enough to mix the modes of the beam and the separate measurement head makes it possible to design it with consideration for the requirements of the application (Määttä et al. 1990). The measurement process is controlled using a microcontroller (Motorola MC68HC11) in the first prototype and a FPGA circuit (Altera EPM 7032LC44-6) in the other two devices. A block diagram of the device presented in paper VI is shown in Fig. 18.

The laser pulse transmitter produces an optical pulse after triggering. The trigger signal in the laser radar device developed in paper VI is optical, in order to reduce the interference between the laser pulse transmitter and the other electronics which was recognised with previous prototypes using electrical triggering. The diode used in the transmitter is a multi-quantum well InGaAs pulsed laser diode (PerkinElmer PGAS3S06), which needs a short current pulse of a few tens of amperes in order to give a short, powerful optical pulse. The current pulse is generated by charging a capacitor to 300V with a transformer and then switching the laser diode and the capacitor in series with an avalanche transistor (Zetex FMMT415). The electronics of the laser pulse transmitter additionally generate time jitter in the switching event, thus preventing it from being synchronous with the clock of the TDC, as asynchronous measurement is essential in order to be able to improve the measurement result by averaging (Hewlett-Packard AN 162-1).

The laser diode is coupled to a transmitter fibre (diameter 365 µm) with a pair of lenses and the peak power of the optical signal is 110 W, the FWHM 6.7 ns and the rise and fall times 2.7 ns and 3.0 ns at the output of the fibre. The wavelength of the optical pulse is 905 nm and the pulsing frequency 10 kHz, which gives an average power of 7.4 mW. The optical pulse, measured using a sampling oscilloscope with an analogue bandwidth of 15 GHz, is shown in Fig. 19.

The devices described in papers V and VI employ the receiver channel described in paper I. The linearity error of the device described in paper VI was measured using a calibration track with an accuracy of 1 mm. The measured error together with the amplitude of the stop signal at the input to the timing discriminator are shown as a function of distance in Fig. 20.

The non-linearity at distances shorter than ~10 metres is caused by the vignetting phenomenon of the dual-axis optics used in the measurements (Määttä et al. 1990). At short distances the receiver sees only a small portion of the light spot on the target, actually the portion that comes from the edge of the spot, and these modes of light have travelled a longer way in the fibre, which means that the target seems to be further away. At distances shorter than 3.8m the signal was too weak for measurement with the optics used here. The optics can be customised for the application, and thus the measurement range can also be modified.

The ripple in the amplitude of the signal is caused by the electrical current-mode gain control, which tries to keep the signal at a level of 1.5 V (paper VI). The dynamic range of the optical input signal in the linearity measurement was 1:11 (@ 3.8…33.8 m).

The single-shot precision was calculated from a large number of single-shot measurements at a fixed distance. The level of the input signal, and thus the peak signal amplitude to rms-noise level ratio (SNR), was varied using an optical neutral density filter. The standard deviation of the distance results as a function of SNR is shown in Fig. 21.

The curve in the figure saturates to a level of 4.2 mm with large signals, which is the single-shot precision of the TDC used in the device. The performance of the laser rangefinding devices is summarised in Table 3.

The TDC, which converts the time interval between the logic-level start and stop signals to a digital word, was implemented in an AMS 0.8 µm BiCMOS process, the size of the chip being 12 mm². It operates by digitising the time interval roughly using an 8-bit counter clocked by a 100 MHz external oscillator. This allows a large measurement range. The resolution is improved by interpolating the timing event inside the clock period by means of a dual ramp technique (Räisänen-Ruotsalainen et al. 2000), the final single-shot precision being less than 30 ps, which corresponds to a distance of 4.5 mm. The measured non-linearity of the TDC is ±5 ps in a measurement range from 10 ns to 2.5 µs, which corresponds to less than ±1 mm in the range from 1.5 m to 370 m. The temperature drift is smaller than ±6 mm (±40 ps) in the temperature range –40 to +60°C.

The current consumption of the TDC without the external 100 MHz oscillator is 70 mA from a single 5 V supply (350 mW), and its maximum measurement frequency is 150 kHz.

The amplifier channel and the TDC used in the prototype presented in paper IV are described in more detail by Ruotsalainen et al. (1995) and Räisänen-Ruotsalainen et al. (1995) and the receiver channel and the TDC of that in paper VI by Palojärvi et al. (1997) and Räisänen-Ruotsalainen et al. (2000).

There are not many papers on pulsed TOF laser rangefinding devices in the literature. Samuels et al. (1992) reported on a low-cost, hand-held laser rangefinder for speed detection and law enforcement. The accuracy required in its absolute distance measurements was not very high, as the target would be moving and the primary object of interest was the speed. The resolution of the device was 2.5 feet (~75 cm). The authors decided to use the pulsed TOF topology because a construction using CW techniques with the needed specifications would not have been an eye-safe Class 1 laser product. The parameters of the laser pulser, 30 W output pulse power, a pulse width of 20 ns and a repetition rate of 381 Hz, gave an average output power of 0.23 mW.

Araki & Yoshida (1996) reported on a pulsed time-of-flight optical distance meter for measuring molten steel levels. The measurements were also performed using a piece of black paper as a target, and the measurement range and standard deviation of the measurement error were 1 m and 1 mm respectively.

A commercial pulsed time-of-flight distance meter is available from Banner Engineering Inc. A visible red laser diode is used to produce a pulse with a duration of 10 ns, and the measurement range is from 0.3 m to 5 m with diffuse targets with ±30 mm linearity. The device is intended especially for industrial automation applications.