Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Human-Robot Collaborative Weight Carrying with a Legged Robot using a Force Torque Sensor
Human-Robot collaborative weight carrying experiments have been carried out mainly with wheeled robots. We want to exploit the advantages of using a legged robot for this task. However, this adds additional complexity in robustly using the readings from the force torque sensor at the end-effector.
Keywords: human robot collaboration, legged robotics, force torque sensor
During human-robot collaboration it is important to have accurate force torque feedback on the robot side in order to detect the human behaviour and react safely to it. This is especially true in a weight carrying scenario with a human, as the readings from the force-torque sensor at the end-effector, deliver the robot with important information, such as in which direction to move, how fast to move, when to stop and so on.
Having a legged platform, additional noise is added to the sensor readings, especially when making contact with the ground while trotting, This makes the values unreliable or even unusable in some scenarios. In this project the student should develop a filtering algorithm to overcome the mentioned problem. At the end of the project the filtering algorithm will be validated on the real robot by carrying out cooperative load carrying experiments.
During human-robot collaboration it is important to have accurate force torque feedback on the robot side in order to detect the human behaviour and react safely to it. This is especially true in a weight carrying scenario with a human, as the readings from the force-torque sensor at the end-effector, deliver the robot with important information, such as in which direction to move, how fast to move, when to stop and so on.
Having a legged platform, additional noise is added to the sensor readings, especially when making contact with the ground while trotting, This makes the values unreliable or even unusable in some scenarios. In this project the student should develop a filtering algorithm to overcome the mentioned problem. At the end of the project the filtering algorithm will be validated on the real robot by carrying out cooperative load carrying experiments.
- problem analysis
- literature research and method selection
- developing the filtering algorithm based on the selected method
- testing the setup on real hardware in a collaborative load carrying scenario with ANYmal with arm
- problem analysis - literature research and method selection - developing the filtering algorithm based on the selected method - testing the setup on real hardware in a collaborative load carrying scenario with ANYmal with arm
- Excellent knowledge of C++ and some experience with hardware
- Experience in ROS or sensing related topics is a plus
- Excellent knowledge of C++ and some experience with hardware - Experience in ROS or sensing related topics is a plus