Currently, increasing effort is taken in the robotics community to understand injury mechanisms during a physical human-robot interaction. This is motivated by the fact that human and robot will work intensively and closely together, and therefore, one has to be aware of the potential threats in case such a close cooperation takes place.
To gain insight into injury mechanisms during robot-human collisions, impact experiments with pig tissue and crash test dummies from automobile crash testing were conducted for various contact scenarios [1, 2]. Relevant biomechanical and medical severity indices were evaluated to quantify the human injury probability.
Having obtained the relationship between collision parameters and injury, the aim of this thesis is to embed injury knowledge into control and/or motion planning, in order to minimize the human injury risk. Possible questions that can be addressed are:
How can the robot velocity be limited to a safe value in any situation?
How can a robot motion be modified such that the impacting mass is minimized?
How can collisions against sensitive human body parts be avoided?
After reviewing relevant safety and control literature, a problem definition shall be formulated and possible control schemes be found. Methods for achieving the desired safe robot motions include e.g. model predictive control (MPC) or port-Hamiltonian systems. The control scheme shall be implemented in simulation and then be applied to state of the art robots like the DLR Lightweight Robot III or the DLR Hand-Arm System.
Knowledge in robotics
Knowledge in nonlinear control
Knowledge in MATLAB/Simulink
Ability to work well structured and organized
 Haddadin, Sami, et. al, Requirements for Safe Robots: Measurements, Analysis and New Insights, The International Journal of Robotics Research (IJRR), 2009
 Haddadin, Sami, et. al, Experimental Safety Study on Soft-tissue Injury in Robotics, IEEE Robotics & Automation Magazine, 2011