Engineer (m/f/d) in real-time stereo vision

The Institute of Robotics and Mechatronics is developing a wide range of robots that enable people to interact more safely and efficiently with their environment. The robots are designed to operate in environments that are inaccessible or dangerous for humans and to support people in their everyday lives and at work.

What you can expect

The Perception and Cognition department is concerned with enabling robotic systems to perceive their environment holistically using the two dominant robot senses of touch and vision. Perception and cognition are elementary components for teleoperated assistance and autonomous robot systems, as they make reactive behaviour, attention control and situation interpretation possible in the first place. Stereo vision is the cornerstone for the exteroceptive perception of our robots. In addition to generating colour images, stereo vision enables the passive assessment of depth in a robot\’s environment. This capability enables our robots to safely navigate and interact with their environment – both indoors and outdoors – by providing reliable spatial awareness without active sensors.

Your tasks

• Research work to improve stereo vision algorithms by combining classical and data-driven approaches for depth estimation
• Implementation and development of algorithms for different platforms (CPU, GPU, NVIDIA Jetson, FPGA), depending on the specific constraints and requirements of each application and method
• Data generation and simulation for training data-driven methods. Creation and curation of photorealistic, high-quality data sets with our simulators (OAISYS, BlenderProc2) for specialised robotic environments ranging from planetary exploration to minimally invasive surgery
• Practical integration, testing and maintenance of robotic systems in the laboratory and at external demonstrations and trade fairs
• Literature research and publication of scientific papers
• Participation in international conferences

Your profile

• Completed scientific university degree (Master\’s degree / university diploma) in engineering or other degree programmes relevant to the position
• Good programming skills in C++ and Python and experience with Linux environments
• Familiarity with software frameworks for machine learning (PyTorch,TensorFlow, CUDA) and robotics middleware (ROS)
• Initial research experience in robotics, stereo depth estimation and machine learning
• Experience with 3D computer graphics (e.g. Blender)
• Team orientation and the ability to pursue constructive common goals
• Good English communication skills

Remuneration is up to the German TVöD 13 depending on qualifications and assigned tasks.

We look forward to getting to know you!

If you have any questions about this position (Vacancy-ID 2497) please contact:

Rudolph Triebel
Tel.: +49 8153 28 4289