Data Analytics

As the number of sensors increases, it becomes increasingly difficult to make the right conclusions from the flood of available information. Methods of data analysis, such as Machine learning can help identify essential patterns in the data and ultimately use them to provide better products and better sensors that ultimately optimize business processes.

Our basic motivation is to design and harness robust tracking algorithms and data analysis techniques using both machine learning and statistical methods. The focus is also on hybrid methods that take advantage of both aspects.

 

Sports Analytics

Sport is becoming increasingly mediatized. Nowadays not only the final game score is important, but also the player´s performance. Sport-relevant evaluation parameters can be precisely calculated and used for improving the movement prediction...

 

Reinforcement Learning

We use automation everywhere in our daily lives. So it's even more important that everything works fine. Meanwhile, potential errors can be corrected by the machines themselves...

 

Optical Tracking

The most obvious method of tracking is through optical stimuli, since the largest amount of environmental information is received through the eyes. This can also be transferred to state-of-the-art technologies...

 

Smart Sensing

Finally, in the process of data analytics and machine learning, movements or states are always classified. Learn more about how researchers and developers at Fraunhofer IIS have once again been particularly involved in this...

 

 

Smart Objects

Sensors are an important component of the development towards automation and networking. They are usually meant to measure a specific parameter, but that's often not enough...

 

Machine Learning Forum

Education opportunities for students and developers in industry.

 

embeddif.ai

embeddif.[ai] is changing the way we think about AI. With the help of embeddif.ai, complex machine learning applications can be run locally on embedded systems - without the need for a cloud.

Competences

 

Machine Learning

In machine learning, the system can learn from examples and generalize and implement them after completing the learning phase...

 

Fusion

A single technology can not usually meet all requirements. One solution is the fusion of multiple technologies and information...

 

Positioning

The localization competence of Fraunhofer IIS can also be used in the field of data analytics. Exact position determination can also be elementary here...

 

Quantum Machine Learning

The development of a quantum computer is no longer a fiction. With entangled qubits, complex computation paths can be solved at the same time.

 

TinyML

Use cases and our offering for optimizing and executing AI-based processing chains on embedded systems.

 

Deep Compression

Artificial intelligence is mostly implemented with Deep Learning, which enables more efficient transfer and execution. The savings in time and energy are immediate.

 

Applications and Projects

Learn more about practice-based projects and the possibilities within your field of application.

 

Publications

Here you can find our publications on the topic of data analytics and machine learning.