Active safety in collaborative robots
Student thesis: Master Thesis and HD Thesis
- Simon Bjerre Krogh
4. term, Control and Automation, Master (Master Programme)
This report covers the derivation of a
collision detection scheme based on a
dynamic model of a robotic structure with
three degrees of freedom. First, a PID
controller is derived with the purpose of
controlling the system, then an extended
Kalman filter with the purpose of collision
detection. The extended Kalman filter did
not work and was replaced with a linear
stationary Kalman filter which is capable
of estimating the system states. The
collision detection is based on external
force estimator, which is made by an open
loop approach.
The open loop approach was deemed
insufficient for collision detection, as it
showed a high correlation between force
estimations and inputs.
Due to this
correlation, it was not possible to detect
if a collision had occurred.
In the end a discussion reflecting the
problems of the project and how further
research may solve these are made
collision detection scheme based on a
dynamic model of a robotic structure with
three degrees of freedom. First, a PID
controller is derived with the purpose of
controlling the system, then an extended
Kalman filter with the purpose of collision
detection. The extended Kalman filter did
not work and was replaced with a linear
stationary Kalman filter which is capable
of estimating the system states. The
collision detection is based on external
force estimator, which is made by an open
loop approach.
The open loop approach was deemed
insufficient for collision detection, as it
showed a high correlation between force
estimations and inputs.
Due to this
correlation, it was not possible to detect
if a collision had occurred.
In the end a discussion reflecting the
problems of the project and how further
research may solve these are made
Language | English |
---|---|
Publication date | 7 Jun 2018 |
Number of pages | 97 |