Sensors use deep learning to automatically recognize, inspect and classify objects or features, providing intelligence that was previously reserved for humans. As a sub-area of machine learning, deep learning is therefore probably the most important future technology within the field of artificial intelligence and is also a long-term driver of Industry 4.0.

After reporting on the successful use of deep learning algorithms in initial pilot projects, Sick has developed software applications based on deep learning for the logistics automation system business. In this application, the deep learning system recognizes whether a sorting tray in a logistics hub is actually only loaded with one object. This leads to more efficient flows of goods.

Training for the sensor
Neural networks are used to implement deep learning. In contrast to the process of classic algorithm development, which is mainly characterized by the manual development of a suitable feature representation, a neural network is trained to find the optimal features for its task and can be retrained again and again with suitable data in order to adapt to new conditions.

Sick uses an independent in-house computer and IT base both for building the training data set by capturing and evaluating thousands of images and examples and for training the neural networks. The extensive calculation of the complex operations of the deep learning solution for training is carried out on specially equipped computers with high GPU performance. The new deep learning algorithms generated from this are provided locally on the sensor and are therefore available immediately and fail-safe, for example on an intelligent camera.

In principle, the concept of sensors specialized by artificial intelligence can also be applied to simple sensors such as inductive proximity switches, retro-reflective photoelectric sensors, ultrasonic sensors and others. In addition, system solutions such as the increasingly demanding vehicle classification at toll stations offer potential for a deep learning-based classification of vehicles into toll classes. as