Adaptive targeting of chaos

We report two adaptive methods for directing chaotic trajectories to desired targets that require only a single ''probing'' of the target by the unperturbed dynamics. In contrast to previous targeting algorithms, these methods do not require a priori information about the stable and unstable manifolds associated with the target point and are not restricted to invertible mappings. The methods apply small perturbations to the state variables (as opposed to parameters) and can reduce the waiting time for the system to visit the target by more than two orders of magnitude. Their robustness and lack of stringent requirements should make these methods easily implementable in experimental applications.