In this paper, we study a robust nonparametric estimator of a regression function when the response variable is subject to random right censorship and when the covariate is of functional nature. Under suitable conditions, we establish an almost complete (a.co.) consistency result with rate as well as an asymptotic distribution result of the estimator when observed data exhibit a mixing dependence.

censored data, robust nonparametric regression, kernel estimate, alpha-mixing dependence, functional data, almost complete convergence, asymptotic normality.