Technology

We are thrilled to empower the most challenging industrial applications. Every day, we try to go around the laws of physics and get the most out of every photon. The result of this work is our new 3D camera technology that will push the boundaries of 3D vision to an unprecedented dimension.



Tomas Kovacovsky,
CTO

 

 

Tomas Kovacovsky
CTO

"We are thrilled to empower the most challenging industrial applications. Every day, we try to go around the laws of physics and get the most out of every photon. The result of this work is our new 3D camera technology that will push the boundaries of 3D vision to an unprecedented dimension."


We turn top performance 3D vision into reality.

"A whole portfolio of different 3D technologies is available for mastering the many different tasks. One technology, however, we have painfully missed so far: the high accuracy snap-shot area scan of large work areas in motion. Now Photoneo has closed the gap and is presenting the market with the MotionCam-3D (former Phoxi 3D Camera), a unique product to date."

Vision Award 2018 laudatory speech by Gabriele Jansen, Managing Director of the M&A consulting firm, Vision Ventures and member of the VISION Award jury

Time-of-flight

Active stereo

Structured patterns/dots

Parallel structured light

Low resolution

Low accuracy


Microsoft

Sony

PMD

AMS/Heptagon

Panasonic

Odos/Rockwell

Espros

Trade-off reliability

vs.resolution


Intel

ZED

Ensenso/IDS

Low resolution

Low quality features


Apple

Qualcomm/Himax

Mantis Vision

Orbbec

High resolution

High accuracy

High quality


Photoneo
Existing method Photoneo

Understanding the environment by robotic systems will be game-changer for the next industrial revolution. Robots, as machines, are nearly perfect, being precise and robust. However, they are blind. With our 3D vision, they get their human-like eyes, and their usage becomes much wider, improved flexibility.

The eyes and brains of robots comes in the form of new 3D cameras and deep learning systems, able to identify not only what has happened ased on a past experience but to predict what may happen.


Parameters of 3D Sensing Techniques In a Nutshell
White paper by our CTO Tomas Kovacovsky

Machine vision is one of the driving forces of industrial automation. For a long time, it’s been primarily pushed forward by improvements made in 2D image sensing, and for some applications, 2D sensing is still an optimal tool to solve a problem. But the majority of challenges machine vision is facing today has a 3D character. Market offers a wide variety of 3D sensoric solutions, most of them claiming a superiority over their competition. While a lot of these claims are based on rational reasoning, one needs to understand differences and the fact that there is no one optimal solution that satisfies all needs.


 

PhoXi ® 3D Scanner In a Bin Picking Application
Use case paper by our Director of Robot Vision Adrian Kratky

While the idea of the robotic arm picking of randomly placed objects from a container is rather simple to imagine, real implementation brings big challenges. The article elaborates on such challenges experienced in an actual implementation, and the capability of hardware and software developed by Photoneo to tackle the issues. As a result, the 7-second cycle time covers picking a small metal object from a big scanning volume loaded with 10 000 parts, nailed by oriented placing. The area ratio of the size of parts being picked to the bin size is approximately 1:900.


Pathfinder
Paper on AGV Pathfinder, a predecessor of what is now the AMR Phollower 100, co-authored by our colleague Jan Bacik

The latest progress in the field of autonomous navigation in combination with user-friendly interfaces and modern materials enables the transfer of AGV technology, originally developed for industrial plants, into healthcare environment. A year of intensive development resulted in the introduction of Pathfinder – a prototype of what is now Photoneo's autonomous mobile robot Phollower 100. The paper elaborates both on the development and implementation phase of the mobile platform. See it in action on our YouTube channel .


AnyPick

 

AnyPick
Technical paper on Anypick by Michal Malý, Director of AI & Co-founder, Photoneo

AnyPick is Photoneo’s machine learning solution enabling robots to pick randomly placed objects from 1x1 cm. It comes pre-learned so it is not necessary to provide any CAD models of the objects - the AI-based location detection allows the robot to pick objects that it has not seen before. Photoneo provides a complete solution, including a 3D area sensor, calibration & alignment, set-up in a friendly user interface, large gripper database, trajectory planning and the AI/ML module which evaluates the scene, creates a heatmap and suggests gripping points. This paper explains the motivation behind coming up with AnyPick, its advantages and applications for which it is a perfect choice. It also describes how the application works and the results achieved. The end of the paper offers examples where you can see AnyPick in action.

Looking for robot intelligence?
Check out Bin Picking Studio