Conventional target localization algorithms for the multiple-input multiple-output (MIMO) radar with widely separated antennas are mainly based on the centralized localization framework, which suffer from huge burdens of computation and communication as well as robustness defect in complex environments. To overcome these issues, this article first considers the MIMO radar with widely separated directional transmitters and omnidirectional receivers generating the bistatic range (BR) and angle of incidence (AOI) measurements for target localization, which is proven that the position of the target can be uniquely determined with one transmitter and one receiver. Then, the distributed localization framework using the hybrid BR and AOI measurements is developed, where each receiver acts as a fusion center extracting the BR and AOI measurements to perform the localization process, and the localization problem is formulated as solving a linear matrix equation. The distributed constrained total least squares (DCTLS) algorithm considering errors in both data matrix and observation vector is proposed for each receiver for target localization. The localization performance regarding the Cramér–Rao lower bound is derived for theoretical analysis. Numerical simulations are performed to validate the efficacy and superiority of the proposed DCTLS algorithm over other typical localization algorithms.