Register now After registration you will be able to apply for this opportunity online.
LIDAR Odometry and State Estimation for Space Robotics applications
Together with ESA and Beyond Gravity, we're developing a system for testing microgravity space structures on earth. To do so, we're developing reactive ground robots that are able to support e.g. solar panels while their unfolding from a satellite is tested. Your part of the project is to develop, evaluate and test (on the robot) state estimation solutions based on LIDAR and IMU.
Keywords: State Estimation, Ground Robots, IMU, LIDAR, Odometry, Mapping, SLAM, Kalman filter, recursive estimation, space
This project is based on existing robots - an omnidirectional ground robot with an IMU, a small 360 degree lidar scanner and special sensing bearings for supporting moving space structures under test.
Your semester project will contribute to the development, improvement and evaluation of a sensor fusion framework for the robot. This includes local odometry (e.g. smooth, reliable velocity measurements based on LIDAR, IMU and wheel odometry), as well as global localization using LIDAR data.
Depending on progress and time, there are a few extensions possible (e.g. outlier detection in LIDAR data, exchange of maps for localization between robots, ...).
By the end of the project, you'll have gained very valuable experience with state-of-the-art state estimation frameworks and hands-on testing on real robots.
This project is based on existing robots - an omnidirectional ground robot with an IMU, a small 360 degree lidar scanner and special sensing bearings for supporting moving space structures under test.
Your semester project will contribute to the development, improvement and evaluation of a sensor fusion framework for the robot. This includes local odometry (e.g. smooth, reliable velocity measurements based on LIDAR, IMU and wheel odometry), as well as global localization using LIDAR data.
Depending on progress and time, there are a few extensions possible (e.g. outlier detection in LIDAR data, exchange of maps for localization between robots, ...).
By the end of the project, you'll have gained very valuable experience with state-of-the-art state estimation frameworks and hands-on testing on real robots.
- Literature study
- Familiarization with system and requirements
- Development and integration of local odometry
- Dev and integration of global localization
- Evaluation
- Literature study - Familiarization with system and requirements - Development and integration of local odometry - Dev and integration of global localization - Evaluation
- Motivated and independent student wanting to work on a real-world robotic project
- Good C++ skills with experience
- Experience with Robot Operating System (ROS
- Basic knowledge of recursive estimation/Kalman filters
- Motivated and independent student wanting to work on a real-world robotic project - Good C++ skills with experience - Experience with Robot Operating System (ROS - Basic knowledge of recursive estimation/Kalman filters
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.
Michael Pantic [mpantic(at)ethz.ch]
Lionell Ott [lioott(at)ethz.ch]
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.
Michael Pantic [mpantic(at)ethz.ch] Lionell Ott [lioott(at)ethz.ch]
