(keine Einträge)
Karriere und Jobs
Für Hochschulabsolventen und Berufstätige
Projekte und Jobs für Studierende
Robot Motion Planning
We are seeking a motivated student to join the Environmental Robotics Lab at MIRMI and contribute to cutting-edge projects in underwater manipulation. Research in underwater manipulation focuses on developing robotic systems capable of performing complex tasks in underwater environments. This involves creating autonomous underwater vehicles (AUVs) equipped with manipulators that can handle objects, conduct repairs, and collect samples. Key challenges include dealing with the dynamic nature of underwater environments, such as currents and low visibility, and ensuring precise control and stability of the manipulators]. Researchers are also working on advanced perception systems and machine learning algorithms to improve the autonomy and efficiency of these robots. The ultimate goal is to enable these systems to operate with minimal human intervention, enhancing their utility in areas like marine biology, underwater archaeology, and environmental monitoring.
Tasks:
- Grasping of known and unknown objects underwater
- Pick-and-place task
- Concept for underactuated underwater manipulation
- Implementation for Reach Alpha 5 manipulator in ROS
- Realistic evaluation in underwater robotics pool
Requirements:
- Knowledge in robotics and control theory
- Coding skills in C++ and/or Python
- Experience with ROS
- Hands-on experience with robots is a plus
What We Offer:
- Opportunity to work on innovative projects with real-world applications
- Access to state-of-the-art facilities and resources
- Collaborative and supportive research environment
- Flexible working hours to accommodate your academic schedule
How to Apply: Please send your CV, a brief cover letter including your prior experience and what you expect to learn, as well as any supporting documents to Moritz Graf (moritz.graf@tum.de). Applications will be reviewed on a rolling basis until the position is filled. The Technical University of Munich is an equal opportunity employer committed to excellence through diversity. We explicitly encourage women to apply, and preference will be given to disabled applicants with equivalent qualifications.
We are seeking a highly motivated Masters student to join the Service Robotics Team at MIRMI and perform research on cloth manipulation. The research will be jointly supervised by collaborators and will focus on developing novel solutions to address the open problem dynamic real-time cloth folding. Most efforts along the same lines use domain randomization in simulations for sim2real transfer, often overlooking domain adaptation techniques that allow dynamic adjustment based on real-time feedback due to perceptual challenges.Research: In this investigation, we will develop a framework that respects the sim2real transfer using a high level trajectory optimization combined with a low level planning policy. More in detail, the first task is to set-up differentiable simulation environment with a cloth. Then, gather data to set up a learning pipeline for behavior cloning with a real-time visual input. Finally, we carry out experiments on a real robotic system.
Research:
In this investigation, we will develop a framework that respects the sim2real transfer using a high level trajectory optimization combined with a low level planning policy. More in detail, the first task is to set-up differentiable simulation environment with a cloth. Then, gather data to set up a learning pipeline for behavior cloning with a real-time visual input. Finally, we carry out experiments on a real robotic system.
Requirements:
- Experiences with physical simulators (e.g., PyBullet, DAXBench, MuJoCo)
- Strong background, expertise or high interest in machine learning tools.
- Knowledge of robot kinematics/dynamics
- Experience with Git
Contact:
Dr. Tianyu Ren (tianyu.ren@tum.de)
Riddhiman Laha (riddhiman.laha@tum.de)
Hamid Sadeghian (hamid.sadeghian@tum.de)
Georg-Brauchle-Ring 60-62, 80992 München
Refenrences:
https://arxiv.org/pdf/2407.01361
To apply:
Send your personal info as attachment to the contact person with the following naming
FirstnameLastname_ Forschungspraxis/Thesis_Staringdate.pdf
e.g., TianyuRen_Thesis_26062024.pdf
The goal of this 3 month internship (with the possibility to extend to 6 months) will be to propose and benchmark real-time robot planning algorithms. To that end, we are seeking a highly motivated HiWi to join the Service Robotics Team at MIRMI and perform high-impact motion planning research combining model-based and machine learning approaches. The research will focus on developing novel solutions to address the open problem of dynamic real-time whole body motion planning for complex robotic systems. Current approaches usually based on sampling-based methods and/or vector fields are either non-reactive or introduce artifacts like local minima with competing multi-objectives in cluttered scenarios. Prior work has already proposed solutions for reactive planning that scale up to complex bimanual manipulation planning.
Research: In this investigation, we will propose a multi-agent environment exploration strategy combined with a data-driven module that plans (sub)optimal trajectories using various maneuvering strategies. This intelligent look-ahead would facilitate adaptive performance at different levels. Furthermore, large scale simulation studies would corroborate the fact the developed framework is system agnostic.
Related Literature: Predictive Multi-Agent-Based Planning and Landing Controller for Reactive Dual-Arm Manipulation (IEEE Transactions on Robotics, 2023)
The main tasks would be (in close collaboration) :
- Implement and extend existing motion planner;
- Port existing planning algorithm in a simulation environment like Issac Sim, Mujoco;
- Set-up comparison planners like STOMP, CHOMP, and TrajOpt;
- Come up with easy and hard scenarios to test the planners;
- Integrate tested code to the humanoid planning stack and perform experiments;
Pre-requisites:
- Master's degree (currently enrolled) in robotics, artificial intelligence, mechanical engineering, computer science, or a related field;
- Excellent programming skills (e.g., C++, Python, Matlab) and good experience with robotic simulation environments like Issac Sim, Mujoco;
- Familiarity with ROS, MoveIt, OMPL;
- Basics of Manipulator Kinematics and Dynamics;
- Basic working knowledge of Manipulator Control;
Helpful but not required
Knowledge of search-based, optimization-based, and reactive planning, running simple controllers on robots.
For more information, please contact:
Riddhiman Laha (riddhiman.laha@tum.de)
Dr. Tianyü Ren (tianyu.ren@tum.de)
Mensch-Roboter-Interaktion
Robot Control
Electronics
Pos-type: Forschungspraxis/Internship, possible thesis extention. New application deadline: November 15, 2024
Brushless motors are growing in popularity for Robotics applications. In particular, due to their high power density, these motors can be used with smaller gear ratios to deliver the torque and speed requirements. For example, the key to MIT mini cheetah's success was BLDC adaptation within the Proprioceptive Actuator concept [1]. You haven't seen its performance, check it here.
Interesting part of using BLDC actuators is the versatility of its electronics, since it requires a logic unit (usually a microcontroller) for its control. It requires 3 phases to be coordinated in a specific manner to supply needed power. This concept is not new, one of the most popular mainstream approaches is FOC control for BLDC. Underlying system design is quite important, which in this case consists of the custom PCB controller and the motor itself. At the end of the internship, one will gain knowledge and understanding of the power PCB design and improve current design to better handle more extreme working conditions: high power loads, high speeds, etc.
Having in mind the challenge, the person should be familiar with the PCB design. Willing to do the signal hacking, debugging, working with the oscilloscope, understand logic of different components, etc. Brushless motors are commonly used in the hobby project, thus experiance can provide great foundation for "build your own things" mindset. At least we do expect the hands-on and genuine curiosity in the engineering & design aspects from the students joining us.
What you will gain:
- Hands-on experience and in-depth understanding of Brushless Motors, and their control
- Best practices for PCB Design and Project Collaboration
- Experience building, prototyping
- Hacking electronic signals (via oscilloscope, etc.)
- Insights in our System Development and access to our community
Requirements from candidates:
- Knowledge of Electronics/Power electronics
- Experience with PCB design and soldering
- Knowledge of one of the popular PCB Design software
- Basic Matlab skills
- Plus are:
- Understanding how Motors work
- Familiarity with GIT
- Embedded system programming, knowledge of C language
- Working skills in Ubuntu operating system
To apply, you can send your CV, and short motivation to:
Supervisor
M.Sc. Vasilije Rakcevic
[1] P. M. Wensing, A. Wang, S. Seok, D. Otten, J. Lang and S. Kim, "Proprioceptive Actuator Design in the MIT Cheetah: Impact Mitigation and High-Bandwidth Physical Interaction for Dynamic Legged Robots," in IEEE Transactions on Robotics, vol. 33, no. 3, pp. 509-522, June 2017, doi: 10.1109/TRO.2016.2640183.
Mechatronik
Brain Computer Interfaces
Dexterous manipulation
Robotic hand modeling
Studentische Hilfskräfte (HiWi)
Andere Kategorien
Weitere Forschungsmöglichkeiten finden Sie an den Lehrstühlen unserer Principal Investigators!
Hinweis zum Datenschutz:
Im Rahmen Ihrer Bewerbung um eine Stelle an der Technischen Universität München (TUM) übermitteln Sie personenbezogene Daten. Beachten Sie bitte hierzu unsere Datenschutzhinweise gemäß Art. 13 Datenschutz-Grundverordnung (DSGVO) zur Erhebung und Verarbeitung von personenbezogenen Daten im Rahmen Ihrer Bewerbung. Durch die Übermittlung Ihrer Bewerbung bestätigen Sie, dass Sie die Datenschutzhinweise der TUM zur Kenntnis genommen haben.