Improving landfill monitoring programs with the aid of geoelectrical - imaging techniques and geographical information systems Master’s Thesis in the Master Degree Programme, Civil Engineering

Recent industrial development of dual-arm robots has considerably leveraged interest of researchers and robot manufacturer. The big difference for this concept compared with normal single arm robots, is that these robots can save space and are able to do multitasking. The way human plan and control bimanual operation was investigated and it was learned that there was much that could be learned from human motion planning. A dual-arm robot from pi4 robotics which is the same size as a human has been used. One bimanual motion operation, where two arms are holding a beam has been analyzed in this thesis. In order to not damage the robot, a simulation model of the robot was made in Simulink. It is desired that no internal forces or moments is applied to the beam and that such forces would be compensated when the beam is gripped. To ensure a safe system, the simulation functions has been programmed in C++ so that it is possible to try these in the simulation environment before they are used by the programming environment of the robot. In this thesis several difference dual-arm control methods for controlling two arms simultaneously have been investigated. The benefit and disadvantages between the difference methods has been pointed out and finally an impedance controller with position mode was selected to control the motion of the arms. The main advantage with this method is that it can control the behavior of the beam and does not need to switch between different kind of controllers. The impedance control algorithm was implemented and motions were the grippers compressed a beam was simulated. During the simulations special attention has been put on reducing the internal forces of the external object while solving the desired motion task. An equation to calculate the internal forces of the beam was derived and the estimated force was given to an impedance controller. The results show a promising response of the controller where internal force is approaching zero without having any undershot. It takes about half second for the internal forces to be compensated. The algorithm works well when the robot is moving in the free space as well as during the contact phase.