Explore GEN family of websites, social media experiences, blogs and more.

Unleashing ideas.
Strengthening startups.

Back to search results
    Date submitted
  • 17-Nov-2016
    Categories
  • IT

3D vision system on Chip

Abstract

Fastree3D solid-state single chip motion sensor delivers 3D imaging for the new markets of automotive safety and autonomous robotics. We enable high resolution LIDARs at a consumer electronics costs, typically 2 cm voxels, 100 fps, at 30m for less than 200$. The innovation consists of a system on chip CMOS circuit integrating an array(QQVGA) of single photon avalanche diodes (SPADs) together with digital processing for direct time-of-flight and (ToF) dand the control of laser illuminators (VCSEL arrays).

Video

Original YouTube URL: Open

Introduction Video

Photos

Additional Questions

Who is your customer?

We serve two markets : machine vision and automotive safety - To machine vision integrators, we offer hardware development kits and camera modules that can be used in applications including autonomous guided vehicles, bin picking, smart building sensing, people counting. This addressable market segment is $185M growing 30%. To key Tier I - II automotive integrators we deliver flash LIDARs modules in to integrate in advanced driving assistance systems (ADAS). This addressable market segment is $150M growing 34% forecasted $ 2.2B by 2022 .

What problem does this idea/product solve or what market need does it serve?

3D vision is essential for automotive safety, machine vision and even gesture control. Such vision has to be achieved in adverse lighting (low light or too much sunlight). Fastree3D aims at delivering the required performance at mid-range ( 1cm resolution x 19'000 points at 30m) at the cost of consumer technologies (as used in gesture control, gaming). Specifically current 3D cameras have either a limited range and resolution (10cm @ 10m), do not operate in adverse lighting (sunshine or low light), provide ambiguous results (derived from video streams, inaccurate scene segmentation) or are simply too expensive.

What attributes will make this idea/product successful? Why do you believe that those features will create success?

We implement arrays single photon detectors (SPAD) on standard CMOS semi-conductors. Sensors measure digitally the direct time-of-flight from illumination by near infra-red lasers (such as VCSEL arrays). In addition we include in a single chip a processor that can dynamically control illumination based on range / speed requirements and pre-processing, avoiding slow frame based post processing.

Explain how you (your team) will execute to make this idea/product successful? What gives you (your team) an advantage over others already in the market or new to this market?

We have validated each element of the technology and demonstrated at trade shows a lab prototype. Our team has been at the forefront of several 3D sensing projects in the past (100M units shipped). We have been granted 8 patents for various components. A hardware developer kit will be available in 2017 and a full system on chip in 2018.