Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
Next revisionBoth sides next revision
jobs [2015/03/19 12:56] – [Theses and Jobs] balintbejobs [2022/12/22 12:04] – [Theses and Student Jobs] kording
Line 1: Line 1:
-~~NOTOC~~ 
-=====Theses and Jobs===== 
-If you are looking for a bachelor/master thesis or a job as a student research assistant, you may find some interesting opportunities on this page. 
  
 +=====Open researcher positions=====
  
 +We currently don't have any open positions for researchers.
 +=====Theses and Student Jobs=====
 +If you are looking for a bachelor/master thesis or a job as a student research assistant, you may find some interesting opportunities on this page.
  
-== GPU-based Parallelization of Numerical Optimization Techniques (BA/MA/HiWi)== 
  
-In the field of Machine Learning, numerical optimization techniques play a focal role. However, as models grow larger, traditional implementations on single-core CPUs suffer from sequential execution causing a severe slow-down. In this thesis, state-of-the-art GPU frameworks (e.g. CUDA) are to be investigated in order implement numerical optimizers that substantially profit from parallel execution.+<html><!-- 
 +== Physics-based grasping in VR with finger tracking(Student Job / HiWi) ==
  
-Requirements: +Implementing physics-based grasping models in virtual environments, 
-  * Skills in numerical optimization algorithms +using Manus VR.
-  * Good programming skills in Python and C/C++ +
- +
-Contact: [[team:daniel_nyga|Daniel Nyga]] +
- +
-== Online Learning of Markov Logic Networks for Natural-Language Understanding (MA)== +
- +
-Markov Logic Networks (MLNs) combine the expressive power of first-order logic and probabilistic graphical models. In the past, they have been successfully applied to the problem of semantically interpreting and completing natural-language instructions from the web. State-of-the-art learning techniques mostly operate in batch mode, i.e. all training instances need to be known in the beginning of the learning process. In context of this thesis, online learning methods for MLNs are to be investigated, which allow incremental learning, when new examples come in one-by-one.+
  
 Requirements: Requirements:
-  * Experience in Machine Learning. +  * Good C++ programming skills 
-  * Experience with statistical relational learning (e.g. MLNs) is helpful. +  * Familiar with skeletal animations 
-  * Good programming skills in Python.+  * Experience with simulators/physics-/game- engines 
 +  * Familiar with Unreal Engine API 
 +  * Familiar with version-control systems (git) 
 +  * Able to work independently with minimal supervision
  
-Contact: [[team:daniel_nyga|Daniel Nyga]]+Contact: [[team:andrei_haidu|Andrei Haidu]] 
 +--></html>
  
  
-==HiWi-Position: Knowledge Representation & Language Understanding for Intelligent Robots==+<html><!-- 
 +== Lisp / CRAM support assistant (HiWi) ==
  
-In the context of the European research project RoboHow.Cog [1,2] we +Technical support for the group for Lisp and the CRAM framework. \\ 
-are investigating methods for combining multimodal sources of knowledge (e.g. video, natural-language recipes or computer games), in order to enable mobile robots to autonomously acquire new high level skills like cooking meals or straightening up rooms.   +8+ hours per week for up to 1 year (paid).
- +
-The Institute for Artificial Intelligence is hiring a student researcher for the +
-development and the integration of probabilistic methods in AI, which enable intelligent robots to understand, interpret and execute natural-language instructions from recipes from the World Wide Web. +
- +
-This HiWi-Position can serve as a starting point for future Bachelor's, Master's or Diploma Theses. +
- +
-Tasks: +
-  * Implementation of an interface to the Robot Operating System (ROS). +
-  * Linkage of the knowledge base to the executive of the robot. +
-  * Support for the scientific staff in extending and integrating components onto the robot platform PR2.+
  
 Requirements: Requirements:
-  * Studies in Computer Science (Bachelor's, Master's or Diploma) +  * Good programming skills in Common Lisp 
-  * Basic skills in Artificial Intelligence +  * Basic ROS knowledge
-  * Optional: basic skills in Probability Theory +
-  * Optional: basic skills in Machine Learning +
-  * Good programming skills in Python and Java +
- +
-Hours: 10-20 h/week +
- +
-Contact: [[team:daniel_nyga|Daniel Nyga]]+
  
-[1] www.robohow.eu\\ +The student will be introduced to the CRAM framework at the beginning of the job, which is a robot programming framework written in Lisp. The student will then be responsible for assisting not familiar with the framework people, explaining them the parts they don't understand and pointing them to the relevant documentation sources.
-[2] http://www.youtube.com/watch?v=0eIryyzlRwA+
  
 +Contact: [[team:gayane_kazhoyan|Gayane Kazhoyan]]
 +--></html>
  
-== Kitchen Activity Games in a Realistic Robotic Simulator (BA/MA/HiWi)== +<html><!-- 
- {{ :research:gz_env1.png?200|}} +== Mesh Editing / Mesh Segmentation/Cutting (Student Job / HiWi)== 
 + {{ :research:human_hand_cutting.png?150|}}
  
-Developing new activities and improving the current simulation framework done under the [[http://gazebosim.org/|Gazebo]] robotic simulator. Creating a custom GUI for the game, in order to launch new scenarios, save logs etc.+ Editing and cutting a human mesh into different parts in Blender / Maya (or other).
  
 Requirements: Requirements:
-  * Good programming skills in C/C++ +  * Good knowledge in 3D Modeling 
-  * Basic physics/rendering engine knowledge +  * Familiar with Blender / Maya (or other)
-  * Gazebo simulator basic tutorials+
  
-Contact: [[team:andrei_haidu|Andrei Haidu]]+Contact: [[team/mona_abdel-keream|Mona Abdel-Keream]] 
 +--></html>
  
-== Integrating Eye Tracking in the Kitchen Activity Games (BA/MA)== 
- {{ :research:eye_tracker.png?200|}}  
- 
-Integrating the eye tracker in the [[http://gazebosim.org/|Gazebo]] based Kitchen Activity Games framework and logging the gaze of the user during the gameplay. From the information typical activities should be inferred. 
- 
-Requirements: 
-  * Good programming skills in C/C++ 
-  * Gazebo simulator basic tutorials 
- 
-Contact: [[team:andrei_haidu|Andrei Haidu]] 
  
-== Hand Skeleton Tracking Using Two Leap Motion Devices (BA/MA)== +<html><!-- 
- {{ :research:leap_motion.jpg?200|}} +== 3D Animation and Modeling (Student Job / HiWi)== 
 + {{ :research:kitchen_unreal.jpg?200|}} 
  
-Improving the skeletal tracking offered by the [[https://developer.leapmotion.com/|Leap Motion SDK]], by using two devices (one tracking vertically the other horizontally) and switching between them to the one that has the best current view of the hand.+Developing and improving existing or new 3D (static/skeletal) 
 +models in Blender / Maya (or other). Further importing and testing the 
 +models against Unreal Engine.
  
-The tracked hand can then be used as input for the Kitchen Activity Games framework.+Bonus: Working with state of the art 3D Scanners [[https://www.goscan3d.com/|Go!SCAN]] for creating sketching new models.
  
 Requirements: Requirements:
-  * Good programming skills in C/C+++  * Experience with Blender / Maya (or other) 
 +  * Knowledge of Unreal Engine material / lightning development 
 +  * Familiar with version-control systems (git) 
 +  * Able to work independently with minimal supervision
  
 Contact: [[team:andrei_haidu|Andrei Haidu]] Contact: [[team:andrei_haidu|Andrei Haidu]]
 +--></html>
  
-== Fluid Simulation in Gazebo (BA/MA)== +== Representing knowledge in a robot-agnostic ontology system for assistive robotics (MA Thesis) ==
- {{ :research:fluid.png?200|}} +
  
-[[http://gazebosim.org/|Gazebo]] currently only supports rigid body physics engines (ODE, Bullet etc.), however in some cases fluids are preferred in order to simulate as realistically as possible the given environment.+The thesis will be jointly supervised by German Aerospace Center (DLR) in Munich and the Institute for Artificial Intelligence (IAI) at the University of Bremen.
  
-Currently there is an [[http://gazebosim.org/tutorials?tut=fluids&cat=physics|experimental version]] of fluids  in Gazebo, using the [[http://onezero.ca/fluidix/|Fluidix]] library to run the fluids computation on the GPU.+Summary: 
 +  * Investigate about existing ontologies, like the Socio-physical Model of Activities (SOMA) from the University of Bremen. 
 +  * Refactor an existing knowledge database (known as the Object DataBase, ODB) used by the DLR into an ontology. 
 +  * Use the ontology to design tasks in the assistive robotics domain using assistive robots provided by the DLR
  
-The computational method for the fluid simulation is SPH (Smoothed-particle Dynamics), however newer and better methods based on SPH are currently present +Full offer:
-and should be implemented (PCISPH/IISPH).+
  
-The interaction between the fluid and the rigid objects is a naive one, the forces and torques are applied only from the particle collisions (not taking into account pressure and other forces).+{{ :teaching:offer_2020-01_op2_wedanjustin.pdf |}}
  
-Another topic would be the visualization of the fluid, currently is done by rendering every particle. For the rendering engine [[http://www.ogre3d.org/|OGRE]] is used.+Contact: [[team:daniel_bessler|Daniel Beßler]]
  
-Here is a [[https://vimeo.com/104629835|video]] example of the current state of the fluid in Gazebo. +== Chatbot zur Erfassung von Vorwissen in Informatik (BA Thesis) ==
  
-Requirements: +In dieser Arbeit soll ein Chatbot in einem jupyter notebook erstellt werden, der das Vorwissen von Nutzer*Innen in verschiedenen Bereichen der Informatik abfragt und bewertet.
-  * Good programming skills in C/C++ +
-  * Interest in Fluid simulation +
-  * Basic physics/rendering engine knowledge +
-  * Gazebo simulator and Fluidix basic tutorials+
  
-Contact: [[team:andrei_haidu|Andrei Haidu]]+Aufgaben: 
 +  * Arbeit mit jupyter notebook 
 +  * Recherche zu existierenden chatbots 
 +  * Erstellung eines chatbots 
 +  * Durchführung einer (kleinen) Nutzerstudie zur Evaluation des erstellten chatbots 
 +  * Bewertung der Nutzerangaben (zur weiteren Verwertung)
  
 +Contact: [[team:michaela_kümpel|Michaela Kümpel]]
  
-== Automated sensor calibration toolkit (MA)==+== Integration of novel objects into Digital Twin Knowledge Bases (MA Thesis) ==
  
-Computer vision is an important part of autonomous robots. For robots the image sensors are the main source of information of the surrounding world. Each camera is different, even if they are from the same production line. For computer vision, especially for robots manipulating their environment, it is important that the parameters for the cameras in use are well known. The calibration of a camera is a time consuming task, and the result depends highly on the chosen setup and the accuracy of the operator.+In this thesis, the goal is to make a robotic system learn new objects automatically. 
 +The system should be able to generate the necessary models required for re-detecting it again and also consult online information sources to automatically acquire knowledge about it.
  
-The topic for this master thesis is to develop an automated system for calibrating cameras, especially RGB-D cameras like the Kinect v2.+The focus of the thesis would be two-fold: 
 +  * Develop methods to automatically infer the object class of new objects. This would include perceiving it with state of the art sensors, constructing a 3d model of it and then infer the object class from online information sources. 
 +  * In the second step the system should also infer factual knowledge about the object from the internet and assert it into a robotic knowledgebase. Such knowledge could for example include the category of this product, typical object properties like its weight or typical location and much more.
  
-The system should be: 
-  * independent of the camera type 
-  * estimate intrinsics and extrinsics 
-  * have depth calibration (case of RGBD) 
-  * integrate capabilities from Halcon [1] 
  
-Requirements:  +Requirements: 
-  * Good programming skills in Python and C/C++ +  * Knowledge about sensor data processing 
-  * ROS, OpenCV+  * Interest in model construction from sensory data 
 +  * Work with KnowRob knowledge processing framework
  
-[1] http://www.halcon.de/ 
  
-Contact: [[team:thiemo_wiedemeyer|Thiemo Wiedemeyer]]+Contact: [[team:patrick_mania|Patrick Mania]]
  
-== On-the-fly 3D CAD model creation (MA)==+<html><!-- 
 +== Development of Modules for Robot Perception (Student Job / HiWi) == 
 +In our research group, we focus on the development of modern robots that can make use of the potential of game engines. One particular research direction, is the combination of computer vision with game engines. 
 +In this context, we are currently offering multiple Hiwi positions / student jobs for the following tasks: 
 +  * Software development to create Interfaces between ROS and Unreal Engine 4 (mainly C++) 
 +  * Software development for our Robot Perception framework [[http://robosherlock.org/|RoboSherlock]] which contains many algorithms for computer vision problems (mainly C++)
  
-Create models during runtime for unknown textured objets based on depth and color information. Track the object and update the model with more detailed information, completing it's 3D model from multiple views improving redetection. Using the robots manipulator pick up the object and complete the model by viewing it from multiple viewpoints.+Requirements: 
 +  * Experience in C++. 
 +  * Basic understanding of the ROS middleware and Linux. 
 +The spoken language in this job is german or english, based on your preference.
  
-Requirements:  +Contact: [[team:patrick_mania|Patrick Mania]] 
-  * Good programming skills in C/C++ +--></html>
-  * strong background in computer vision  +
-  * ROS, OpenCV, PCL+
  
-Contact: [[team:thiemo_wiedemeyer|Thiemo Wiedemeyer]]+== Game Engine Developer and 3D-Modelling  (Student Job / HiWi) == 
 +A recent development in the field of AI is the usage of photorealistic simulations, for example to create Digital Twins of environments like households or  [[https://iottechnews.com/news/2021/apr/13/bentley-systems-nvidia-omniverse-create-real-time-digital-twins/|factories]]. 
 +In our research group, we focus on the development of modern robots that can make use of the potential of game engines. This requires a high degree of specialized game engine plugins that can simulate certain aspects of our research. Another important task is the creation of 3d models.
  
-== Simulation of a robots belief state to support perception(MA) ==+Therefore, we are currently offering multiple Hiwi positions / student jobs for the following tasks: 
 +  * Modelling of objects for the use in Unreal Engine 4. 
 +  * Creation of specific simulation aspects in Unreal Engine 4. For example the development of interactable objects.
  
-Create a simulation environment that represents the robots current belief state and can be updated frequently. Use off-screen rendering to investigate the affordances these objects possess, in order to support segmentation, detection and tracking of these in the real world.  +Requirements: 
- +  * Knowledge of 3D-Modelling tools. Blender would be highly preferred. 
-Requirements:  +  * Experience in Game Engine development. Ideally Unreal Engine 4 and C++.
-  * Good programming skills in C/C++ +
-  * strong background in computer vision  +
-  * Gazebo, OpenCV, PCL +
- +
-Contact: [[team:ferenc_balint-benczedi|Ferenc Balint-Benczedi]] +
- +
-== Multi-expert segmentation of cluttered and occluded scenes == +
- +
-Objects in a human environment are usually found in challenging scenes. They can be stacked upon eachother, touching or occluding, can be found in drawers, cupboards, refrigerators and so on. A personal robot assistant in order to execute a task, needs to detect these objects and recognize them. In this thesis a multi-modal approach to interpreting cluttered scenes is going to be investigated, combining the results of multiple segmentation algorithms in order to come up with more reliable object hypotheses. +
- +
-Requirements:  +
-  * Good programming skills in C/C++ +
-  * strong background in 3D vision  +
-  * basic knowledge of ROS, OpenCV, PCL+
  
-Contact: [[team:ferenc_balint-benczedi|Ferenc Balint-Benczedi]]+The spoken language in this job is german or english, based on your preference.
  
 +Contact: [[team:patrick_mania|Patrick Mania]]
- Impressum - Datenschutz



Prof. Dr. hc. Michael Beetz PhD
Head of Institute

Contact via
Andrea Cowley
assistant to Prof. Beetz
ai-office@cs.uni-bremen.de

Discover our VRB for innovative and interactive research


Memberships and associations:


Social Media: