Effects of time delay in P-P telemanipulation architectures

MSc assignment

Telemanipulation allows a human operator (master) to control tele-operated devices (slave) and interact with remote environments. Other than the human operator and the tele-operator, a communication architecture is also required, which informs how these two devices will exchange information. One type of communication architecture is the position-position (P-P) architecture; this configuration transmits a position command from the master to the slave and returns information about the slave position back to the master device. While telemanipulation is essential for an operator when tasks need to be accomplished remotely or in dangerous environments, it is subject to time-delays due to long distances between operator and tele-operator. Even small time-delays can greatly affect the stability and transparency of the teleoperation system.

Properly understanding the effects of time delays in P-P telemanipulation architectures is important to ensure good system performance and stability. Interestingly, small time-delays present in communication channels of a P-P controller result in performance degradation before observing instability. This deterioration in performance presents itself in the form of a damping effect.  However, with larger time delays the system becomes energy generative and results in an unstable system. While this type of architecture provides some robustness to time-delays in terms of stability, the damping effect due to deterioration of performance is an undesirable consequence. Some work on the theory has already been done to try and better understand why P-P architectures behave in such a way, and a solution, in the form of a feed-forward controller, proposed to mitigate the observed performance degradation. Although simulations have proven the addition of a feed-forward controller to be effective, experimental data is needed to verify the theoretical findings.

In this assignment, the behaviour of P-P architectures when subjected to time delays will be further investigated. A feed-forward controller will be introduced to the teleoperator system to attempt to mitigate the observed performance degradation. The goal will be to design, implement and execute experiments to verify whether improvement in performance can be observed.