Aerial Robots, unlike regular drones, are able to perform physical work at height and are becoming true general purpose robots.
High performance and high accuracy control is still an ongoing challenge - after all, precise physical work in constantly changing airflow is challenging!
In this project, we want to explore novel Reinforcement Learning approaches for end-to-end control of a complex aerial robot with 12 degrees of freedom. During the project, you will get hands on experience in aerial robotics, sim-to-real transfer for flying vehicles in turbulence and state of the art reinforcement learning. Your learned controllers will be evaluated on the most advanced over-actuated aerial robot that is currently being researched on.
Aerial Robots, unlike regular drones, are able to perform physical work at height and are becoming true general purpose robots. High performance and high accuracy control is still an ongoing challenge - after all, precise physical work in constantly changing airflow is challenging!
In this project, we want to explore novel Reinforcement Learning approaches for end-to-end control of a complex aerial robot with 12 degrees of freedom. During the project, you will get hands on experience in aerial robotics, sim-to-real transfer for flying vehicles in turbulence and state of the art reinforcement learning. Your learned controllers will be evaluated on the most advanced over-actuated aerial robot that is currently being researched on.
- Literature research
- Choosing an appropriate approach
- Implementation and training
- Testing in simulation
- Testing on Aerial Robot
- Literature research - Choosing an appropriate approach - Implementation and training - Testing in simulation - Testing on Aerial Robot
- Good knowledge of C++ and Python
- Experience with ROS
- Experience with RL in general
- General knowledge of control theory
- Desirable: Knowledge of NVidia Isaac Sim
- Good knowledge of C++ and Python - Experience with ROS - Experience with RL in general - General knowledge of control theory - Desirable: Knowledge of NVidia Isaac Sim
Please contact via Sirop directly, including a recent transcript of records and CV. We would also appreciate a very short statement about your motivation/why you are interested in the topic.
Please contact via Sirop directly, including a recent transcript of records and CV. We would also appreciate a very short statement about your motivation/why you are interested in the topic.
