Effects of time delay in P-P telemanipulation architectures

Position-position telemanipulation architectures are widely used control schemes for remote robotic operation. This type of architecture is known for its stability advantages over Position-Force architectures. However, time delays in communication channels still degrade performance and lead to instability under certain conditions. Understanding how these delays affect system behaviour is crucial for improving the reliability and efficiency of teleoperated systems.

A theoretical analysis was conducted, and through simulation-based experiments, three primary research questions were addressed: first, how the P-P architecture behaves under ideal conditions; second, how the system behaves due to bidirectional time delays; and finally, whether adding a compensation term mitigates any adverse effects present.

The results show that position and forces can be precisely predicted under ideal conditions. However, the remote robot does not perfectly track the operator’s position, even under ideal conditions, due to inherent dynamics present in the system. When time delays are introduced, the system initially exhibits a damping effect before transitioning to non-passive behaviour beyond a critical delay threshold, eventually leading to energy-generative behaviour. This outcome depends not only on the time delay but also on the chosen input frequency. To counteract the negative impacts of time delays, a compensation term, computed as the parasitic force due to time delays, was implemented via a feedforward controller.

Simulations confirmed that this approach improves performance by reducing the parasitic effects due to time delay. This force refers to the unwanted effects of time delay on system performance. It allows for improved alignment between the operator and remote robot forces. However, this compensation term also removes the beneficial dissipative contribution of the parasitic force that helps in maintaining system passivity. Specifically, when the remote robot interacts with a dynamic environment under time delays, such as a virtual wall, the remaining delayed force feedback can inject energy into the system. This was shown to lead to system generative behaviour under conditions that were previously dissipative. Therefore, while this compensation strategy improves force  transparency, it does not guarantee passivity when coupled with dynamic environments.

The findings of this study contribute to a deeper understanding of P-P telemanipulation architectures and the effects of time delay on these systems. Future work should focus on experimental validation with physical teleoperation devices. Additionally, since this method does not inherently solve stability issues, combining this compensation strategy with techniques that aim to solve stability issues may provide additional insight into P-P architectures and their behaviour.