Topics Projektpraktikum Cognitive Systems

In order to discuss your interest in the topics or any other questions, please get in touch with the advisors listed under each topic.

Cybathlon BCI Challenge: Control a Virtual Agent through Simulated Motor-Imagery Brain-Computer Interface

Please contact the advisor by email: Nicolas Berberich

Abstract: Brain-Computer Interfaces based on recording and classifying motor imageries from EEG signals allow to translate intentions into actions without movement. Thus, they can be used to increase agency for people who have lost the ability to move their limbs, e.g. due to quadriplegia. This is the focus of the Cybathlon BCI Challenge for which a student team at TUM is preparing. This project is intended for students who are participating in the team and would like to turn their involvement into a scientific project. Possible projects could be the implementation and comparative evaluation of real-time artifact removal algorithms, supervised motor imagery classification models or the transfer function from classifier output to control input. The project will be based on public EEG datasets and can thus be done from home.


Smart planning of pick and place tasks for WRS “Tidy up there” challenge

Advisor: Julio Rogelio Guadarrama Olvera

Abstract: The “Tidy up there” challenge of the World Robot Summit competition requires a robot to arrange the objects in a room from random places on the ground and an area to specific locations and orientations according to certain object classes. This challenge requires fast and precise pick and place manipulation and in-door navigation. The objectives are two: 1) pick and place as many objects as possible in 15 minutes, 2) maximize the score where every object sum different according to the difficulty for picking and placing. Therefore, any improvement in achieving these goals will represent a bigger chance of winning the competition. In this context, the proposed topic is to develop a high-level planner that decides the order in which the objects must be sorted in order to reduce the transit time and waiting times. Strategies, as using a tray or picking two objects at the same time, can be used by the planner to improve the results. 

Aptitudes: Good programing skills for C++ or python and good coding standards

References:  

  • Kang, M., Kwon, Y., & Yoon, S. E. (2018, June). Automated task planning using object arrangement optimization. In 2018 15th International Conference on Ubiquitous Robots (UR) (pp. 334-341). IEEE. 

  • Rahman Dabbour, A., Erdem, E., & Patoglu, V. (2019). Object Placement on Cluttered Surfaces: A Nested Local Search Approach. arXiv preprint arXiv:1906.08494

  • Kang, M., Kwon, Y., & Yoon, S. E. (2018, June). Automated task planning using object arrangement optimization. In 2018 15th International Conference on Ubiquitous Robots (UR) (pp. 334-341). IEEE. 


Reem-C robot in mc_rtc control framework

Abstract: The mc_rtc control framework is a useful software stack intended to fast implement whole-body controllers in humanoid robots and mobile platforms. It is based on QP task definition and solving to execute a stack of tasks with a strict or flexible hierarchy. The Reem-C robot is a full-size humanoid robot that runs on the ros_control framework which provides a hardware abstraction layer to implement controllers. This topic requires implementing a bridge between the ros_control framework and the mc_rtc framework for the Reem-C robot. 

Aptitudes: Strong programing skills for C++ and good coding standards

References:  


3D Force Sensing from Distributed Barometric Pressure Sensors

Advisor: Quentin Leboutet

Abstract: In robotics, the ability of simultaneously measuring three axis forces and contact angles open the doors to a whole new range of interaction and walking algorithms. In 2018, the Biomimetic Robotics Laboratory at MIT developed a sensor with such capabilities. Their design is based on an array of modified barometric pressure sensors, casted into a flexible polymer, resulting in a both cheap and reliable device. The mapping between the pressure measurements and the actual force and contact point is obtained through a dedicated neural network. Based on the MIT original design, we built our own version of this sensor and verified the feasibility of the concept. Your task this semester will be to calibrate this sensor and to implement an online force and contact point estimator using dedicated learning techniques. 

Aptitudes: Strong programing skills for C++, prior knowledge of ROS, prior knowledge of estimation techniques (least-squares, kalman filter), prior knowledge of machine learning techniques. 

References:

  • Meng Yee (Michael) Chuah, Lindsay Epstein, Donghyun Kim, Juan Romero, and Sangbae Kim, “Bi-Modal Hemispherical Sensor: A Unifying Solution for Three AxisForce and Contact Angle Measurement”, IEEE IROS (2019) 

  • Meng Yee (Michael) Chuah, “Composite Force Sensing Foot Utilizing Volumetric Displacement of a Hyperelastic Polymer”, Carnegie Mellon University (2010) 

  • Meng Yee Chuah and Sangbae Kim, “Enabling Force Sensing During Ground Locomotion: A Bio-Inspired, Multi-Axis, Composite Force Sensor Using Discrete Pressure Mapping”, IEEE Sensor (2014) 

  • Meng Yee (Michael) Chuah and Sangbae Kim, “Improved Normal and Shear Tactile Force Sensor Performance via Least Squares Artificial Neural Network (LSANN)"