Milos Davidovic

A 16

A time-of-flight depth sensor is presented employing a correlation based concept. The pixels are able to suppress contributions of ambient light up to 150 klx autonomously and directly in the electronic circuits consuming only 2 μA, each. All 16 × 16 pixels capture the depth information simultaneously at maximally 16 frames per second using an external A/D converter (ADC). Additionally, an embedded 12-bit ADC can be used for converting the analog output values of the pixels to the digital domain at lower frame rates. Distance measurements in a range of 0-3.2 m show standard deviations in the centimeter range and a linearity error within -1/+2 cm. The influence of extraneous light is expressed as a relative distance error which is also in the centimeter range. A complete impression of the sensor system is given by discussing the pixel circuits, the ADC design, the design of the modulated light source providing 0.9 W optical power in the near infrared, and the auxiliary setup including an FPGA board. Additionally, a sensor model is provided to predict the measurement error due to stochastic sources. Three-dimensional snapshots as well as a comparison with state-of-the-art sensors and products complete the paper.

\, \times

This work covers a correlation-based, optical time-of-flight matrix range sensor which is able to suppress ambient light up to 150klx full-electronically and autonomously in each single pixel. The CMOS-compatible sensor features 16×16 pixels and permits distance measurements with a standard deviation of 1cm (5cm) up to 1m (3.2m), while the linearity error is within ±2cm for the entire measurement range within a measurement time of 50ms per distance point. A 0.9W LED source is used to illuminate the field of view with near-infrared light, which is modulated at 10MHz. In this design, the current consumption could be reduced from 100µA to 2µA per pixel, while increasing the distance measurement performance at the same time.

16 Pixel Distance Sensor With In-Pixel Circuitry That Tolerates 150 klx of Ambient Light

Within this work an integrated range finding single pixel sensor manufactured in a standard 90nm CMOS technology is presented. The sensor works on the time-of-flight principle obtaining the distance information out of the correlated sent and received signals. The implementation of a range-finding sensor in 90nm technology is using the most advanced process for a distance sensor up to now based on the bridge circuit. Background light suppression is inherently provided in the pixel sensor. The pixel facilitates a high fill factor accounting to 90% at an area of 50 × 64 µm² and has a power consumption of 2 µW. Measurement results show a standard deviation of 2 cm at 1.2 m covering the range from 0.2 to 3.2 m at 120 klx background illumination.

Extraneous-light resistant multipixel range sensor based on a low-power correlating pixel-circuit

In this paper an integrated background light (BGL) immune single-pixel range finding sensor based on the time-of-flight (TOF) principle is presented. The sensor is fabricated in a standard 180 nm 1P6M CMOS process, reaching 40 × 40 µm2 total pixel area, at a fill factor of ∼ 67%. As a key element for the BGL suppression a current-sample-and-hold circuit is introduced. A resulting distance variation as a function of BGL remained in a 1.5 cm range for applied dc light of 180 klx, which is, to our best knowledge, the highest BGL immunity reported so far. Measurement results are carried out at 100 fps, showing a standard deviation of 8 mm at 1 m and ∼ 4 cm for distances up to 3.2 m.

Range finding sensor in 90nm CMOS with bridge correlator based background light suppression

In this paper we present an FPGA based characterization system for our 3D TOF distance sensors supporting up to 128 × 128 pixels. The system is capable of flexibly generating all control signals required for a typical TOF measurement. Their properties can be changed within a very broad range. The cycle-to-cycle jitter of those signals was reduced to 1 ps by ECL circuitry. This is equivalent to a standard deviation of the measured distance of 0.15 mm. Furthermore, the system is able to preprocess the distance information before transferring the data to a terminal PC, which reduces the data load on the USB interface. The system includes an averaging function with a maximum of 256 elements to reduce the standard deviation of precision distance measurement sensors. A novel fiber based setup is introduced to systemize the characterization process. By means of averaging a standard deviation of 2 mm could be achieved with one of our 3D TOF distance sensors.

High dynamic range background light suppression for a TOF distance measurement sensor in 180nm CMOS

In this work an integrated filter-less BiCMOS based color sensor is presented. It is capable of determining the center wavelength of a monochromatic light source in the visible range. This can be used e.g. to determine the three different light colors of an RGB-LED. The detector is based on the effect that the penetration depth in silicon depends on the wavelength of the incident light. It is built up of three vertically stacked pn junctions - a shallow pn junction as blue light detector, a middle pn junction for sensing the green part of incident light and a deep diode to detect red light. The resulting RGB output is used to discriminate between the three different light colors of an RGB-LED. The sensor was characterized with a sweep able monochromatic light source. The sensor is fabricated in a standard 0.6 µm BiCMOS process without using any additional filter layers. This makes it possible to integrate the sensor with other circuitry into a single chip without process modifications.

FPGA based time-of-flight 3D camera characterization system

A systematic investigation of the combination of a reference path and a reference pixel as correction method for a systematic error induced by the light source of a time-of-flight (TOF) distance measurement sensor is presented. A change of the bandwidth of the light source, e.g. caused by drifting temperature of the used LEDs results in a bandwidth dependent distance error. The presented method allows reducing this error over a large operating range by ~97 %, i.e. to 3 %.

Integrated filter-less BiCMOS sensor for RGB-LED color determination

In this work an integrated Time-Of-Flight (TOF) based range finding single pixel sensor in 90nm CMOS is presented. The sensor exploits a PN photodiode for the signal detection, achieving the pixel size of 80 × 86.5 µm2 at the fill factor of ∼ 93%. Measurement results show a standard deviation of 2.5 cm at 1 m and less than 8 cm for distances up to 3.2 m. Ambient light suppression is accomplished inherently in pixel through the bridge correlator circuit approach, whereby the measured distance changes only in the range +3.2cm to −1.2cm for applied DC light up to 180 klx. An inverter feedback ensures a long integration of the modulated signal canceling out the influence of the photodiode capacitance. Due to very small structure sizes, the circuit itself occupies an area of only 13 × 17 µm2 and together with a shrunken photodiode high count multipixel sensors can be realized.

Correction of the temperature induced error of the illumination source in a time-of-flight distance measurement setup

Abstract. Time-of-flight (TOF) range sensors acquire distances by means of an optical signal delay measurement. As the signal travels at the speed of light, distance resolutions in the subcentimeters range require a time measurement resolution that is in the picoseconds range. However, typical clock synthesizers and digital buffers possess cycle-to-cycle jitter values of up to hundreds of picoseconds, which can potentially have a noticeable impact on the TOF system performances. In this publication, we investigate the influence of two common types of cycle-to-cycle jitter distributions on the measured distance. This includes a random Gaussian distribution, which is caused by, e.g., stochastic noise sources, and a discrete jitter distribution, which is found when timing constraints fail in synchronous digital designs. It was demonstrated that a Gaussian cycle-to-cycle jitter has only a negligible impact on the performance of the TOF distance sensors up to a standard deviation of 1 ns of the Gaussian jitter distribution. However, even the discrete cycle-to-cycle jitter investigated in its simplest form lowers the distance precision of the TOF sensor by a factor of 2.86, i.e., the standard deviation increases from 2.9 to 8.3 mm.

TOF range finding sensor in 90nm CMOS capable of suppressing 180 klx ambient light

Time-of-Flight (TOF) 3D cameras determine the distance information by means of a propagation delay measurement. The delay value is acquired by correlating the sent and received continuous wave signals in discrete phase delay steps. To reduce the measurement time as well as the resources required for signal processing, the number of phase steps can be decreased. However, such a change results in the arising of a crucial systematic distance dependent distance error. In the present publication we investigate this phase dependent error systematically by means of a fiber based measurement setup. Furthermore, the phase shift is varied with an electrical delay line device rather than by moving an object in front of the camera. This procedure allows investigating the above mentioned phase dependent error isolated from other error sources, as, e.g., the amplitude dependent error. In other publications this error is corrected by means of a look-up table stored in a memory device. In our paper we demonstrate an analytical correction method that dramatically minimizes the demanded memory size. For four phase steps, this approach reduces the error dramatically by 89.4 % to 13.5 mm at a modulation frequency of 12.5 MHz. For 20.0 MHz, a reduction of 86.8 % to 11.5 mm could be achieved.