Tractor-trailer wheeled robots (TTWRs) are highly nonlinear and underactuated dynamical systems. It is necessary to use nonlinear control methods, for the control of wheeled robots. Back-stepping method is a Lyapunov-based systematic technique for designing nonlinear control algorithms. In this paper, an adaptive back-stepping controller is proposed for the TTWRs. The proposed algorithm uses an adaptive layout for the compensation of the system wheel slips, which updates controller parameters based on a combination of error signals and estimated uncertainties. This paper is one of the firsts to propose a control algorithm for off-axle TTWRs in the presence of wheel slips. The control algorithm is designed to track the reference trajectories and make the robot asymptotically stable around the reference trajectories. The stability of the method is proved using Lyapunov theory. In order to compensate the sliding of wheels as the system uncertainties, appropriate adaptive rules have been investigated. Obtained results demonstrate the efficiency of the proposed method. The results for the tracking control of the TTWR in the presence of wheel slips show that the slip effects are effectively compensated using the proposed adaptive back-stepping control algorithm.