Time delay compensation based on Smith Predictor in multiagent formation control

摘要

This paper investigates the use of time delay compensation methodology based on Smith Predictor applied to the control of a group of agents moving on a plane to a desired rigid geometric configuration. The unavoidable presence of time delays inherent to the communication links between agents is known to have a negative impact on the system performance, leading to instability in some cases. The decentralized and nonlinear nature of the underlying control problem has been taken into account to find a suitable control scheme that counteracts the effect of time delays. Moreover, the control scheme works using relative position measurements expressed in local frames, not being necessary a global coordinate reference frame. It is theoretically demonstrated and confirmed by simulation that exponential stability to the prescribed formation is achieved when time delays are constant and known. Finally, further simulation results show that, even when there exist errors on time delays knowledge, the global system performance is significantly improved with respect to the case of no delay compensation.