Output synchronization for teleoperation of robot manipulators
Miller, Patrick Thomas
MetadataShow full item record
With the emergence of innumerable unmanned robotic systems, there is a greater push for a robust and stable means of teleoperated control. Teleoperation is extremely useful in two major ways. First, a remotely control robot may be able to enter an environment which is dangerous or inhabitable by a human operator. This scenario pertains to autonomous vehicle or hazardous waste management applications. Secondly, robot manipulators are capable of detecting and compensating for environmental interactions and making more precise and consistent movements then their human counterparts. This becomes particularly useful robot assisted surgery. In such remote control applications, it is also important to accurately reflect the remote environment back to the human user. This is typically achieved by adding force and torque sensors on the remote system and transmitting force data back to the operator. This is an ideal form of bilateral control since the sense of touch is significantly more sensitive then the human eye. However, problems arise when transmission delays are introduced into the link between host and remote system. Transmission delays are an unavoidable part of remote communications which are typically compensated by adding physical or virtual compliance into the system. However, this has the deleterious effect on 'transparent' perception of the remote environment. Thus, it is the responsibility of the control algorithm to compensate for delay induced instability while providing a transparent feel to the human operator. Maintaining the stability of the overall teleoperation scheme at a minimal loss of performance is the basis of this thesis. The case-study of telemanipulation of an RR planar manipulator and wheeled mobile robot using the stylus endpoint Phantom Omni haptic device will serve to form our efforts in this thesis. First and foremost the devices are kinematically and dynamically dissimilar. These couple with the remaining issues of stable teleoperation over lossy, band limited, imperfect communication media to make this a challenging task. To this end we will study the application of a bilateral teleoperation framework developed  using nonlinear control and energy based stability theory to overcome some of the limitations. Locally an adaptive controller will identify and stabilize the dynamics of the master and slave systems. This local controller will rendered the dynamics passive to a secondary coupling input. A passive mapping is used to couple the output states of the master and slave systems. This mapping is shown to be insensitive to lossy and delayed communication mediums. A battery of simulations and real time experiments are used to verify the proposed controller.