In order to solve the problem of large error and poor stability of mobile robot positioning system in NLOS environment, a novel NLOS error resistant N-CTK combined algorithm was proposed. Firstly, based on the Chan-Taylor collaborative algorithm, a CTK combined positioning algorithm is proposed by integrating the Kalman filter algorithm. Then, the NLOS error model is built based on the TDOA measurement value, the NLOS error conversion factor is introduced, the extended Kalman filter algorithm is fused, and the proposed CTK combined algorithm is combined to obtain the estimated value of the tag finally. Experimental results show that the cumulative distribution function of the N-CTK algorithm is 13.5% higher than that of the CTK algorithm when the error is 6cm in LOS environment, and 55% higher than that of the CTK algorithm when the error is 15cm in NLOS environment, and the positioning accuracy is significantly improved.