Steering policies for controlled Markov chains under a recurrence condition

The authors consider the class of steering policies for controlled Markov c hains under a recurrence condition. A steering policy is defined as one ada ptively alternating between two stationary policies in order to track a sam ple average cost to a desired value. Convergence of the sample average cost s is derived via direct sample path arguments, and the performance of the s teering policy is discussed. Steering policies are motivated by, and partic ularly useful in, the discussion of constrained Markov chains with a single constraint.