We study the problem of distributed state-space estimation, where a set of nodes are required to estimate the state of a nonlinear state-space system based on their observations. We extend our previous work on distributed Kalman filtering to the nonlinear case, and propose algorithms for Extended and Unscented Kalman filtering. The resulting algorithms are robust to node and link failure, scalable, and fully distributed, in the sense that no fusion center is required, and nodes communicate with their neighbors only. We apply the algorithms to the problem of estimating the position of every node in an ad-hoc network, also known as wireless localization. Simulation results illustrate the performance of the proposed algorithms.