This paper presents MRAS speed sensorless control of induction motor using type-2 fuzzy logic controller (T2FLC). These controllers replace the PI ones, in the new MRAS strategy proposed in Benlaloui et al. (IEEE Trans Energy Convers 30(2):588–595, 2015), in order to improve the induction motor performances and robustness at low speed region. Indeed, the choice of these controllers is made because of their adaptation-based schemes which permit to handle nonlinear uncertain systems without the need of precise mathematical model required when using PI controllers. Comparative study had shown a better rejection of disturbance and high insensitivity to stator resistance compared to the PI and the T1FLC controllers. The effectiveness of the proposed speed-based T2FLC estimation method and its good robustness are validated by simulation and by experimental results.