AI based object detection
Customized to the required use case, AI networks are selected, and an AI is trained with existing data, or a new AI is trained with data to be annotated.
These data can be, for example,
- recognition of persons,
- warnings or warning signs,
- predefined faces,
- goods specifications,
- vehicle models, and many more.
In case the required hardware (such as RealSense or Jetson Nano) is not available on the respective industrial vehicles, we can remedy the situation with a mobile device, for example to fully automate forklifts of the past generation.
Algorithm based, autonomous loading and navigation
Sensor-assisted collision warning on time
Both mechanical and optical sensors can be used here. Mechanical sensors are more suitable for soiled environments and confrontation with various precipitation (IMU-Robo-Kit).For indoor applications and optimal outdoor environmental conditions, optical sensors can be used for collision warnings. Depending on the application, these can be 2D or 3D depth imaging cameras, as well as lasers and/or TOF cameras.
We are specialized in the analysis of imagery, including point cloud data, and the conversion of 2D to 3D imagery to provide in-depth information. Our advantage is that through our know-how and sensor-based application, there can be no collision between man and machine.
One unique feature is not only the ability to detect downward obstacles such as steps or holes from depth images, but also to count them using AI.
Examples of hardware for the detection of the environment
Applications under iOS with the iPhone – incl. AR
With today’s modern possibilities of 3D-object-recognition, objects can be recognized, captured and even measured. Down to centimetres it is possible to measure goods, rooms, areas and prepare them for further steps. This is used, for example, in logistics to measure the loading area of a truck and find out how many europallettes find space on it – even completely individual non-standard pallettes!
Thanks to the continuous development in the field of AI (artificial intelligence), it is now possible for us to classify objects that have been detected, such as person-recognition, but also for individual applications to build their own AI – for example, if object X looks like a car, it may pass, if not, the barrier remains closed; or people with stored faces may approach the heavy construction machine, for other people, the machine sends alarm. Also, anomaly detection is possible for us! This is used, for example, in production procedures to detect deficiencies. Other applications are sound classification, lane detection (so that a machine follows a line), and many more.
Above the sceneries of the ‘real world’, there is AR – augmented reality. With our algorithms we can set points, which are valid for further measurements as a reference. This can be helpful e.g., for the docking of a truck at the ramp to dock exactly in the centre, or e.g., for the separation of an object in equal distances, without having to measure with the scale. Many other data can be used: Angle, orientation, speed, etc.
Our solution to your problem consists of an iPhone with an integrated, highly efficient processor, a high-resolution camera and lidar sensor also integrated, our algorithm and a multi-member & motivated team of developers who make all this possible for you.
Together, we’ll have a look at the problems to be solved and consider the best solutions as a team. Then we’ll be able to start with a prototype.
If you need support with applying for funding for further steps, you can count on our experience.
We are looking forward to your request!
- the correct loading of vehicles on low loaders
- the correct alignment of vehicles for their application
- control of safe loading of goods
- precise docking of the truck on the loading ramp
- measuring of loading areas and calculation of the goods to be loaded
- beim präzisen Andocken von LKW zu Anhänger, Tieflader oder Baumaschine
- the optimization of delivery times
- and much more
- the calculation for the optimal loading of a truck (quantity of goods) – individually
- the automation of loading by modern forklifts
- faster localization of goods
- to assign goods reference values for further processes
- the measurement and automated calculation of the quantity of goods
- line tracking for transport machines, e.g., forklifts
- and much more
- Detection of defects by anomaly detection (deviations from the normal)
- e.g., damage to a package, a crooked seam on clothing, etc.
- Process optimization through automation
- e.g., by measuring of parts (accuracy +/- 0,5cm), alignment of the goods to be processed, etc.
- Noise detection to detect defects
- Alignment of robots to certain, previously set, reference points
- and much more
Our algorithm can also pass on control commands to machines without interfering with the control system,
because safety is our TOP priority!
Request without obligation now!
Application example: Alignment to a mobile home
Industry Meets Makers
More Infos: LINK